#  @systematicls systematic longshort systematic longshort posts on X about if you, money, stocks, in the the most. They currently have [------] followers and [---] posts still getting attention that total [-------] engagements in the last [--] hours. ### Engagements: [-------] [#](/creator/twitter::1779775861589504000/interactions)  - [--] Week [---------] +1,770% - [--] Month [---------] -77% - [--] Months [----------] +5,553% - [--] Year [----------] +603% ### Mentions: [--] [#](/creator/twitter::1779775861589504000/posts_active)  - [--] Week [--] -48% - [--] Month [---] -51% - [--] Months [---] +2,925% - [--] Year [---] +168% ### Followers: [------] [#](/creator/twitter::1779775861589504000/followers)  - [--] Week [------] +0.37% - [--] Month [------] +2.90% - [--] Months [------] +298% - [--] Year [------] +374% ### CreatorRank: [-------] [#](/creator/twitter::1779775861589504000/influencer_rank)  ### Social Influence **Social category influence** [finance](/list/finance) [technology brands](/list/technology-brands) [cryptocurrencies](/list/cryptocurrencies) [social networks](/list/social-networks) [stocks](/list/stocks) [countries](/list/countries) [exchanges](/list/exchanges) [celebrities](/list/celebrities) [currencies](/list/currencies) [vc firms](/list/vc-firms) **Social topic influence** [if you](/topic/if-you), [money](/topic/money), [stocks](/topic/stocks), [in the](/topic/in-the), [ai](/topic/ai), [crypto](/topic/crypto), [investment](/topic/investment), [hedging](/topic/hedging), [finance](/topic/finance), [quant](/topic/quant) **Top accounts mentioned or mentioned by** [@paleologo](/creator/undefined) [@0xfdf](/creator/undefined) [@scottph77711570](/creator/undefined) [@quantarb](/creator/undefined) [@therobotjames](/creator/undefined) [@macrocephalopod](/creator/undefined) [@pedma7](/creator/undefined) [@friendscallmeap](/creator/undefined) [@cryptohound666](/creator/undefined) [@oxbridgequant](/creator/undefined) [@quantymacro](/creator/undefined) [@experquisite](/creator/undefined) [@buffetsalpha](/creator/undefined) [@maruushae](/creator/undefined) [@elletwocache](/creator/undefined) [@palashiam](/creator/undefined) [@coastalquant](/creator/undefined) [@paleologos](/creator/undefined) [@movietimedev](/creator/undefined) [@agustinlebron3](/creator/undefined) **Top assets mentioned** [Quant (QNT)](/topic/quant) [Alpha Technology Group Limited (ATGL)](/topic/alpha) [Bitcoin (BTC)](/topic/bitcoin) [Alphabet Inc Class A (GOOGL)](/topic/$googl) [Premia (PREMIA)](/topic/premia) [Robinhood Markets, Inc. (HOOD)](/topic/$hood) [CITADEL (CITADEL)](/topic/citadel) [CVS Health Corp (CVS)](/topic/cvs) ### Top Social Posts Top posts by engagements in the last [--] hours "huge proponent of building own projects around the entire pipeline of trading if you're a qr or aspiring qr. often times you will be siloed into specific responsibilities (e.g. alpha generation data cleaning etc); personal projects round out the skillsets. This is not far off. Time spent has more effect when it's spent working on your own projects. https://t.co/lsNyRzPzsb This is not far off. Time spent has more effect when it's spent working on your own projects. https://t.co/lsNyRzPzsb" [X Link](https://x.com/systematicls/status/1786636864151007518) 2024-05-04T05:59Z [----] followers, [---] engagements "@therobotjames @_deepdiveStocks if you look at the most illiquid (by adv) penny stocks you might conclude that their variance is low because they are simply not trading. where the analysis fails is that the impact from actually trading will lead to disproportionate variance above his static analysis" [X Link](https://x.com/systematicls/status/1787709247909732378) 2024-05-07T05:01Z [--] followers, [--] engagements "@macrocephalopod @ScottPh77711570 @PtrPomorski How are you guys walking around with no market-neutral mandate. I can barely get 5% delta exposure because my phone rings and risk is on the line" [X Link](https://x.com/systematicls/status/1788053662125662561) 2024-05-08T03:49Z [--] followers, [--] engagements "@ScottPh77711570 @macrocephalopod @PtrPomorski i mean when ure retail you're not charging yourself 2/20. so. returns are returns enjoy the risk premia" [X Link](https://x.com/systematicls/status/1788058084415246661) 2024-05-08T04:07Z [--] followers, [--] engagements "@ScottPh77711570 @therobotjames @SpacChad @readysetliqd @notgodcomplex96 @liquiditygoblin @weaponizedFOMO found an interview with @therobotjames https://www.youtube.com/shorts/HAsemyRhNj8 https://www.youtube.com/shorts/HAsemyRhNj8" [X Link](https://x.com/systematicls/status/1788840433151246641) 2024-05-10T07:56Z [--] followers, [--] engagements "@ScottPh77711570 @Larryjamieson_ What made you decide to move to Thailand Was it just to stretch the US dollar you got from trading" [X Link](https://x.com/systematicls/status/1789220180775612437) 2024-05-11T09:05Z [---] followers, [---] engagements "is it discretionary trading still if an AI does it or is it black-box systematic trading from https://ibkrcampus.com/ibkr-quant-news/dr-ernest-chans-journey-the-promise-and-limits-of-ai-and-machine-learning-for-investing/ https://ibkrcampus.com/ibkr-quant-news/dr-ernest-chans-journey-the-promise-and-limits-of-ai-and-machine-learning-for-investing/" [X Link](https://x.com/systematicls/status/1789373225446543635) 2024-05-11T19:13Z [----] followers, [---] engagements "The intern's signal signal - dataset1_feature1 + dataset1_feature2 + dataset1_feature3 + dataset1_feature4 + dataset1_feature5 + . + dataset1_feature20 + dataset1_feature21 no operators no transformations just linear addition of 20+ features from the same dataset" [X Link](https://x.com/systematicls/status/1789552597755412664) 2024-05-12T07:06Z [----] followers, 28.7K engagements "@arnaud710 keeping it simple - if not data driven features should ideally be causal to returns e.g. we have reason to believe that stocks with increasing dividend yield have greater demand and therefore drive prices up signal - ts_delta(dividend_yield 90) 1/2" [X Link](https://x.com/systematicls/status/1789695655021314248) 2024-05-12T16:34Z [----] followers, [----] engagements "Created this acct to vibe with the crypto crowd since i'm exploring a new asset class; been in equities my whole career. Then got 30k views from a thread about signal correlation. Seems like there's a demand for the ol quanty tweets. What should I shitpost (vaguely) about" [X Link](https://x.com/systematicls/status/1789840004740427992) 2024-05-13T02:08Z [----] followers, 14.5K engagements "Anyone who has spent real R&D dollars on processes know that every discovery and knowledge has a cost. But there's still a way to glean knowledge from the people who have played in institutional grounds. 2/7" [X Link](https://x.com/systematicls/status/1789948333118693378) 2024-05-13T09:18Z [----] followers, [---] engagements "Thankfully there are still [--] types of content I've found useful in my time lurking on twitter and listening to podcasts on spotify: [--]. Guys thinks everyone knows this [--]. Accidental alpha leaks [--]. Guy just really needs the world to think he is smart 3/7" [X Link](https://x.com/systematicls/status/1789948336792813901) 2024-05-13T09:18Z [----] followers, [----] engagements "@ScottPh77711570 @__paleologo Yeah I thought you would really like it. He's not just a good practitioner but a good business owner and thinks about life from an extreme sports / special force vantage point. Very on the nose about considerations on risk reward factor hedging talent and leverage too" [X Link](https://x.com/systematicls/status/1789974358875652317) 2024-05-13T11:01Z [----] followers, [---] engagements "Are there any shops where the "in-house proprietary MFM" isn't just a wrapper around Barra/Axioma + (2-3 in-house models)" [X Link](https://x.com/systematicls/status/1790436760138260576) 2024-05-14T17:39Z [----] followers, [----] engagements "@bzzz_anon And what did u tell them" [X Link](https://x.com/systematicls/status/1790450636279919093) 2024-05-14T18:34Z [----] followers, [---] engagements "@JaredKubin So awesome. And they are being used for explainatory purposes ex-post or for hedging If for hedging have you benchmarked against Barra" [X Link](https://x.com/systematicls/status/1790815236032741668) 2024-05-15T18:43Z [----] followers, [---] engagements "Extremely large and successful podshops pass-through all costs to investors. They only hurt when there are a small number of large PnL generating teams that cannot offset a large number of losing teams. Then they would have to still pay the positive teams according to the formulaic compensation and this payment would have to come from retained earnings/treasury. Extremely leveraged long/short portfolios are in vogue right now (e.g. pod shops). They rely heavily on stops to manage risk. Could the failure mode be rogue wave short squeezes where things just gap up hundreds of percent in a single" [X Link](https://x.com/systematicls/status/1791580531882410306) 2024-05-17T21:24Z [----] followers, 74K engagements "Secondly risk limits exists not only on individual pods but as a firm as well. These risks are extremely well-distributed over a staggeringly large surface. For every team that is packing their things into a box after getting blown-up there is another team with equal and opposite positions popping champagne for having made their quarter. The probability of largewell-run podshops imploding from such undiversified carelessness is not zero but extremely minute" [X Link](https://x.com/systematicls/status/1791580534084419692) 2024-05-17T21:24Z [----] followers, [----] engagements "I'm endeavoring to share some thoughts on the markets quantitative research portfolio management building a career and life amassed over more than a decade of risk-taking. There are plenty of great long-form content covering all of the above in great detail. My goal is not to compete with them but perhaps in as few words and as little equations as possible try to impart the salient point intuitively and then refer you to the long-form content for rigor if necessary. I believe that strong one-liners are more easily incorporated into heuristics than [--] pages of terse content and good heuristics" [X Link](https://x.com/systematicls/status/1791730475243086330) 2024-05-18T07:20Z [----] followers, 22.1K engagements "My favorite part of this is you calling schonfeld mediocre. That's how I know you're a player. Also I think you are talking about two different points here crowding in delevering/hedging and withdrawal runs. In crowded deleverings most firms now want PMs to be orthogonal to some custom crowding factors; but I feel that misses the point. Hedgefunds become correlated to each other during systemic events BECAUSE we are all trying to hedge to the same factors. 95% of risk models are a copy of Barra/Axioma based on Barra/Axioma or benchmarked to Barra/Axioma. When we see exposures and form hedging" [X Link](https://x.com/systematicls/status/1791738400393109815) 2024-05-18T07:51Z [----] followers, [--] engagements "@__paleologo @X [--] + [--] + [--] + [--] + [--] + . = -1/12" [X Link](https://x.com/systematicls/status/1791856291520073905) 2024-05-18T15:40Z [----] followers, [---] engagements "http://x.com/i/article/1791446922177458178 http://x.com/i/article/1791446922177458178" [X Link](https://x.com/systematicls/status/1792152309574808051) 2024-05-19T11:16Z [----] followers, 70.4K engagements "For those of you who have enjoyed the posts on ergodicity you would also enjoy Taleb's work - the source of my inspiration on the subject. I want to point out specifically [--] things that I think are worth paying attention to that I haven't mentioned in the accompanying post: [--]. The first thing I would like to point out is to take reference from Taleb's definition of "ensemble probability" where observations are independent of each other (1000 people going to casino) and "time probability" where observations have path dependency (1 person going to the casino [----] times). In backtesting a" [X Link](https://x.com/systematicls/status/1793015700279398469) 2024-05-21T20:27Z [----] followers, 26.9K engagements "Should global equities be moved on-chain in the form of perpetuals Off CeFi exchange hours we could allow price discovery on-chain on CeFi exchange hours (30 mins post auction) we can implement funding rates to tether prices to the CeFi exchange hours. Pros: [--]. One wallet for global margin management [--]. 24/7 Trading and price discovery [--]. No implementation shortfall no missing locates [--]. Transparent automated PnL Tx reports [--]. Vault vertical for easy integration of institutional players (funds/mm/prop) Cons: [--]. Will not solve long tail of thin liquidity past first [---] instruments for every" [X Link](https://x.com/systematicls/status/1793153567861440790) 2024-05-22T05:35Z [----] followers, [----] engagements "@palash_i_am hilarious and a valid point" [X Link](https://x.com/systematicls/status/1793155620025294911) 2024-05-22T05:43Z [----] followers, [---] engagements "I think really with any kind of automated signal research the most important question is how do you overcome the immense amount of noise in your signal pool. If you have some cause - effect you are looking for do you really need to automate the signal discovery Perhaps you could argue that you know the function but not yet the form. And if you don't then you are essentially data-mining which is not atrocious but it will introduce plenty of noise into your signal pool. How do you overcome that then becomes the principal question" [X Link](https://x.com/systematicls/status/1793216483172204616) 2024-05-22T09:45Z [----] followers, [----] engagements "This is one of those areas which are a siren song to those who have just started. It sounds like a dream come through and I think even @quant_arb has written extensively about it. You want to believe that taking [-----] features and throwing it into a process will give you [--] sharpe but so does every graduate with some rudimentary understanding of sklearn. The fact is this: from a market efficiency point of view if it were so easy this would be the dominant way to extract alpha out of the market. The fact that it is conventional wisdom to advise AGAINST it tells you that the approach is riddled" [X Link](https://x.com/systematicls/status/1793217337602277480) 2024-05-22T09:48Z [----] followers, [----] engagements "@cryptohound666 @oxbridgequant @quant_arb @0xfdf Even Piper agrees" [X Link](https://x.com/systematicls/status/1793217734744170499) 2024-05-22T09:49Z [----] followers, [---] engagements "Whenever I'm inevitably asked about alpha mining - I always default to pizzas. I love myself a good new york slice. There are many ways to get a pizza going. Tomato sauce base mozerella pepperoni Fucking amazing. [--] sharpe. White sauce base mozzy mushrooms Give it to me. [--] sharpe Lemon ricotta base spinach lamb sausage I'm all over it. [---] sharpe. You look at all these pizzas and you gotta respect the pizzalo for having some game. One day your friend who has never baked a day in his life tells you "you know what I'm going to get into the pizza game. I'm going to get into the business of" [X Link](https://x.com/systematicls/status/1793224610085060654) 2024-05-22T10:17Z [----] followers, [----] engagements "@quantymacro @ScottPh77711570 @oxbridgequant @cryptohound666 @quant_arb @0xfdf next month it won't be halved. it will be complaining about [--] productivity" [X Link](https://x.com/systematicls/status/1793230995967352835) 2024-05-22T10:42Z [----] followers, [---] engagements "@0xfdf @experquisite @oxbridgequant @cryptohound666 @quant_arb Selection bias under multiple testing 10.3905/jpm.2014.40.5.094" [X Link](https://x.com/systematicls/status/1793252874677530833) 2024-05-22T12:09Z [----] followers, [---] engagements "@0xfdf @experquisite @oxbridgequant @cryptohound666 @quant_arb pink one looks good. count me in for 5M" [X Link](https://x.com/systematicls/status/1793253246011842704) 2024-05-22T12:11Z [----] followers, [---] engagements "@0xfdf @experquisite @oxbridgequant @cryptohound666 @quant_arb count me in for the green one at -5M too" [X Link](https://x.com/systematicls/status/1793253303062819082) 2024-05-22T12:11Z [----] followers, [---] engagements "@0xfdf @experquisite @oxbridgequant @cryptohound666 @quant_arb mentions rentech this guy is retail" [X Link](https://x.com/systematicls/status/1793253876059324751) 2024-05-22T12:13Z [----] followers, [---] engagements "@0xfdf @experquisite @oxbridgequant @cryptohound666 @quant_arb always nice to read your stuff. can tell you're so passionate about the game" [X Link](https://x.com/systematicls/status/1793255824707686414) 2024-05-22T12:21Z [----] followers, [---] engagements "Have you heard of Renaissance bakery though Do you know they have this baking unit that is only available for employees It's been rumored that nobody knows what goes into the pizzas from that baking unit. It's been delivering deliciousness [--] folds above the industry average for [--] years straight" [X Link](https://x.com/systematicls/status/1793260017535349197) 2024-05-22T12:37Z [----] followers, [---] engagements "@quantymacro @0xfdf @YannickKae @0xfdf how does one hedge against the PnL factor" [X Link](https://x.com/systematicls/status/1793327979000648183) 2024-05-22T17:08Z [----] followers, [---] engagements "@quantymacro @0xfdf @YannickKae Omg I know for a portfolio P my hedging basket is -P. PnL neutral" [X Link](https://x.com/systematicls/status/1793328161217990943) 2024-05-22T17:08Z [----] followers, [---] engagements "@Norsay7 @0xfdf @portfoliooptim @experquisite @oxbridgequant @cryptohound666 @quant_arb This is a GOOD point. And is really where a serious conversation on data mining should start" [X Link](https://x.com/systematicls/status/1793343705241604608) 2024-05-22T18:10Z [----] followers, [--] engagements "With the advent of liquidity venues for private markets (NPM and the like) I wonder how much longer it will be before quants are all over the private markets -- from my understanding and limited market research the only competitor/proof-of-concept now is EQUIAM. It seems like we (quants) would deliver a strictly better product from an investment offering perspective. Hear me out: 1) Ability to learn from vast amounts of historical data on past non-intuitive patterns that generalize (pitchbook crunchbase linkedIn SP etc). For every founder/start-up trait that represents a traditional signal" [X Link](https://x.com/systematicls/status/1793493004306768031) 2024-05-23T04:03Z [----] followers, 11.8K engagements "Example of a private equity liquidity venue: https://www.nasdaqprivatemarket.com/product/second-market/ https://www.nasdaqprivatemarket.com/product/second-market/" [X Link](https://x.com/systematicls/status/1793493387578142941) 2024-05-23T04:05Z [----] followers, [----] engagements "The only actually quant vc I'm aware of: https://equiam.com/ https://equiam.com/" [X Link](https://x.com/systematicls/status/1793493722040283410) 2024-05-23T04:06Z [----] followers, [----] engagements "@sersabr Which other funds do you know of That are taking a purely quant approach to the problem" [X Link](https://x.com/systematicls/status/1793500109952725182) 2024-05-23T04:32Z [----] followers, [---] engagements "Post about quant in private markets. I'm quote retweeting this to see if this bypasses the X algo on quashing links. With the advent of liquidity venues for private markets (NPM and the like) I wonder how much longer it will be before quants are all over the private markets -- from my understanding and limited market research the only competitor/proof-of-concept now is EQUIAM. It seems like With the advent of liquidity venues for private markets (NPM and the like) I wonder how much longer it will be before quants are all over the private markets -- from my understanding and limited market" [X Link](https://x.com/systematicls/status/1793718589536731273) 2024-05-23T19:00Z [----] followers, [----] engagements "http://x.com/i/article/1792999650309578752 http://x.com/i/article/1792999650309578752" [X Link](https://x.com/systematicls/status/1793975236242907298) 2024-05-24T12:00Z [----] followers, 39.7K engagements "The source of the quote from @0xfdf https://x.com/0xfdf/status/1793655052692759036 When interviewing quants I often encounter people who are qualified on paper from firms I'm not familiar with but who can't credibly do the work. I've picked up a few consistent signals about those firms that I'll share so people entering the industry know what to avoid 🧵 https://x.com/0xfdf/status/1793655052692759036 When interviewing quants I often encounter people who are qualified on paper from firms I'm not familiar with but who can't credibly do the work. I've picked up a few consistent signals about" [X Link](https://x.com/systematicls/status/1793975892445933580) 2024-05-24T12:02Z [----] followers, [----] engagements "I wake up everyday thankful I am a quant. I go to sleep everyday thankful I am a quant. I genuinely believe it is good for the soul and mind. If we could run for basis points it might even be good for the body" [X Link](https://x.com/systematicls/status/1794135472119382160) 2024-05-24T22:36Z [----] followers, 93.5K engagements "Play the game for what it is. A chance to compete with your mind a chance to prove yourself in a system that is as close to meritocratic as it can be. One has to revere the challenge and be able to show up every day rain or shine and compete. If you can do it day on day for decades you will be rewarded with the challenge of your life brilliant colleagues (really some of the smartest people around) and oftentimes more money than you know what to do with. Quants who can't enjoy the game will one day drop out and find themselves lurking fintwit and complaining about the soullessness of other" [X Link](https://x.com/systematicls/status/1794138229429649812) 2024-05-24T22:47Z [----] followers, [----] engagements "I'd also like to believe that being a quant is more than just a career but a way of life of reasoning about the difficult choices we have to make in life. There are few careers where you have to research a topic stake real money on it get noisy feedback in real-time and manage the risk of being wrong. These skills make for a pretty well-rounded toolkit for dealing with most problems in life" [X Link](https://x.com/systematicls/status/1794139931562131695) 2024-05-24T22:54Z [----] followers, [----] engagements "One of the things I really enjoy being a quant but don't often get to talk about is just how absolutely diverse the ways one can be smart in. I think it's a function of the many ways one could make money in the markets. The longer you stay in the industry the more you appreciate that people are multi-dimensional and can be geniuses in ways you are retarded in and vice versa" [X Link](https://x.com/systematicls/status/1794141110849437953) 2024-05-24T22:59Z [----] followers, [----] engagements "Have a mind to start a weekend side project where I'll look at private equity data and see if there are any signals that are persistent and exploitable" [X Link](https://x.com/systematicls/status/1794259090275000619) 2024-05-25T06:47Z [----] followers, [----] engagements "Skill stones in Diablo immortal were so extremely inefficient when they were first released. They were a combination of some constituent fungible commodity e.g. skill_stone = skill_1 + skill_2 + skill_3. Hence the fair way to price a skill_stone would just be to do price(skill_stone) = price(skill_1) + price(skill_2) + price(skill_3). The catch was that there was no product or market for individual skills only skill_stones that were a random ensemble of skills. Firstly liquidity was so thin you could find a certain skill and buy up all the stones below a certain price (floor) with that skill" [X Link](https://x.com/systematicls/status/1794266077314310289) 2024-05-25T07:15Z [----] followers, [----] engagements "A realistic fixed transaction cost in US D1 is 5bps on average if you are in a large fund and [--] bps if you are in a small fund. As a heuristic you can model another [--] bps of slippage/impact for half most liquid part of your universe and 10bps for the half most illiquid parts of the universe. This is a harsher estimation than reality. Its dirty and is a gross approximation but it should do" [X Link](https://x.com/systematicls/status/1794467255453204884) 2024-05-25T20:35Z 10.3K followers, [----] engagements "Sorry when I say "universe". I mean the Russel3000. If you are constructing your universe out of yahoo finance (noob) and getting sp500 you can model all-in costs as 5bps-10bps and life will be fine" [X Link](https://x.com/systematicls/status/1794469752913760699) 2024-05-25T20:45Z [----] followers, [----] engagements "@mipt_filipp @0xfdf Yeah you can tell from his reply that he likely trades a more liquid universe. If you trade 3-4% ADV on the most [----] illiquid names in Russel3000 you will pay for it in slippage and impact" [X Link](https://x.com/systematicls/status/1794636262894407989) 2024-05-26T07:46Z [----] followers, [--] engagements "@KimchiQuant I absolutely know that - which is why I am offering corrections. The entire premise of this is that SVB free data is expensive and I know most people reading this aren't going to have it" [X Link](https://x.com/systematicls/status/1794636508449956213) 2024-05-26T07:47Z [----] followers, [--] engagements "@therobotjames de beers" [X Link](https://x.com/systematicls/status/1794654620415091061) 2024-05-26T08:59Z [----] followers, [--] engagements "@__paleologo Maybe I am cringe" [X Link](https://x.com/systematicls/status/1794670430181339163) 2024-05-26T10:02Z [----] followers, [---] engagements "@__paleologo "For once I am hoping people confuse us with BlackRock" was pretty funny to me" [X Link](https://x.com/systematicls/status/1794670865210319010) 2024-05-26T10:04Z [----] followers, [---] engagements "Why do you guys refer to this as crowding literature It's just an investigation on flows to me and we know for the most part that there are flows have a temporary and permanent component - the interesting thing about this paper is how long-lasting the temporary component is because it has a positive feedback loop that keeps it going. -- Crowding in my mind is nomenclature served for phenomena of the masses" [X Link](https://x.com/systematicls/status/1794796171141271657) 2024-05-26T18:22Z [----] followers, [---] engagements "The funny thing about this paper is that the hedging the "ponzi" risks part isn't clear but you could probably monetize the flows indicator since its so persistent. You would also expect it to be positive in the long run since not all impact of the flows is temporary and the total PnL should be greater than the reversion. Hah" [X Link](https://x.com/systematicls/status/1794802139019215206) 2024-05-26T18:45Z [----] followers, [---] engagements "@AmeliaBarty I want to believe its our destiny I also have to hedge and say it is an **optimistic articulation** of our destiny. Have you read [--] body problem" [X Link](https://x.com/systematicls/status/1794809808622850463) 2024-05-26T19:16Z [----] followers, [---] engagements "Being on FinTweet and reading some accounts whose entire existence is to propagate some negative remark or incendiary comment really invokes "Genius of the Crowd" vibes by Charles Bukowski but there is genius in their hatred there is enough genius in their hatred to kill you to kill anybody not wanting solitude not understanding solitude they will attempt to destroy anything that differs from their own not being able to create art they will not understand art they will consider their failure as creators only as a failure of the world not being able to love fully they will believe your love" [X Link](https://x.com/systematicls/status/1794990271563956312) 2024-05-27T07:13Z [----] followers, [----] engagements "Movers Risk of Ruin recommended by @turbokapital The story starts with a banger hook of money lost a shoe flipper at Nike getting counter-measured by their employer. and then the quote "Arbers never sleep the ones who are trying to look for an advantage play are thinking about it 24/7 while the ones who are protecting the system are getting their [--] hours shut-eye" Great stories about adverse selection and counter-measures from the house. Then goes on to talk about exotic illiquid bets in small games (e.g. college basketball games) and how the house has a large edge in them. Apparently they" [X Link](https://x.com/systematicls/status/1795077128465400287) 2024-05-27T12:58Z [----] followers, [----] engagements "One of the places where I think it is very logical for a +EV vendor to sell their signals is sports betting. This is because it is difficult to get liquidity as a large known player in this market. If you truly have an edge you might actually be able to profit more from your edge selling your signals to the masses than trying to overcome the liquidity problem on your own. Reading this @ Bloomberg There are plenty of charlatans who are exceedingly good at selling courses on how to get rich quick and with material that is 1) ambiguous and 2) subject to your interpretation 3) with very low" [X Link](https://x.com/systematicls/status/1795077693798797369) 2024-05-27T13:00Z [----] followers, 15.6K engagements "Something is interesting about multi-manager systems always opting to start with discretionary PMs in some cases even discretionary analysts (Norias). I believe some part of it is because there is a long build-out required for quant teams in general but a part of me also believes it is because the CLB/Alpha Capture model works best when signals are strong. Quants have low information correlation on each bets they take on average. In the IR = IC x SQRT(BREADTH) model a systematic approach will have very low IC spread over a large number of bets. This is because most signals are faint patterns" [X Link](https://x.com/systematicls/status/1795395053038555232) 2024-05-28T10:01Z [----] followers, 20K engagements "One thing that I would like to add is that we know that if ER and VarR is estimated without error MVO is mathematically optimal. We also know that a large problem with MVO is that ER is very hard to estimate. I can imagine that if IC is sufficiently high from an ensemble of signals each with high IR the error on ER becomes smaller and MVO approaches optimality. This happens with quant systems as well of course. I think the question is if the effect is significantly more pronounced with discretionary PMs" [X Link](https://x.com/systematicls/status/1795396877824970969) 2024-05-28T10:09Z [----] followers, [----] engagements "@palash_i_am Which part I am not familiar with EP. I know they exist but nothing beyond that" [X Link](https://x.com/systematicls/status/1795397173506638270) 2024-05-28T10:10Z [----] followers, [---] engagements "@HiddeMulder2 I hear you but costs at the scale Jain has raised should not be the primary concern. I would imagine that not being able to have a strong start without be more of a concern. Unless your equation includes the cost of time as in this allows them to start earning asap" [X Link](https://x.com/systematicls/status/1795398587830218941) 2024-05-28T10:15Z [----] followers, [---] engagements "@IndEqQnt @__paleologo I think when he said AC book he meant ALPHA CAPTURE" [X Link](https://x.com/systematicls/status/1795402477896130667) 2024-05-28T10:31Z [----] followers, [---] engagements "@__paleologo Yes - that's what I mean speed of execution on the business level. On one hand I guess it's Occam's razor all the way down on the other hand kind of disappointed that it's just business strategy" [X Link](https://x.com/systematicls/status/1795409809136845038) 2024-05-28T11:00Z [----] followers, [---] engagements "@__paleologo And I'd have to take your word for it because I've heard great things about Citadel's CLB and I would assume you've seen how the sausages are made" [X Link](https://x.com/systematicls/status/1795410038787645759) 2024-05-28T11:01Z [----] followers, [---] engagements "Listening to @alexisohanian with @tseides on Capital Allocators podcast Really enjoyed the idea of community building on Reddit by playing as an internet host and the implication of online forums being a persistent "party" which led to some organic growth as people stumbled onto old threads. Secondly a good reminder on 1) the virtues of giving immediate direct and compassionate feedback and building a team centered around that; and 2) the importance of letting go of under-performers and retaining a passionate performant team Lastly the idea on social media tending towards proof of humanity as" [X Link](https://x.com/systematicls/status/1795504630119674356) 2024-05-28T17:17Z [----] followers, [----] engagements "Are there any papers I have missed other than the obvious ones I.E The effects of trading suspensions in China Anyone has any interesting anecdotes to contribute to the ether" [X Link](https://x.com/systematicls/status/1795770946864968087) 2024-05-29T10:55Z [----] followers, [----] engagements "Sorry I forgot to add my main point - this leads me to believe that national-chinese PMs have a data-source they are not talking about or some secret cave where they are discussing which stocks are undergoing suspension and for how long" [X Link](https://x.com/systematicls/status/1795771700430450781) 2024-05-29T10:58Z [----] followers, [----] engagements "Signals in china L/S are also unique because your long side often has no alpha. If you decompose your signals into long + short and look only at the residuals on the long side it's [--]. It is not a trivial exercise to get non-zero residuals on the long side. It is extremely competitive. Practically any signal that is worth its salt in the L/S space comes from the short side - precisely because it cannot be realized it cannot be competed away. This has some implications because 1) It implies that searching for signals that are index-hedged is functionally equivalent to searching for LO signals" [X Link](https://x.com/systematicls/status/1795784416410906847) 2024-05-29T11:49Z [----] followers, [----] engagements "It is pretty severe. Your asshole is puckered for an entire month. Your entire book is also thrown off its exposures. Hedging is difficult especially when the suspended stocks can be part of factor mimicking portfolios. Trivial to say that you can backtest this with historical suspended data but in live it is not trivial to predict which stocks are going to be suspended ex-ante" [X Link](https://x.com/systematicls/status/1795795444679029214) 2024-05-29T12:32Z [----] followers, [---] engagements "I work in statarb space so my signals are cross-sectional in nature. A simple example would be directional_dividend_yield - ts_delta(dividend_yield 63) cross_sectional_dividend_yield - directional_dividend_yield - cs_mean(directional_dividend_yield) The cross_sectional_dividend_yield signal would be dollar-neutral and the intended bets are on the relative outperformance of stocks with increasing dividend yield" [X Link](https://x.com/systematicls/status/1795857311191413002) 2024-05-29T16:38Z [----] followers, [---] engagements "@quanterson @__paleologo @palash_i_am There is a well-trodden path with clear requirements and a path less taken where you are free to pit yourself against the jungle If you are pitting yourself against the jungle your knuckles are probably going to be more useful than where you went to school" [X Link](https://x.com/systematicls/status/1796287070304096637) 2024-05-30T21:06Z [----] followers, [--] engagements "yet again the claim out there that risk models drive crowding. Sigh. Could you expound on this a little I wouldn't go as far as to say "risk factors drive crowding" but the idea that there is crowding from using the same risk factors seems fairly intuitive to me. If we are all looking at a certain factor when exposures to that factor rise we are all going to be shorting our exposure's worth of that factor to hedge. If we are using the same product - our factor-mimicking portfolio are going to be identical. There is going to be impact from this. Would this not be considered as crowding" [X Link](https://x.com/systematicls/status/1796459122147336648) 2024-05-31T08:30Z [----] followers, [----] engagements "I think something missing from this discussion is that 1) exposures are not constant 2) target portfolios are not formed instantaneously and we need to trade into them. 1) Exposures wax and wane. Exposures are decided by the same risk models. As long as any two strategies share some common variation to a factor and that exposure increases beyond some threshold hedging is expected 2) If you residualizing your portfolio to F it will result in resid_port = curr_port + hedging_port This hedging_port is going to be in the same direction as every other strategy with the same common variation to F." [X Link](https://x.com/systematicls/status/1796993326211842314) 2024-06-01T19:52Z [----] followers, [----] engagements "@0xfdf @__paleologo @quantymacro @lloydchrist84 @coastal_quant Let me model this when I have time" [X Link](https://x.com/systematicls/status/1796996303324590205) 2024-06-01T20:04Z [----] followers, [---] engagements "@0xfdf @__paleologo @quantymacro @lloydchrist84 @coastal_quant Sorry I wanna be more active but wife is really angry I'm not helping more with the house moving" [X Link](https://x.com/systematicls/status/1796997522772365362) 2024-06-01T20:09Z [----] followers, [---] engagements "@__paleologo @0xfdf @quantymacro @lloydchrist84 @coastal_quant Would you also like to meet my grandma in person" [X Link](https://x.com/systematicls/status/1797002819125616807) 2024-06-01T20:30Z [----] followers, [---] engagements "One reason why I am inclined to say this is because anectodally I've experienced a few instances where I see exposures of a portfolio change - trading activity in fmp - factor PnL starts to move - more trading activity in the fmp. Since then the idea that using the same risk models leads to crowding in the hedging baskets has always been an idea I've taken to be true but I've not investigated this in detail" [X Link](https://x.com/systematicls/status/1797004469277970810) 2024-06-01T20:37Z [----] followers, [---] engagements "But in a real world setting two strategies are definitely correlated. I can't imagine two index-rebal PMs running strategies that are completely orthogonal to each other. Using the same datasets using the same know-how. Its hard to imagine successful strategies of the same flavor being completely uncorrelated" [X Link](https://x.com/systematicls/status/1797005474266829119) 2024-06-01T20:41Z [----] followers, [----] engagements "@__paleologo @SmityAndre82456 @0xfdf @quantymacro @lloydchrist84 @coastal_quant Index rebal is also a good example because it almost always has positive loading on the momentum factor. Yet it's expensive to put on a constant hedge so most PMs opt not to until they get dinged for having indecent exposures" [X Link](https://x.com/systematicls/status/1797006658029359255) 2024-06-01T20:45Z [----] followers, [----] engagements "One final thing I'd also like to contribute to the ether is that two strategies can have a high correlation to each other in their target position space but trade at a different pace to it. The annoying thing about hedging is that all strategies with the same exposures have to put on hedges pretty much all at once. In a very real way it really does "expose" crowding" [X Link](https://x.com/systematicls/status/1797008668422246826) 2024-06-01T20:53Z [----] followers, [---] engagements "To the dms and comments on what book I should read or why don't I have x or y: 1) we travel alot and have a "buy what you are willing to dispose" mantra 2) this means unfortunately books get thrown out every moving cycle sometimes only to be bought again 3) i mostly use e-books. I buy paperback if i REALLY like it or to passed as a reference 4) guys I do read fiction literature. Read Great gatsby (its there). Love Kafka and Flowers more 5) There are more books in the library not shown here" [X Link](https://x.com/systematicls/status/1797840276435980510) 2024-06-04T03:58Z [----] followers, 19.6K engagements "@BlackSwan_ptf Thorpe is a fucking legend. Go for that first. Its a great read. Non technical but really is one of the books on the differentiation of mental giants vs rest of the humans" [X Link](https://x.com/systematicls/status/1797873771443085501) 2024-06-04T06:11Z [----] followers, [----] engagements "@bfpill Phemonenal read. More than a story about robert moses felt like a commentary on how a shadow govt works in practice" [X Link](https://x.com/systematicls/status/1797885248854020512) 2024-06-04T06:57Z [----] followers, [---] engagements "I think there is a rather interesting lesson to be drawn from @TheRoaringKitty's bet on GME. It may look like absolute lunacy but it's actually a fairly sound move even from a portfolio optimisation perspective. We often talk about diversification in the industry as a means of reducing variance - we want to smooth out our risks over a large number of instruments so no one instrument can hurt us too much. And this idea is taken as gospel and pushed to an extreme L/S equities often end up having portfolios that bet on 10000+ instruments globally at the same time over millions of signals. This" [X Link](https://x.com/systematicls/status/1797937480832053403) 2024-06-04T10:24Z [----] followers, 36.9K engagements "This is not a commentary on market manipulation. I for one am up for any strategy that can generate returns. I think it something that is easy to observe its association but won't be easy trying to --prove-- causation. If insider trading were legal I would not be a quant though. I'll be a discretionary L/S PM with ceos of nano-cap companies in sleazy clubs everyday of the week banging out 1000+% returns a year" [X Link](https://x.com/systematicls/status/1797939427597271050) 2024-06-04T10:32Z [----] followers, [----] engagements "@HiddeMulder2 @TheRoaringKitty Odd that you would call [--] stock + cash as diversification but I understand your point" [X Link](https://x.com/systematicls/status/1797940357696770097) 2024-06-04T10:36Z [----] followers, [---] engagements "What's the most intuitive explanation of factors that you guys know Every young quant (in a hf) is exposed to factors basically within [--] weeks into the job whether they are in a central training program or a pod. I find the mechanical explanation of a factor bland unintuitive and doesn't explain clearly the distaste for factors. I've always liked to get people to imagine factors as the first principal component of all the implementations of a known replicable alpha/signal. That captures the idea that 1) Factors are just another TxN weights matrix 2) Factors are basically spent alphas/signals" [X Link](https://x.com/systematicls/status/1798393715016483032) 2024-06-05T16:37Z [----] followers, [----] engagements ""Most systems have a few driving factors" is a nice generalization. In the army we used to understand that great units are seldom the sum of great men but the product of great leaders orchestrating the performance. Take out the leaders you cripple performance. @0xfdf @systematicls I don't know that I can answer that. But I also believe that there are two possible answers. [--]. The first is economic-theory based. There is one (SDF) factor whose return in a future state of the world is proportional to the marginal utility of of your consumption in that https://t.co/oxPrIchhht @0xfdf @systematicls" [X Link](https://x.com/systematicls/status/1798410354026066005) 2024-06-05T17:43Z [----] followers, [----] engagements "I think I've discovered something interesting but I need to test this in more detail. It appears that within the first [--] mins of a tweet one can roughly extrapolate the popularity of the tweet. There is a strong momentum effect that gains steam at the start or never does. I wish Analytics captured the time-series evolution of tweet impressions so I can study this in detail without writing some poller to track impressions on my own" [X Link](https://x.com/systematicls/status/1798436344588587264) 2024-06-05T19:26Z [----] followers, 10.1K engagements "I really like NNT's writings and ramblings. I know he has a reputation as a pompous prick but I find it endearing that he lives his life as a free man. I rage vicariously through him. https://medium.com/incerto/how-i-write-8b495eae0330 https://medium.com/incerto/how-i-write-8b495eae0330" [X Link](https://x.com/systematicls/status/1798604009521230244) 2024-06-06T06:33Z [----] followers, [----] engagements "Investigating factors in crypto and using only pairs in @HyperliquidX. Factors are orthogonalized to each other using gramschmidt. Observations: 1) Supermajority of returns is explained by the market factor 2) A tiny amount is explained by time-series momentum 3) The rest are vanishingly small and not statistically significant It could be that HL only lists cryptos of a certain size so the effect of the size factor is more muted in this universe. Still a surprising finding. Does anyone have contradictory results" [X Link](https://x.com/systematicls/status/1798622771595157520) 2024-06-06T07:47Z [----] followers, 36.3K engagements "@HyperliquidX To me at least it's counter-intuitive to what I'd expect from equities" [X Link](https://x.com/systematicls/status/1798624048760684708) 2024-06-06T07:52Z [----] followers, [----] engagements "@ScottPh77711570 If I don't residualize it to market this is what I get which I assume you should see something similar in the arithmetic returns space" [X Link](https://x.com/systematicls/status/1798639051999986066) 2024-06-06T08:52Z [----] followers, [----] engagements "@Andrew30482844 curious where you interpreted the requirement for a consistent drift" [X Link](https://x.com/systematicls/status/1798761559826813290) 2024-06-06T16:59Z [----] followers, [--] engagements "Ah okie that's fair. 1) I do not think all factors have positive premia all the time think they have pockets of predictability just like most signals. For the ones who do have long-term positive premia I do expect that they have some undesirable characteristics that allow them to persist. I should word that better. 2) I do believe however that factors are the results of the impact from persistent trading. E.g. That the covid factor is likely the impact of gross delevering across multiple strategies" [X Link](https://x.com/systematicls/status/1798765384985923622) 2024-06-06T17:14Z [----] followers, [---] engagements "@weaponizedFOMO It's true that the bar to be competent for a somewhat competitive end-to-end system has gotten progressively higher :-( How do you deal with it" [X Link](https://x.com/systematicls/status/1799025511844446385) 2024-06-07T10:28Z [----] followers, [---] engagements "@pedma7 Thank you Trend is a very sticky. It's relatively high vol low sharpe and has high drawdowns. BUT it is one of the very few things you can say for sure has positive premia" [X Link](https://x.com/systematicls/status/1799037586184540213) 2024-06-07T11:15Z [----] followers, [---] engagements "@pedma7 you can try adding carry to your repertoire next It also has positive premia higher sharpe has a lower drawdown but requires you to manage margin" [X Link](https://x.com/systematicls/status/1799037923947667702) 2024-06-07T11:17Z [----] followers, [---] engagements "Why are there so little interest in prediction markets Why have they traditionally failed to garner any significant interest" [X Link](https://x.com/systematicls/status/1799089430214619592) 2024-06-07T14:42Z [----] followers, [----] engagements "Roaring Kitty's biggest flaw is unveiling the mystique he had built up over the [--] years and showing everyone that there was no magic after all. Only a man and his dream to be a billionaire through a meme" [X Link](https://x.com/systematicls/status/1799336330507501666) 2024-06-08T07:03Z [----] followers, [----] engagements "If I were his game theorist I would not have recommended that move. He did not backtest live streaming before using it in production. Typically strategy drift and copium in production" [X Link](https://x.com/systematicls/status/1799339463711306169) 2024-06-08T07:15Z [----] followers, [----] engagements "@vitrovic1 This is interesting we were watching too. And was a learning/point-of-interest to us. I don't know what dataset the "algos" were using we don't have anything that has such high fidelity. It's probably some raw live datafeed. https://x.com/systematicls/status/1799121040192242157 So interesting that the instant he said "lose it all" prices dropped by [---] bps. https://x.com/systematicls/status/1799121040192242157 So interesting that the instant he said "lose it all" prices dropped by [---] bps" [X Link](https://x.com/systematicls/status/1799342822065410547) 2024-06-08T07:28Z [----] followers, [---] engagements "@vvsmoe Sorry do you mean people speculated that he had some plans Or that he announced somewhere that he had some plans" [X Link](https://x.com/systematicls/status/1799350026772328819) 2024-06-08T07:57Z [----] followers, [---] engagements "@quant_spence @TheFlowHorse @0xfdf I'm just another dumbass larp" [X Link](https://x.com/systematicls/status/1799470599829704782) 2024-06-08T15:56Z [----] followers, [--] engagements "You can bet on a global portfolio of [-----] stocks that stocks with increasing dividend yield will have higher relative returns while stocks with decreasing dividend yield will have lower relative returns. You would express such an idea as dividend_yield_delta - ts_zscore(dividend_yield 63) ptfl_mean - cs_mean(dividend_yield_delta market) cs_dividend_yield_signal - dividend_yield_delta - ptfl_mean" [X Link](https://x.com/systematicls/status/1799935418651275377) 2024-06-09T22:43Z [----] followers, [---] engagements "You need to get the factor mimicking portfolio (the weights of the factor portfolio) for the characteristic portfolio approach you need to estimate the fmp. If you are do fama-french N-sorting you get the fmp as a result. Once you have the fmp and the factor returns you can regress to find your exposures and hedging is then -exposure*fmp" [X Link](https://x.com/systematicls/status/1800052607530221880) 2024-06-10T06:29Z [----] followers, [---] engagements "@wiltonhotz Do note that how you construct your factors will affect your exposures your hedging and the return characteristics you see. @CliffordAsness's AQR has a good paper on this: https://www.aqr.com/Insights/Research/Journal-Article/Craftsmanship-Alpha-An-Application-to-Style-Investing https://www.aqr.com/Insights/Research/Journal-Article/Craftsmanship-Alpha-An-Application-to-Style-Investing" [X Link](https://x.com/systematicls/status/1800054471537688752) 2024-06-10T06:36Z [----] followers, [---] engagements "@itsalexjune @__paleologo **Talking about** is the key phrase here" [X Link](https://x.com/systematicls/status/1800065191461450044) 2024-06-10T07:19Z [----] followers, [---] engagements "@HPierre_ @__paleologo Infeasible area" [X Link](https://x.com/systematicls/status/1800246271078547583) 2024-06-10T19:18Z [----] followers, [----] engagements "In the event that your corr(signal returns_t+1) is held constant across your universe this is true Which is fascinating because your lower bound will be held constant while your upper bound will increase basically giving you more opportunities to juice your returns. Very intuitive" [X Link](https://x.com/systematicls/status/1800260353961685160) 2024-06-10T20:14Z [----] followers, [----] engagements "Fascinating read. A summary in a picture: https://www.bloomberg.com/news/features/2024-06-06/sam-altman-bent-the-world-to-his-will-long-before-chatgpt-s-rise https://www.bloomberg.com/news/features/2024-06-06/sam-altman-bent-the-world-to-his-will-long-before-chatgpt-s-rise https://www.bloomberg.com/news/features/2024-06-06/sam-altman-bent-the-world-to-his-will-long-before-chatgpt-s-rise https://www.bloomberg.com/news/features/2024-06-06/sam-altman-bent-the-world-to-his-will-long-before-chatgpt-s-rise" [X Link](https://x.com/systematicls/status/1800501834739863972) 2024-06-11T12:14Z [----] followers, [----] engagements "@therobotjames years oops I meant this. He's a very niche very indie artist. More famous for playing games (Rust) than he is a musician but he uses his soundtracks in his videos. https://open.spotify.com/track/3ShromO1Sf75HXGFABMNlU https://open.spotify.com/track/3ShromO1Sf75HXGFABMNlU" [X Link](https://x.com/systematicls/status/1800757574758977623) 2024-06-12T05:10Z [----] followers, [---] engagements "Unironically true for strategies in quant finance as well. Strategies get crowded after they do well as they become the new hot thing and then they eventually collapse under their weight and deflate. In the collapse the space clears up and a new pecking order emerges. If your opponents are opportunists one way to beat them is to outlast them. Opportunists almost by definition lack staying power. If your opponents are opportunists one way to beat them is to outlast them. Opportunists almost by definition lack staying power" [X Link](https://x.com/systematicls/status/1802096376970695156) 2024-06-15T21:50Z [----] followers, [----] engagements "@__paleologo book Advanced Portfolio Management is my go-to recommendation for learning more about factors with a good balance between intuition and rigor. There is also material on hedging in the appendix. For a very light (and hopefully intuitive) introduction to factors you can refer to a recent article: https://x.com/systematicls/status/1799941636618867011 https://x.com/systematicls/status/1799941636618867011" [X Link](https://x.com/systematicls/status/1802690600871063893) 2024-06-17T13:11Z [----] followers, [----] engagements "All examples here can be trivially reproduced. Data used here is "retail grade" data from Polygon but it shouldn't matter even if its plucked from yfinance because its OHCLV data for Sp500 constituents. The banks variable is actually dollar volume as in banks=close*volume. Lastly .shift(2) is used because I cannot assume being able to get my entire position at the close. Having an entire day to vwap into my position is more realistic for me. If you are retail trading less than 1m you can probably use .shift() for your signals" [X Link](https://x.com/systematicls/status/1802811642876711115) 2024-06-17T21:12Z [----] followers, [----] engagements "Redbone by LAV8 on spotify is real musical alpha for coding" [X Link](https://x.com/systematicls/status/1803329579387646137) 2024-06-19T07:30Z [----] followers, [----] engagements "@__paleologo The time series approach is useful for high-complexity factors where its not trivial to define a known characteristic for an instrument e.g. index-rebalness" [X Link](https://x.com/systematicls/status/1803346255139279280) 2024-06-19T08:37Z [----] followers, [----] engagements "It is a factor. Everything that has been discussed is a factor. But these factors differ in construction. There are [--] ways to calculate factors: [--]. Time Series Where you have the factor's portfolio and therefore returns and you need to estimate the instrument's loadings on the factor. [--]. Fundamental Where you have each instrument's loading to the factor and you need estimate the factor's returns and from returns back out its portfolio. --- Then there are statistical factors where you are doing some kind of statistical method to back out latent factors from the returns e.g. PCA" [X Link](https://x.com/systematicls/status/1803442983444025458) 2024-06-19T15:01Z [----] followers, [---] engagements "@macrocephalopod @0xfdf @SmityAndre82456 @__paleologo @palash_i_am This means you can only hedge by doing some kind of optimization on your original portfolio right - what gappy refers to as "internal hedging" where your factor exposures are a part of the constraint/objective" [X Link](https://x.com/systematicls/status/1803446956490104974) 2024-06-19T15:17Z [----] followers, [----] engagements "@macrocephalopod @0xfdf @SmityAndre82456 @__paleologo @palash_i_am I guess this works I don't particularly like it though. I've found the extra constraints make the portfolio worse in testing" [X Link](https://x.com/systematicls/status/1803447723817107771) 2024-06-19T15:20Z [----] followers, [----] engagements "The one sticky use-case for blockchains that have actual PMF is a decentralized marketplace of all shapes and sizes: amms orderbooks gambling prediction markets etc. I think it is only natural to imagine that the space will continue doubling down on it and trying to expand vertically and horizontally into the domain of TradFi. Liquidity also tends to consolidate there is a strong winner takes all phenomenon in exchanges/marketplaces. The first one who can crack RWAs at scale will likely have a significant advantage at venturing into more and more esoteric offerings" [X Link](https://x.com/systematicls/status/1803708412980846609) 2024-06-20T08:36Z [----] followers, [----] engagements "Have just begun to read it and already a banger. Foresee this to be the new reference book to all the materials needed in the quant landscape wer alpha tho MAJOR DROP: "The Element of Quantitative Investing" latest book draft. Links at the end of the 🧵. Short introduction: I am writing a book on quantitative investing for Wiley (April 2025). It's the book I wanted to write all along but I strayed and wrote another book first. MAJOR DROP: "The Element of Quantitative Investing" latest book draft. Links at the end of the 🧵. Short introduction: I am writing a book on quantitative investing for" [X Link](https://x.com/systematicls/status/1804395833879843133) 2024-06-22T06:07Z [----] followers, [----] engagements "@ScottPh77711570 You might be interested in such frameworks for crypto flow around special sits" [X Link](https://x.com/systematicls/status/1804405187123314874) 2024-06-22T06:44Z [----] followers, [----] engagements "@ScottPh77711570 If you think about it even DEXes in new chains are subject to these binary events. They either make it have a winner's take all effect and have their valuations 100x or don't have [--] liquidity and eventually die" [X Link](https://x.com/systematicls/status/1804408061387518335) 2024-06-22T06:56Z [----] followers, [----] engagements "Can construct a market-neutral long trump short everyone else to collapse the reality spread. If you're risk-averse long trump biden short everyone else to collapse the illusion of choice spread. Betfair US election probabilities - 58% Trump 17% Biden 19% Newsom Harris Obama or Clinton 5% Anyone else Kind of amazing. Betfair US election probabilities - 58% Trump 17% Biden 19% Newsom Harris Obama or Clinton 5% Anyone else Kind of amazing" [X Link](https://x.com/systematicls/status/1806607483705946210) 2024-06-28T08:36Z [----] followers, [----] engagements "I remember in [--] the institutional equities guys were all laughing at the crypto guys for how manic the crypto markets were. Funny how no one escapes the meme mania" [X Link](https://x.com/systematicls/status/1806817756215730655) 2024-06-28T22:31Z [----] followers, [----] engagements "Can't agree with this take. I'm fairly certain Ken Griffin can get into Citadel. The point is never to find someone perfect in CS math and finance. But to look for a genuine and deep interest in at least one of them as a signal that they have a strong desire for curiousity AND competence. One of the most outdated or mid curve things in quant is the talent selection process. No one can be perfect in CS math and finance. I am not sure if Ken Griffin can get into citadel now. One of the most outdated or mid curve things in quant is the talent selection process. No one can be perfect in CS math" [X Link](https://x.com/systematicls/status/1806860035093680248) 2024-06-29T01:19Z [----] followers, 58K engagements "@blueprintsmb22 @JaredKubin understood apologies" [X Link](https://x.com/systematicls/status/1807368379327713548) 2024-06-30T10:59Z [----] followers, [--] engagements "For those in the game what are popular institutional datasets in crypto Price is not an issue. E.g. Is ravenpack still the go-to news provider for crypto" [X Link](https://x.com/systematicls/status/1807417600818708768) 2024-06-30T14:15Z [----] followers, 22K engagements "To make sure I get my message across. My original post: --- Imagine selling a course on "end-to-end infrastructure" for "complex alpha representations" so you have "all you need to begin quantitative trading". Only to run a funding-rate arb bot. What people sell tells you a lot about what they believe is profitable. (hint: they don't sell anything they believe is profitable) The fact he blocked me right after tells you a lot about the validity of my post. I assume most of my followers are smart rational people. Let's have a simple thought experiment. You discover an investment process that" [X Link](https://x.com/systematicls/status/1808076484017639476) 2024-07-02T09:53Z [----] followers, 45.7K engagements "nothing wrong with teaching funding rate arbs His post was about his own funding rate arb implementation which I found funny because [--]. His entire content is teaching people to build a system for backtesting and alpha discovery [--]. In the same post he talked about moving off manual PnL recording into a .csv file. I'm not sure someone who found it more convenient to manually record PnL should be teaching courses on automation" [X Link](https://x.com/systematicls/status/1808150371124367396) 2024-07-02T14:46Z [----] followers, [----] engagements "@TimMeggs @spxudi @quant_arb @TimMeggs Gave me a great breakdown of each of the popular vendors and their pros and cons. I think he has a strong understanding of what each competitor's strengths and weaknesses are and how they compare to his product" [X Link](https://x.com/systematicls/status/1808155330809061466) 2024-07-02T15:06Z [----] followers, [----] engagements "@m1ha313 @quant_arb HFT Infra for ML How does that work (genuine question)" [X Link](https://x.com/systematicls/status/1808448580048584764) 2024-07-03T10:31Z [----] followers, [---] engagements "@m1ha313 @quant_arb hahaha understood" [X Link](https://x.com/systematicls/status/1808449483468079491) 2024-07-03T10:35Z [----] followers, [--] engagements "Let me tell you guys its shocking how many MFT equity statarb teams are up 3% on GMV YTD. The only question is. are you up 3% on GMV with [--] sharpe (underperformer) [--] sharpe (average) or [--] sharpe (best-in-class)" [X Link](https://x.com/systematicls/status/1808825177398260090) 2024-07-04T11:28Z [----] followers, 17.4K engagements "@BearNecessiti15 I've always wondered what its like to play poker with weak currencies and huge denominations. All in. [--] million (they normally convert to cr) rupees Must feel good to say" [X Link](https://x.com/systematicls/status/1815658184310407621) 2024-07-23T08:00Z [----] followers, [----] engagements "@rosswern @macrocephalopod Very interesting. I've not looked at KR before because its relatively thin compared to HK/IN. This is great thank you" [X Link](https://x.com/systematicls/status/1815671193556463762) 2024-07-23T08:52Z [----] followers, [---] engagements "Perspective on running a global book seems to be interesting to you guys. Heres one more - there are two ways to run a global statarb book. You can run every country as a seperate portfolio and so all your signals are country-wide signals and you can't make bets on cross-country patterns and can't easily hedge global factors. Doing it this way is easy to manage and requires no further thought. However if you are a mega chadster the other way you can run a global book is to trade all [-----] stocks in a single portfolio. You can now make bets on global patterns and hedge out global asset-class" [X Link](https://x.com/systematicls/status/1815714654699872306) 2024-07-23T11:44Z [----] followers, 14.2K engagements "Yes but then you don't have issues of your signals being misaligned cross-sectionally across countries. Lets try another rather than value lets imagine cross-sectional momentum instead. If you ranked your signal across the globe for cs-momentum it would be a substantially different portfolio from ranking your signals country-wide and concatening them together" [X Link](https://x.com/systematicls/status/1815723202917851298) 2024-07-23T12:18Z [----] followers, [---] engagements "I think you're getting stuck on the trading aspect the salient point is the signal aspect. Are your signals global Or are they country-wide. If they are global then you have to deal with implementation shortfalls of your signals across the different countries. Not thinking about this is catastrophic. If they are country-wide then you can combine them into a single portfolio or trade them separately I don't think the difference is catastrophic" [X Link](https://x.com/systematicls/status/1815724908623851845) 2024-07-23T12:25Z [----] followers, [---] engagements "@0xfdf @IronEconomist @macrocephalopod AFAIK there are many many competent statarb teams that have never created global signals in their lifetime" [X Link](https://x.com/systematicls/status/1815726952562741276) 2024-07-23T12:33Z [----] followers, [---] engagements "@macrocephalopod hey maybe I furiously plant trees on the weekends you never know mate" [X Link](https://x.com/systematicls/status/1816375092479418620) 2024-07-25T07:29Z [----] followers, [----] engagements "It seems like analytics and optimizations of portfolios and risks are the truest expressions of quant skills. I suppose that makes enterprise risk managers quintessential quants. There are so much of a portfolio manager's skillset that have nothing to do with expressing quant skills but I would consider to be integral in uncovering alpha (e.g. project management deciding the search path along the alpha search space)" [X Link](https://x.com/systematicls/status/1817111853195268166) 2024-07-27T08:16Z [----] followers, 10.4K engagements "Because short + long index is a vastly inferior product to long/short. For the simple reason that you are only capturing short-side alpha and forgoing long-side alpha AND your long index cannot hedge out your factor exposures on the short side efficiently I dont understand why there are not more short only hedge funds Own index short get paid for alpha. Median stock does not create much value. This chart is exaggerated but upside is narrow Used to do generalist short selling & loved it. Wish that was still a thing I dont understand why there are not more short only hedge funds Own index short" [X Link](https://x.com/systematicls/status/1817426418030305591) 2024-07-28T05:06Z [----] followers, 18.3K engagements "My job is to make the long-run EV calculation using a flawed model given asymmetric information right Yes I know you are passionate and would kill for a chance to be in XXX I know you would want to stay with me till the end of the time of course you would never leave. Of course you'll make the team 100M PnL in your first year. I'm going to assume you are going to come in for 2-3 years learn the know-how contribute fuck-all on average and leave. I'm going to put a hair-cut on everything you tell me. Your modal contribution is going to be [--] post tx cost. And so my modal answer is going to be" [X Link](https://x.com/systematicls/status/1817470994224930926) 2024-07-28T08:03Z [----] followers, 13.1K engagements "You can tell someone's seniority by when they go to the gym. Before and after work Not in management. After lunch Likely senior/management but not top echelon. Comes in at 1pm in running shoes and changes straight into a suit CXO" [X Link](https://x.com/systematicls/status/1817846201276117002) 2024-07-29T08:54Z [----] followers, 19.8K engagements "Quote: Practice. Thats all it is. If you practice enough you can sense where the open ice will be. Watching the three men interact I could not help but notice the banality of genius. It occurred to me that Carbon did not have any superpowers beyond its boring and total efficiency. People worked like machines: they made goals they accomplished them this was the genius of following through. There would be no deliberations no excuses. There was will followed by action. Mean it do it. There was no such thing as a slump. But I wondered if the genius here could also be the horror. You might through" [X Link](https://x.com/systematicls/status/1819063824483602710) 2024-08-01T17:33Z [----] followers, [--] engagements "Gappy woke up today and chose violence. But +1 quants generally stay away from economists and more generally economics/business/finance graduates. If you are a student and want to be a quant math/physics/cs is the way to go @stevehouf Whether economists make great traders or make money in markets actually doesn't say anything about the correctness of the economic models. Youd expect that *some* economists would at least make good traders/PMs or risk managers. Where are they Geneakoplos Thaler Fama @stevehouf Whether economists make great traders or make money in markets actually doesn't say" [X Link](https://x.com/systematicls/status/1820368433315246564) 2024-08-05T07:57Z [----] followers, 21.7K engagements "I think the highest density are engineers/physics/math/cs graduates" [X Link](https://x.com/systematicls/status/1820369104299061572) 2024-08-05T07:59Z [----] followers, [----] engagements "@kelxyz_ I think having a MSc in a technical discipline is very advantageous to be a quant. It does have some signal. I can only speak for this industry and not for others" [X Link](https://x.com/systematicls/status/1820375061661941774) 2024-08-05T08:23Z [----] followers, [---] engagements ""Hedge funds have underperformed the market. Hedge funds have underperformed the market." John sobs quietly to himself as his global long-only ETF portfolio gets decimated. Meanwhile - how's just another Monday for all my L/S broskis - better check that size and momentum exposure and we should be good to go" [X Link](https://x.com/systematicls/status/1820475589029175799) 2024-08-05T15:02Z [----] followers, [----] engagements "How does FinTwit feel about platforms like https://numer.ai/home https://numer.ai/home" [X Link](https://x.com/systematicls/status/1822514372578890171) 2024-08-11T06:04Z 10.3K followers, [----] engagements "So the main reason I'm asking this is because I'm curious why none of these have achieved significant scale - some have come and gone like Quantopian. Some have come and never managed a significant amount of capital and most have fizzled out of the news" [X Link](https://x.com/systematicls/status/1822548467044241672) 2024-08-11T08:19Z 10.3K followers, [----] engagements "If I wasn't a quant and if all whimsical jobs in an advanced society were equally financially rewarding so there was nothing to min-max one of the more interesting jobs is to go around the world and look for very interesting hero journeys (a la campbell)) or people with exceptional character and document all the features that make them the way they are. Eventually all features will either be endogenous or exogenous. Create a test to find endogenous traits and create a school to assert exogenous traits. Raise money to go to subsaharan africa and find all impoverished children who match the" [X Link](https://x.com/systematicls/status/1823135260646973508) 2024-08-12T23:11Z 10.3K followers, 15K engagements "@PinguCapital @pleaseconsider4 @__paleologo Here's some plain speak: you are antiquated. The best discretionary PMs hire quants in their teams today precisely because they have learnt that a quant overlay to guide portfolio management decisions significantly improves performance" [X Link](https://x.com/systematicls/status/1826817114113732807) 2024-08-23T03:01Z 10.3K followers, [---] engagements "Small timer wonders why he can't break into the big leagues. #Quants are evolving into religious proselytizers with their equations like a new gospel in finance. Theres nothing new about using maths when trading or investing. History shows theres been many extremely successful investors/traders before quants became mainstream. #Quants are evolving into religious proselytizers with their equations like a new gospel in finance. Theres nothing new about using maths when trading or investing. History shows theres been many extremely successful investors/traders before quants became mainstream" [X Link](https://x.com/systematicls/status/1826817344708247617) 2024-08-23T03:02Z 10.3K followers, [----] engagements "@PinguCapital Yeah I know some guys who call their $10K PA a "family office" too" [X Link](https://x.com/systematicls/status/1826829348613357911) 2024-08-23T03:50Z 10.3K followers, [---] engagements "If we are to carve Olympus into our image it will not be from one mighty blow but a million chips to the surface - each insignificant but in totality world-changing" [X Link](https://x.com/systematicls/status/1828547250911695003) 2024-08-27T21:36Z 10.3K followers, [----] engagements "Can't speak for all careers obviously but at least for a traditional quant career the first few years are slow but normally you'll get to a point where PnL becomes positive and attributable and it's a much more comfortable ride from there on out. Your purchasing power should outpace everything short of financial armageddon" [X Link](https://x.com/systematicls/status/1830397323475824753) 2024-09-02T00:08Z 10.3K followers, 21.6K engagements "How does one start building systematic strategies for these markets Where is a good place to get normalized data Being able to trade all the states of the world a little step at a time. https://t.co/OelD7Jg21j Being able to trade all the states of the world a little step at a time. https://t.co/OelD7Jg21j" [X Link](https://x.com/systematicls/status/1833381459064983743) 2024-09-10T05:46Z 10.3K followers, 11.9K engagements "@saplash_i_am I think that depends on the driver right. If two bets are implicitly betting on let's say the health of joe biden then theoretically it should hold even at the smallest discrete timesteps - any deviation is a statistical arbitrage opportunity" [X Link](https://x.com/systematicls/status/1833394644086100259) 2024-09-10T06:38Z 10.3K followers, [---] engagements "Taking a poll on behalf of a friend - A start-up fintech company has a novel idea at the end game it wants to crowdsource portfolio managers and allow thousands of people to participate in the upside of the financial markets while risking no personal capital. Here are the mechanisms: [--]. You first have to participate on their crowdsourcing platform to submit signals. These signals are very simple mathematical models that take about 1-60 minutes to create one. You are paid for each signal you submit. If the firm uses your signals in live trading you are paid a profit share on the PnL as well." [X Link](https://x.com/systematicls/status/1834834494182113561) 2024-09-14T06:00Z 10.3K followers, 13K engagements "Every hopeful quant research candidate in every interview: "You don't bring any special skillsets to our team. Seems like a generic profile and we only want the best." Candidate: "Not me there's only one of me in all the world. I'm one in a krillion https://www.youtube.com/watchv=01Runa3neb0 https://www.youtube.com/watchv=01Runa3neb0" [X Link](https://x.com/systematicls/status/1836040241192612193) 2024-09-17T13:51Z 10.3K followers, [----] engagements "@0xFaust12 @choffstein @0xfdf You sound like you work in Citadel coverage" [X Link](https://x.com/systematicls/status/1836632302895427854) 2024-09-19T05:03Z 10.3K followers, [---] engagements "I'm curious to know what is the intersection of the demand for quant knowledge. Imagine I've stumbled onto an infinite article factory. Comment below on every quant finance article you wish you had access to and I might be able to make it happen" [X Link](https://x.com/systematicls/status/1836679295621828767) 2024-09-19T08:10Z 10.3K followers, 22.3K engagements "@__paleologo @ScottPh77711570 @pedma7 @macrocephalopod @buffetsalpha @therobotjames Based on all my time on X that's basically Scott's game" [X Link](https://x.com/systematicls/status/1837070172437500230) 2024-09-20T10:03Z 10.3K followers, [---] engagements "@ScottPh77711570 @__paleologo @pedma7 @macrocephalopod @buffetsalpha @therobotjames never better honestly thanks for asking. heard you were dishing out life saving advice to a certain oxbridge quant" [X Link](https://x.com/systematicls/status/1837071155389210817) 2024-09-20T10:07Z 10.3K followers, [---] engagements "@pedma7 @__paleologo @ScottPh77711570 @macrocephalopod @buffetsalpha @therobotjames Is there still premium to harvest in ICOs I remember there was a study where if you went long anything on Binance Launchpad and exited [--] hours later it was like a 300% returns strategy. Was not true for other launchpads" [X Link](https://x.com/systematicls/status/1837072140962779411) 2024-09-20T10:11Z 10.3K followers, [---] engagements "@pedma7 @ScottPh77711570 @__paleologo @macrocephalopod @buffetsalpha @therobotjames @__paleologo Imo this [--] tweets tell you everything you need to know for the retail quant crowd. At least the ones who aren't steeped in the institutional L/S stuff - my anecdotal experience is that most of them have become masochists and have learnt to embrace pain" [X Link](https://x.com/systematicls/status/1837072978598842759) 2024-09-20T10:15Z 10.3K followers, [---] engagements "@ScottPh77711570 @pedma7 @__paleologo @macrocephalopod @buffetsalpha @therobotjames I don't even want to click on http://Dumpy.Fun http://Dumpy.Fun" [X Link](https://x.com/systematicls/status/1837073789395554627) 2024-09-20T10:18Z 10.3K followers, [---] engagements "@friendscallmeap @ScottPh77711570 @pedma7 @__paleologo @macrocephalopod @buffetsalpha @therobotjames Have you guys checked out RWA perps vs CeFI exchanges @kaledora's @OstiumLabs is trying to put RWA perps on-chain. Might have some alpha there not sure what's the institutional participation like" [X Link](https://x.com/systematicls/status/1837074655984898558) 2024-09-20T10:21Z 10.3K followers, [---] engagements "PM: "Our team moves fast we can put risk on your ideas very early if you demonstrate potential" QR: Waiting for risks to be put on his ideas after demonstrating potential" [X Link](https://x.com/systematicls/status/1838051178065907969) 2024-09-23T03:02Z 10.3K followers, [----] engagements "It's a good read - at least worth thinking about for retail portfolios. For quants from my understanding I can summarize "return stacking": [--]. Portfolio optimisation exercise where you have at least one alpha generating strategy/manager lets call it A1. [--]. Hedge all factors from A1 you don't want. [--]. Add exposures to A1 that you do want. Use levered exposures for capital efficiency. They've combined [--] and [--] and called it "beta swapping". [--]. Combining [--] and [--] is "returns stacking". It doesn't seem like stacking anything here because we start from A1 but the article implicitly assumes starting" [X Link](https://x.com/systematicls/status/1839145373043941481) 2024-09-26T03:30Z 10.8K followers, [----] engagements ""Never give in. Never give in. Never never never neverin nothing great or small large or pettynever give in except to convictions of honour and good sense. Never yield to force. Never yield to the apparently overwhelming might of the enemy."" [X Link](https://x.com/systematicls/status/1843990464808222737) 2024-10-09T12:22Z 10.3K followers, [----] engagements "On the mindless pursuit of power. https://youtu.be/w_MSFkZHNi4 https://youtu.be/w_MSFkZHNi4" [X Link](https://x.com/systematicls/status/1849104669039284535) 2024-10-23T15:04Z 13.9K followers, [----] engagements "Learnt a ton from this podcast about running large models on GPUs. https://signalsandthreads.com/the-uncertain-art-of-accelerating-ml-models/ https://signalsandthreads.com/the-uncertain-art-of-accelerating-ml-models/" [X Link](https://x.com/systematicls/status/1850727893044965872) 2024-10-28T02:34Z 13.9K followers, [----] engagements "In [----] if DL in finance still doesn't work as far as you know then you don't really know a lot @systematicls Well what you're saying makes sense in theory but it doesn't work in practice as far as I know. Some of the best firms have tried to make it work but the problem is that deep learning requires a lot of data to avoid overfitting. This problem exists even with high-frequency @systematicls Well what you're saying makes sense in theory but it doesn't work in practice as far as I know. Some of the best firms have tried to make it work but the problem is that deep learning requires a lot of" [X Link](https://x.com/systematicls/status/1866342067624182117) 2024-12-10T04:39Z 10.5K followers, 16.5K engagements "Idiots on fintwit: no real firms uses DL in practice. Also firms:" [X Link](https://x.com/systematicls/status/1866850732017848751) 2024-12-11T14:21Z 10.7K followers, 23.4K engagements "@arnaud710 @macrocephalopod @quant_arb If what you are saying is literal and true it makes what he's done even more impressive. And I'm not being kind on larps - I mean guys lets be objective here" [X Link](https://x.com/systematicls/status/1867237080000340244) 2024-12-12T15:56Z 10.7K followers, [----] engagements "Just for the uninitiated. When convering inr to usd. Lakh 1k usd Crore 100k usd You're welcome. i can tolerate lakhs but i draw the line at crore. wtf is a crore just tell me in normal numbers i can tolerate lakhs but i draw the line at crore. wtf is a crore just tell me in normal numbers" [X Link](https://x.com/systematicls/status/1870079466015519186) 2024-12-20T12:11Z 10.7K followers, [----] engagements "Playing POE2 and really liking the currency market especially as a market practitioner. In general every game that has a free market allows a patient speculator to profit either from trends or market making; but there is some nuance in every one of them. In POE2 the primary mode of exchange is a barter trade around valuable commodities like the one shown below. There is a currency market that facilitates this barter trade and it allows you to buy "5 exalts for [--] chaos". The most reliable method I've found so far to profit from the markets is to quote a spread around liquid markets (example" [X Link](https://x.com/systematicls/status/1875921840742551981) 2025-01-05T15:06Z 12.9K followers, 21.4K engagements "Land of the guiseppes and it was beautiful and delicious It was mind blowing to see thousands of people flock to appreciate work done thousands of years ago" [X Link](https://x.com/systematicls/status/1896167464188473822) 2025-03-02T11:55Z 10.9K followers, [----] engagements "An unironically clear guide on various arbitrages (mechanical arbitrages triangular arbitrages constituent arbitrages) through a game medium: https://maxroll.gg/poe2/resources/flipping-with-the-currency-exchange https://maxroll.gg/poe2/resources/flipping-with-the-currency-exchange" [X Link](https://x.com/systematicls/status/1903692890443850063) 2025-03-23T06:18Z 11.3K followers, [----] engagements "Guessing from a portfolio standpoint: If you knew with certainty (P=1) how the stock price of a single instrument will move tomorrow the rational optimal thing to do is to borrow from every outlet you can and bet the farm house and tomorrow's farm and house on it. The more certain you are the less you want to dilute that alpha. Even in the context of having [--] stocks with P=1 you would just pick the one with higher abs returns; in cases of complete information there is no need for diversification. However when P is small diversification becomes the only free lunch. Noise gets averaged to 0" [X Link](https://x.com/systematicls/status/1924356671255761189) 2025-05-19T06:49Z 12.2K followers, 19.9K engagements "Poetry Patriotism does not come naturally to meI have always been far too fond of the foreign. Aesthetically I yearn for the rich history and beautiful architecture of Europe; politically I dream of the competent governance and high state capacity of Singapore or Taiwan. I am a Patriotism does not come naturally to meI have always been far too fond of the foreign. Aesthetically I yearn for the rich history and beautiful architecture of Europe; politically I dream of the competent governance and high state capacity of Singapore or Taiwan. I am a" [X Link](https://x.com/systematicls/status/1941118705817297179) 2025-07-04T12:55Z 12.7K followers, [----] engagements "Somehow I think the young white collars are underestimating how devasting AI is going to be in their job prospects. At least in the "short run" until we all become redundant. I foresee that the first tranche of jobs that are going to be impacted are the junior white collars that are traditionally the byproduct of being a median performer of a fairly technical degree (e.g. your 50th to 90th percentile CS grad). These median performers who would otherwise end up with jobs at SMEs with simple requirements for "run-of-the-mill computer jobs" will be the first to have their place in the economy" [X Link](https://x.com/systematicls/status/1941895906435158191) 2025-07-06T16:23Z 12.8K followers, 12.4K engagements "PSA: Do not put skilled in excel or list excel as your first skill when applying for quant researcher roles. Its a negative signal that suggests you are probably going to be a bad swe" [X Link](https://x.com/systematicls/status/1942508964106932572) 2025-07-08T09:00Z 12.8K followers, [----] engagements "@coastal_quant Post more on this please" [X Link](https://x.com/systematicls/status/1957203447402680789) 2025-08-17T22:10Z 12.9K followers, [---] engagements "If you want to have a finger on how the median US statistical arbitrage pod is doing just take a look at some heuristic of w_mom * momentum factor + w_size * -size factor. Momentum is down AND size is up It's a quant quake" [X Link](https://x.com/systematicls/status/1959603281396924688) 2025-08-24T13:06Z 12.9K followers, 25.7K engagements "Quality LinkedIn interactions as usual" [X Link](https://x.com/systematicls/status/1963494529685672319) 2025-09-04T06:49Z 13.7K followers, 11.9K engagements "The world does not hand crowns to the idle nor mercy to the weak" [X Link](https://x.com/systematicls/status/1974048673693880592) 2025-10-03T09:47Z 14K followers, [----] engagements "Wisdom is wasted on the old they best they can do is to try to give it away to oft unwilling receivers" [X Link](https://x.com/systematicls/status/1974673572330393694) 2025-10-05T03:10Z 14K followers, [----] engagements Limited data mode. Full metrics available with subscription: lunarcrush.com/pricing
@systematicls systematic longshort
systematic longshort posts on X about if you, money, stocks, in the the most. They currently have [------] followers and [---] posts still getting attention that total [-------] engagements in the last [--] hours.
Engagements: [-------] #
- [--] Week [---------] +1,770%
- [--] Month [---------] -77%
- [--] Months [----------] +5,553%
- [--] Year [----------] +603%
Mentions: [--] #
- [--] Week [--] -48%
- [--] Month [---] -51%
- [--] Months [---] +2,925%
- [--] Year [---] +168%
Followers: [------] #
- [--] Week [------] +0.37%
- [--] Month [------] +2.90%
- [--] Months [------] +298%
- [--] Year [------] +374%
CreatorRank: [-------] #
Social Influence
Social category influence finance technology brands cryptocurrencies social networks stocks countries exchanges celebrities currencies vc firms
Social topic influence if you, money, stocks, in the, ai, crypto, investment, hedging, finance, quant
Top accounts mentioned or mentioned by @paleologo @0xfdf @scottph77711570 @quantarb @therobotjames @macrocephalopod @pedma7 @friendscallmeap @cryptohound666 @oxbridgequant @quantymacro @experquisite @buffetsalpha @maruushae @elletwocache @palashiam @coastalquant @paleologos @movietimedev @agustinlebron3
Top assets mentioned Quant (QNT) Alpha Technology Group Limited (ATGL) Bitcoin (BTC) Alphabet Inc Class A (GOOGL) Premia (PREMIA) Robinhood Markets, Inc. (HOOD) CITADEL (CITADEL) CVS Health Corp (CVS)
Top Social Posts
Top posts by engagements in the last [--] hours
"huge proponent of building own projects around the entire pipeline of trading if you're a qr or aspiring qr. often times you will be siloed into specific responsibilities (e.g. alpha generation data cleaning etc); personal projects round out the skillsets. This is not far off. Time spent has more effect when it's spent working on your own projects. https://t.co/lsNyRzPzsb This is not far off. Time spent has more effect when it's spent working on your own projects. https://t.co/lsNyRzPzsb"
X Link 2024-05-04T05:59Z [----] followers, [---] engagements
"@therobotjames @_deepdiveStocks if you look at the most illiquid (by adv) penny stocks you might conclude that their variance is low because they are simply not trading. where the analysis fails is that the impact from actually trading will lead to disproportionate variance above his static analysis"
X Link 2024-05-07T05:01Z [--] followers, [--] engagements
"@macrocephalopod @ScottPh77711570 @PtrPomorski How are you guys walking around with no market-neutral mandate. I can barely get 5% delta exposure because my phone rings and risk is on the line"
X Link 2024-05-08T03:49Z [--] followers, [--] engagements
"@ScottPh77711570 @macrocephalopod @PtrPomorski i mean when ure retail you're not charging yourself 2/20. so. returns are returns enjoy the risk premia"
X Link 2024-05-08T04:07Z [--] followers, [--] engagements
"@ScottPh77711570 @therobotjames @SpacChad @readysetliqd @notgodcomplex96 @liquiditygoblin @weaponizedFOMO found an interview with @therobotjames https://www.youtube.com/shorts/HAsemyRhNj8 https://www.youtube.com/shorts/HAsemyRhNj8"
X Link 2024-05-10T07:56Z [--] followers, [--] engagements
"@ScottPh77711570 @Larryjamieson_ What made you decide to move to Thailand Was it just to stretch the US dollar you got from trading"
X Link 2024-05-11T09:05Z [---] followers, [---] engagements
"is it discretionary trading still if an AI does it or is it black-box systematic trading from https://ibkrcampus.com/ibkr-quant-news/dr-ernest-chans-journey-the-promise-and-limits-of-ai-and-machine-learning-for-investing/ https://ibkrcampus.com/ibkr-quant-news/dr-ernest-chans-journey-the-promise-and-limits-of-ai-and-machine-learning-for-investing/"
X Link 2024-05-11T19:13Z [----] followers, [---] engagements
"The intern's signal signal - dataset1_feature1 + dataset1_feature2 + dataset1_feature3 + dataset1_feature4 + dataset1_feature5 + . + dataset1_feature20 + dataset1_feature21 no operators no transformations just linear addition of 20+ features from the same dataset"
X Link 2024-05-12T07:06Z [----] followers, 28.7K engagements
"@arnaud710 keeping it simple - if not data driven features should ideally be causal to returns e.g. we have reason to believe that stocks with increasing dividend yield have greater demand and therefore drive prices up signal - ts_delta(dividend_yield 90) 1/2"
X Link 2024-05-12T16:34Z [----] followers, [----] engagements
"Created this acct to vibe with the crypto crowd since i'm exploring a new asset class; been in equities my whole career. Then got 30k views from a thread about signal correlation. Seems like there's a demand for the ol quanty tweets. What should I shitpost (vaguely) about"
X Link 2024-05-13T02:08Z [----] followers, 14.5K engagements
"Anyone who has spent real R&D dollars on processes know that every discovery and knowledge has a cost. But there's still a way to glean knowledge from the people who have played in institutional grounds. 2/7"
X Link 2024-05-13T09:18Z [----] followers, [---] engagements
"Thankfully there are still [--] types of content I've found useful in my time lurking on twitter and listening to podcasts on spotify: [--]. Guys thinks everyone knows this [--]. Accidental alpha leaks [--]. Guy just really needs the world to think he is smart 3/7"
X Link 2024-05-13T09:18Z [----] followers, [----] engagements
"@ScottPh77711570 @__paleologo Yeah I thought you would really like it. He's not just a good practitioner but a good business owner and thinks about life from an extreme sports / special force vantage point. Very on the nose about considerations on risk reward factor hedging talent and leverage too"
X Link 2024-05-13T11:01Z [----] followers, [---] engagements
"Are there any shops where the "in-house proprietary MFM" isn't just a wrapper around Barra/Axioma + (2-3 in-house models)"
X Link 2024-05-14T17:39Z [----] followers, [----] engagements
"@bzzz_anon And what did u tell them"
X Link 2024-05-14T18:34Z [----] followers, [---] engagements
"@JaredKubin So awesome. And they are being used for explainatory purposes ex-post or for hedging If for hedging have you benchmarked against Barra"
X Link 2024-05-15T18:43Z [----] followers, [---] engagements
"Extremely large and successful podshops pass-through all costs to investors. They only hurt when there are a small number of large PnL generating teams that cannot offset a large number of losing teams. Then they would have to still pay the positive teams according to the formulaic compensation and this payment would have to come from retained earnings/treasury. Extremely leveraged long/short portfolios are in vogue right now (e.g. pod shops). They rely heavily on stops to manage risk. Could the failure mode be rogue wave short squeezes where things just gap up hundreds of percent in a single"
X Link 2024-05-17T21:24Z [----] followers, 74K engagements
"Secondly risk limits exists not only on individual pods but as a firm as well. These risks are extremely well-distributed over a staggeringly large surface. For every team that is packing their things into a box after getting blown-up there is another team with equal and opposite positions popping champagne for having made their quarter. The probability of largewell-run podshops imploding from such undiversified carelessness is not zero but extremely minute"
X Link 2024-05-17T21:24Z [----] followers, [----] engagements
"I'm endeavoring to share some thoughts on the markets quantitative research portfolio management building a career and life amassed over more than a decade of risk-taking. There are plenty of great long-form content covering all of the above in great detail. My goal is not to compete with them but perhaps in as few words and as little equations as possible try to impart the salient point intuitively and then refer you to the long-form content for rigor if necessary. I believe that strong one-liners are more easily incorporated into heuristics than [--] pages of terse content and good heuristics"
X Link 2024-05-18T07:20Z [----] followers, 22.1K engagements
"My favorite part of this is you calling schonfeld mediocre. That's how I know you're a player. Also I think you are talking about two different points here crowding in delevering/hedging and withdrawal runs. In crowded deleverings most firms now want PMs to be orthogonal to some custom crowding factors; but I feel that misses the point. Hedgefunds become correlated to each other during systemic events BECAUSE we are all trying to hedge to the same factors. 95% of risk models are a copy of Barra/Axioma based on Barra/Axioma or benchmarked to Barra/Axioma. When we see exposures and form hedging"
X Link 2024-05-18T07:51Z [----] followers, [--] engagements
"@__paleologo @X [--] + [--] + [--] + [--] + [--] + . = -1/12"
X Link 2024-05-18T15:40Z [----] followers, [---] engagements
"http://x.com/i/article/1791446922177458178 http://x.com/i/article/1791446922177458178"
X Link 2024-05-19T11:16Z [----] followers, 70.4K engagements
"For those of you who have enjoyed the posts on ergodicity you would also enjoy Taleb's work - the source of my inspiration on the subject. I want to point out specifically [--] things that I think are worth paying attention to that I haven't mentioned in the accompanying post: [--]. The first thing I would like to point out is to take reference from Taleb's definition of "ensemble probability" where observations are independent of each other (1000 people going to casino) and "time probability" where observations have path dependency (1 person going to the casino [----] times). In backtesting a"
X Link 2024-05-21T20:27Z [----] followers, 26.9K engagements
"Should global equities be moved on-chain in the form of perpetuals Off CeFi exchange hours we could allow price discovery on-chain on CeFi exchange hours (30 mins post auction) we can implement funding rates to tether prices to the CeFi exchange hours. Pros: [--]. One wallet for global margin management [--]. 24/7 Trading and price discovery [--]. No implementation shortfall no missing locates [--]. Transparent automated PnL Tx reports [--]. Vault vertical for easy integration of institutional players (funds/mm/prop) Cons: [--]. Will not solve long tail of thin liquidity past first [---] instruments for every"
X Link 2024-05-22T05:35Z [----] followers, [----] engagements
"@palash_i_am hilarious and a valid point"
X Link 2024-05-22T05:43Z [----] followers, [---] engagements
"I think really with any kind of automated signal research the most important question is how do you overcome the immense amount of noise in your signal pool. If you have some cause - effect you are looking for do you really need to automate the signal discovery Perhaps you could argue that you know the function but not yet the form. And if you don't then you are essentially data-mining which is not atrocious but it will introduce plenty of noise into your signal pool. How do you overcome that then becomes the principal question"
X Link 2024-05-22T09:45Z [----] followers, [----] engagements
"This is one of those areas which are a siren song to those who have just started. It sounds like a dream come through and I think even @quant_arb has written extensively about it. You want to believe that taking [-----] features and throwing it into a process will give you [--] sharpe but so does every graduate with some rudimentary understanding of sklearn. The fact is this: from a market efficiency point of view if it were so easy this would be the dominant way to extract alpha out of the market. The fact that it is conventional wisdom to advise AGAINST it tells you that the approach is riddled"
X Link 2024-05-22T09:48Z [----] followers, [----] engagements
"@cryptohound666 @oxbridgequant @quant_arb @0xfdf Even Piper agrees"
X Link 2024-05-22T09:49Z [----] followers, [---] engagements
"Whenever I'm inevitably asked about alpha mining - I always default to pizzas. I love myself a good new york slice. There are many ways to get a pizza going. Tomato sauce base mozerella pepperoni Fucking amazing. [--] sharpe. White sauce base mozzy mushrooms Give it to me. [--] sharpe Lemon ricotta base spinach lamb sausage I'm all over it. [---] sharpe. You look at all these pizzas and you gotta respect the pizzalo for having some game. One day your friend who has never baked a day in his life tells you "you know what I'm going to get into the pizza game. I'm going to get into the business of"
X Link 2024-05-22T10:17Z [----] followers, [----] engagements
"@quantymacro @ScottPh77711570 @oxbridgequant @cryptohound666 @quant_arb @0xfdf next month it won't be halved. it will be complaining about [--] productivity"
X Link 2024-05-22T10:42Z [----] followers, [---] engagements
"@0xfdf @experquisite @oxbridgequant @cryptohound666 @quant_arb Selection bias under multiple testing 10.3905/jpm.2014.40.5.094"
X Link 2024-05-22T12:09Z [----] followers, [---] engagements
"@0xfdf @experquisite @oxbridgequant @cryptohound666 @quant_arb pink one looks good. count me in for 5M"
X Link 2024-05-22T12:11Z [----] followers, [---] engagements
"@0xfdf @experquisite @oxbridgequant @cryptohound666 @quant_arb count me in for the green one at -5M too"
X Link 2024-05-22T12:11Z [----] followers, [---] engagements
"@0xfdf @experquisite @oxbridgequant @cryptohound666 @quant_arb mentions rentech this guy is retail"
X Link 2024-05-22T12:13Z [----] followers, [---] engagements
"@0xfdf @experquisite @oxbridgequant @cryptohound666 @quant_arb always nice to read your stuff. can tell you're so passionate about the game"
X Link 2024-05-22T12:21Z [----] followers, [---] engagements
"Have you heard of Renaissance bakery though Do you know they have this baking unit that is only available for employees It's been rumored that nobody knows what goes into the pizzas from that baking unit. It's been delivering deliciousness [--] folds above the industry average for [--] years straight"
X Link 2024-05-22T12:37Z [----] followers, [---] engagements
"@quantymacro @0xfdf @YannickKae @0xfdf how does one hedge against the PnL factor"
X Link 2024-05-22T17:08Z [----] followers, [---] engagements
"@quantymacro @0xfdf @YannickKae Omg I know for a portfolio P my hedging basket is -P. PnL neutral"
X Link 2024-05-22T17:08Z [----] followers, [---] engagements
"@Norsay7 @0xfdf @portfoliooptim @experquisite @oxbridgequant @cryptohound666 @quant_arb This is a GOOD point. And is really where a serious conversation on data mining should start"
X Link 2024-05-22T18:10Z [----] followers, [--] engagements
"With the advent of liquidity venues for private markets (NPM and the like) I wonder how much longer it will be before quants are all over the private markets -- from my understanding and limited market research the only competitor/proof-of-concept now is EQUIAM. It seems like we (quants) would deliver a strictly better product from an investment offering perspective. Hear me out: 1) Ability to learn from vast amounts of historical data on past non-intuitive patterns that generalize (pitchbook crunchbase linkedIn SP etc). For every founder/start-up trait that represents a traditional signal"
X Link 2024-05-23T04:03Z [----] followers, 11.8K engagements
"Example of a private equity liquidity venue: https://www.nasdaqprivatemarket.com/product/second-market/ https://www.nasdaqprivatemarket.com/product/second-market/"
X Link 2024-05-23T04:05Z [----] followers, [----] engagements
"The only actually quant vc I'm aware of: https://equiam.com/ https://equiam.com/"
X Link 2024-05-23T04:06Z [----] followers, [----] engagements
"@sersabr Which other funds do you know of That are taking a purely quant approach to the problem"
X Link 2024-05-23T04:32Z [----] followers, [---] engagements
"Post about quant in private markets. I'm quote retweeting this to see if this bypasses the X algo on quashing links. With the advent of liquidity venues for private markets (NPM and the like) I wonder how much longer it will be before quants are all over the private markets -- from my understanding and limited market research the only competitor/proof-of-concept now is EQUIAM. It seems like With the advent of liquidity venues for private markets (NPM and the like) I wonder how much longer it will be before quants are all over the private markets -- from my understanding and limited market"
X Link 2024-05-23T19:00Z [----] followers, [----] engagements
"http://x.com/i/article/1792999650309578752 http://x.com/i/article/1792999650309578752"
X Link 2024-05-24T12:00Z [----] followers, 39.7K engagements
"The source of the quote from @0xfdf https://x.com/0xfdf/status/1793655052692759036 When interviewing quants I often encounter people who are qualified on paper from firms I'm not familiar with but who can't credibly do the work. I've picked up a few consistent signals about those firms that I'll share so people entering the industry know what to avoid 🧵 https://x.com/0xfdf/status/1793655052692759036 When interviewing quants I often encounter people who are qualified on paper from firms I'm not familiar with but who can't credibly do the work. I've picked up a few consistent signals about"
X Link 2024-05-24T12:02Z [----] followers, [----] engagements
"I wake up everyday thankful I am a quant. I go to sleep everyday thankful I am a quant. I genuinely believe it is good for the soul and mind. If we could run for basis points it might even be good for the body"
X Link 2024-05-24T22:36Z [----] followers, 93.5K engagements
"Play the game for what it is. A chance to compete with your mind a chance to prove yourself in a system that is as close to meritocratic as it can be. One has to revere the challenge and be able to show up every day rain or shine and compete. If you can do it day on day for decades you will be rewarded with the challenge of your life brilliant colleagues (really some of the smartest people around) and oftentimes more money than you know what to do with. Quants who can't enjoy the game will one day drop out and find themselves lurking fintwit and complaining about the soullessness of other"
X Link 2024-05-24T22:47Z [----] followers, [----] engagements
"I'd also like to believe that being a quant is more than just a career but a way of life of reasoning about the difficult choices we have to make in life. There are few careers where you have to research a topic stake real money on it get noisy feedback in real-time and manage the risk of being wrong. These skills make for a pretty well-rounded toolkit for dealing with most problems in life"
X Link 2024-05-24T22:54Z [----] followers, [----] engagements
"One of the things I really enjoy being a quant but don't often get to talk about is just how absolutely diverse the ways one can be smart in. I think it's a function of the many ways one could make money in the markets. The longer you stay in the industry the more you appreciate that people are multi-dimensional and can be geniuses in ways you are retarded in and vice versa"
X Link 2024-05-24T22:59Z [----] followers, [----] engagements
"Have a mind to start a weekend side project where I'll look at private equity data and see if there are any signals that are persistent and exploitable"
X Link 2024-05-25T06:47Z [----] followers, [----] engagements
"Skill stones in Diablo immortal were so extremely inefficient when they were first released. They were a combination of some constituent fungible commodity e.g. skill_stone = skill_1 + skill_2 + skill_3. Hence the fair way to price a skill_stone would just be to do price(skill_stone) = price(skill_1) + price(skill_2) + price(skill_3). The catch was that there was no product or market for individual skills only skill_stones that were a random ensemble of skills. Firstly liquidity was so thin you could find a certain skill and buy up all the stones below a certain price (floor) with that skill"
X Link 2024-05-25T07:15Z [----] followers, [----] engagements
"A realistic fixed transaction cost in US D1 is 5bps on average if you are in a large fund and [--] bps if you are in a small fund. As a heuristic you can model another [--] bps of slippage/impact for half most liquid part of your universe and 10bps for the half most illiquid parts of the universe. This is a harsher estimation than reality. Its dirty and is a gross approximation but it should do"
X Link 2024-05-25T20:35Z 10.3K followers, [----] engagements
"Sorry when I say "universe". I mean the Russel3000. If you are constructing your universe out of yahoo finance (noob) and getting sp500 you can model all-in costs as 5bps-10bps and life will be fine"
X Link 2024-05-25T20:45Z [----] followers, [----] engagements
"@mipt_filipp @0xfdf Yeah you can tell from his reply that he likely trades a more liquid universe. If you trade 3-4% ADV on the most [----] illiquid names in Russel3000 you will pay for it in slippage and impact"
X Link 2024-05-26T07:46Z [----] followers, [--] engagements
"@KimchiQuant I absolutely know that - which is why I am offering corrections. The entire premise of this is that SVB free data is expensive and I know most people reading this aren't going to have it"
X Link 2024-05-26T07:47Z [----] followers, [--] engagements
"@therobotjames de beers"
X Link 2024-05-26T08:59Z [----] followers, [--] engagements
"@__paleologo Maybe I am cringe"
X Link 2024-05-26T10:02Z [----] followers, [---] engagements
"@__paleologo "For once I am hoping people confuse us with BlackRock" was pretty funny to me"
X Link 2024-05-26T10:04Z [----] followers, [---] engagements
"Why do you guys refer to this as crowding literature It's just an investigation on flows to me and we know for the most part that there are flows have a temporary and permanent component - the interesting thing about this paper is how long-lasting the temporary component is because it has a positive feedback loop that keeps it going. -- Crowding in my mind is nomenclature served for phenomena of the masses"
X Link 2024-05-26T18:22Z [----] followers, [---] engagements
"The funny thing about this paper is that the hedging the "ponzi" risks part isn't clear but you could probably monetize the flows indicator since its so persistent. You would also expect it to be positive in the long run since not all impact of the flows is temporary and the total PnL should be greater than the reversion. Hah"
X Link 2024-05-26T18:45Z [----] followers, [---] engagements
"@AmeliaBarty I want to believe its our destiny I also have to hedge and say it is an optimistic articulation of our destiny. Have you read [--] body problem"
X Link 2024-05-26T19:16Z [----] followers, [---] engagements
"Being on FinTweet and reading some accounts whose entire existence is to propagate some negative remark or incendiary comment really invokes "Genius of the Crowd" vibes by Charles Bukowski but there is genius in their hatred there is enough genius in their hatred to kill you to kill anybody not wanting solitude not understanding solitude they will attempt to destroy anything that differs from their own not being able to create art they will not understand art they will consider their failure as creators only as a failure of the world not being able to love fully they will believe your love"
X Link 2024-05-27T07:13Z [----] followers, [----] engagements
"Movers Risk of Ruin recommended by @turbokapital The story starts with a banger hook of money lost a shoe flipper at Nike getting counter-measured by their employer. and then the quote "Arbers never sleep the ones who are trying to look for an advantage play are thinking about it 24/7 while the ones who are protecting the system are getting their [--] hours shut-eye" Great stories about adverse selection and counter-measures from the house. Then goes on to talk about exotic illiquid bets in small games (e.g. college basketball games) and how the house has a large edge in them. Apparently they"
X Link 2024-05-27T12:58Z [----] followers, [----] engagements
"One of the places where I think it is very logical for a +EV vendor to sell their signals is sports betting. This is because it is difficult to get liquidity as a large known player in this market. If you truly have an edge you might actually be able to profit more from your edge selling your signals to the masses than trying to overcome the liquidity problem on your own. Reading this @ Bloomberg There are plenty of charlatans who are exceedingly good at selling courses on how to get rich quick and with material that is 1) ambiguous and 2) subject to your interpretation 3) with very low"
X Link 2024-05-27T13:00Z [----] followers, 15.6K engagements
"Something is interesting about multi-manager systems always opting to start with discretionary PMs in some cases even discretionary analysts (Norias). I believe some part of it is because there is a long build-out required for quant teams in general but a part of me also believes it is because the CLB/Alpha Capture model works best when signals are strong. Quants have low information correlation on each bets they take on average. In the IR = IC x SQRT(BREADTH) model a systematic approach will have very low IC spread over a large number of bets. This is because most signals are faint patterns"
X Link 2024-05-28T10:01Z [----] followers, 20K engagements
"One thing that I would like to add is that we know that if ER and VarR is estimated without error MVO is mathematically optimal. We also know that a large problem with MVO is that ER is very hard to estimate. I can imagine that if IC is sufficiently high from an ensemble of signals each with high IR the error on ER becomes smaller and MVO approaches optimality. This happens with quant systems as well of course. I think the question is if the effect is significantly more pronounced with discretionary PMs"
X Link 2024-05-28T10:09Z [----] followers, [----] engagements
"@palash_i_am Which part I am not familiar with EP. I know they exist but nothing beyond that"
X Link 2024-05-28T10:10Z [----] followers, [---] engagements
"@HiddeMulder2 I hear you but costs at the scale Jain has raised should not be the primary concern. I would imagine that not being able to have a strong start without be more of a concern. Unless your equation includes the cost of time as in this allows them to start earning asap"
X Link 2024-05-28T10:15Z [----] followers, [---] engagements
"@IndEqQnt @__paleologo I think when he said AC book he meant ALPHA CAPTURE"
X Link 2024-05-28T10:31Z [----] followers, [---] engagements
"@__paleologo Yes - that's what I mean speed of execution on the business level. On one hand I guess it's Occam's razor all the way down on the other hand kind of disappointed that it's just business strategy"
X Link 2024-05-28T11:00Z [----] followers, [---] engagements
"@__paleologo And I'd have to take your word for it because I've heard great things about Citadel's CLB and I would assume you've seen how the sausages are made"
X Link 2024-05-28T11:01Z [----] followers, [---] engagements
"Listening to @alexisohanian with @tseides on Capital Allocators podcast Really enjoyed the idea of community building on Reddit by playing as an internet host and the implication of online forums being a persistent "party" which led to some organic growth as people stumbled onto old threads. Secondly a good reminder on 1) the virtues of giving immediate direct and compassionate feedback and building a team centered around that; and 2) the importance of letting go of under-performers and retaining a passionate performant team Lastly the idea on social media tending towards proof of humanity as"
X Link 2024-05-28T17:17Z [----] followers, [----] engagements
"Are there any papers I have missed other than the obvious ones I.E The effects of trading suspensions in China Anyone has any interesting anecdotes to contribute to the ether"
X Link 2024-05-29T10:55Z [----] followers, [----] engagements
"Sorry I forgot to add my main point - this leads me to believe that national-chinese PMs have a data-source they are not talking about or some secret cave where they are discussing which stocks are undergoing suspension and for how long"
X Link 2024-05-29T10:58Z [----] followers, [----] engagements
"Signals in china L/S are also unique because your long side often has no alpha. If you decompose your signals into long + short and look only at the residuals on the long side it's [--]. It is not a trivial exercise to get non-zero residuals on the long side. It is extremely competitive. Practically any signal that is worth its salt in the L/S space comes from the short side - precisely because it cannot be realized it cannot be competed away. This has some implications because 1) It implies that searching for signals that are index-hedged is functionally equivalent to searching for LO signals"
X Link 2024-05-29T11:49Z [----] followers, [----] engagements
"It is pretty severe. Your asshole is puckered for an entire month. Your entire book is also thrown off its exposures. Hedging is difficult especially when the suspended stocks can be part of factor mimicking portfolios. Trivial to say that you can backtest this with historical suspended data but in live it is not trivial to predict which stocks are going to be suspended ex-ante"
X Link 2024-05-29T12:32Z [----] followers, [---] engagements
"I work in statarb space so my signals are cross-sectional in nature. A simple example would be directional_dividend_yield - ts_delta(dividend_yield 63) cross_sectional_dividend_yield - directional_dividend_yield - cs_mean(directional_dividend_yield) The cross_sectional_dividend_yield signal would be dollar-neutral and the intended bets are on the relative outperformance of stocks with increasing dividend yield"
X Link 2024-05-29T16:38Z [----] followers, [---] engagements
"@quanterson @__paleologo @palash_i_am There is a well-trodden path with clear requirements and a path less taken where you are free to pit yourself against the jungle If you are pitting yourself against the jungle your knuckles are probably going to be more useful than where you went to school"
X Link 2024-05-30T21:06Z [----] followers, [--] engagements
"yet again the claim out there that risk models drive crowding. Sigh. Could you expound on this a little I wouldn't go as far as to say "risk factors drive crowding" but the idea that there is crowding from using the same risk factors seems fairly intuitive to me. If we are all looking at a certain factor when exposures to that factor rise we are all going to be shorting our exposure's worth of that factor to hedge. If we are using the same product - our factor-mimicking portfolio are going to be identical. There is going to be impact from this. Would this not be considered as crowding"
X Link 2024-05-31T08:30Z [----] followers, [----] engagements
"I think something missing from this discussion is that 1) exposures are not constant 2) target portfolios are not formed instantaneously and we need to trade into them. 1) Exposures wax and wane. Exposures are decided by the same risk models. As long as any two strategies share some common variation to a factor and that exposure increases beyond some threshold hedging is expected 2) If you residualizing your portfolio to F it will result in resid_port = curr_port + hedging_port This hedging_port is going to be in the same direction as every other strategy with the same common variation to F."
X Link 2024-06-01T19:52Z [----] followers, [----] engagements
"@0xfdf @__paleologo @quantymacro @lloydchrist84 @coastal_quant Let me model this when I have time"
X Link 2024-06-01T20:04Z [----] followers, [---] engagements
"@0xfdf @__paleologo @quantymacro @lloydchrist84 @coastal_quant Sorry I wanna be more active but wife is really angry I'm not helping more with the house moving"
X Link 2024-06-01T20:09Z [----] followers, [---] engagements
"@__paleologo @0xfdf @quantymacro @lloydchrist84 @coastal_quant Would you also like to meet my grandma in person"
X Link 2024-06-01T20:30Z [----] followers, [---] engagements
"One reason why I am inclined to say this is because anectodally I've experienced a few instances where I see exposures of a portfolio change - trading activity in fmp - factor PnL starts to move - more trading activity in the fmp. Since then the idea that using the same risk models leads to crowding in the hedging baskets has always been an idea I've taken to be true but I've not investigated this in detail"
X Link 2024-06-01T20:37Z [----] followers, [---] engagements
"But in a real world setting two strategies are definitely correlated. I can't imagine two index-rebal PMs running strategies that are completely orthogonal to each other. Using the same datasets using the same know-how. Its hard to imagine successful strategies of the same flavor being completely uncorrelated"
X Link 2024-06-01T20:41Z [----] followers, [----] engagements
"@__paleologo @SmityAndre82456 @0xfdf @quantymacro @lloydchrist84 @coastal_quant Index rebal is also a good example because it almost always has positive loading on the momentum factor. Yet it's expensive to put on a constant hedge so most PMs opt not to until they get dinged for having indecent exposures"
X Link 2024-06-01T20:45Z [----] followers, [----] engagements
"One final thing I'd also like to contribute to the ether is that two strategies can have a high correlation to each other in their target position space but trade at a different pace to it. The annoying thing about hedging is that all strategies with the same exposures have to put on hedges pretty much all at once. In a very real way it really does "expose" crowding"
X Link 2024-06-01T20:53Z [----] followers, [---] engagements
"To the dms and comments on what book I should read or why don't I have x or y: 1) we travel alot and have a "buy what you are willing to dispose" mantra 2) this means unfortunately books get thrown out every moving cycle sometimes only to be bought again 3) i mostly use e-books. I buy paperback if i REALLY like it or to passed as a reference 4) guys I do read fiction literature. Read Great gatsby (its there). Love Kafka and Flowers more 5) There are more books in the library not shown here"
X Link 2024-06-04T03:58Z [----] followers, 19.6K engagements
"@BlackSwan_ptf Thorpe is a fucking legend. Go for that first. Its a great read. Non technical but really is one of the books on the differentiation of mental giants vs rest of the humans"
X Link 2024-06-04T06:11Z [----] followers, [----] engagements
"@bfpill Phemonenal read. More than a story about robert moses felt like a commentary on how a shadow govt works in practice"
X Link 2024-06-04T06:57Z [----] followers, [---] engagements
"I think there is a rather interesting lesson to be drawn from @TheRoaringKitty's bet on GME. It may look like absolute lunacy but it's actually a fairly sound move even from a portfolio optimisation perspective. We often talk about diversification in the industry as a means of reducing variance - we want to smooth out our risks over a large number of instruments so no one instrument can hurt us too much. And this idea is taken as gospel and pushed to an extreme L/S equities often end up having portfolios that bet on 10000+ instruments globally at the same time over millions of signals. This"
X Link 2024-06-04T10:24Z [----] followers, 36.9K engagements
"This is not a commentary on market manipulation. I for one am up for any strategy that can generate returns. I think it something that is easy to observe its association but won't be easy trying to --prove-- causation. If insider trading were legal I would not be a quant though. I'll be a discretionary L/S PM with ceos of nano-cap companies in sleazy clubs everyday of the week banging out 1000+% returns a year"
X Link 2024-06-04T10:32Z [----] followers, [----] engagements
"@HiddeMulder2 @TheRoaringKitty Odd that you would call [--] stock + cash as diversification but I understand your point"
X Link 2024-06-04T10:36Z [----] followers, [---] engagements
"What's the most intuitive explanation of factors that you guys know Every young quant (in a hf) is exposed to factors basically within [--] weeks into the job whether they are in a central training program or a pod. I find the mechanical explanation of a factor bland unintuitive and doesn't explain clearly the distaste for factors. I've always liked to get people to imagine factors as the first principal component of all the implementations of a known replicable alpha/signal. That captures the idea that 1) Factors are just another TxN weights matrix 2) Factors are basically spent alphas/signals"
X Link 2024-06-05T16:37Z [----] followers, [----] engagements
""Most systems have a few driving factors" is a nice generalization. In the army we used to understand that great units are seldom the sum of great men but the product of great leaders orchestrating the performance. Take out the leaders you cripple performance. @0xfdf @systematicls I don't know that I can answer that. But I also believe that there are two possible answers. [--]. The first is economic-theory based. There is one (SDF) factor whose return in a future state of the world is proportional to the marginal utility of of your consumption in that https://t.co/oxPrIchhht @0xfdf @systematicls"
X Link 2024-06-05T17:43Z [----] followers, [----] engagements
"I think I've discovered something interesting but I need to test this in more detail. It appears that within the first [--] mins of a tweet one can roughly extrapolate the popularity of the tweet. There is a strong momentum effect that gains steam at the start or never does. I wish Analytics captured the time-series evolution of tweet impressions so I can study this in detail without writing some poller to track impressions on my own"
X Link 2024-06-05T19:26Z [----] followers, 10.1K engagements
"I really like NNT's writings and ramblings. I know he has a reputation as a pompous prick but I find it endearing that he lives his life as a free man. I rage vicariously through him. https://medium.com/incerto/how-i-write-8b495eae0330 https://medium.com/incerto/how-i-write-8b495eae0330"
X Link 2024-06-06T06:33Z [----] followers, [----] engagements
"Investigating factors in crypto and using only pairs in @HyperliquidX. Factors are orthogonalized to each other using gramschmidt. Observations: 1) Supermajority of returns is explained by the market factor 2) A tiny amount is explained by time-series momentum 3) The rest are vanishingly small and not statistically significant It could be that HL only lists cryptos of a certain size so the effect of the size factor is more muted in this universe. Still a surprising finding. Does anyone have contradictory results"
X Link 2024-06-06T07:47Z [----] followers, 36.3K engagements
"@HyperliquidX To me at least it's counter-intuitive to what I'd expect from equities"
X Link 2024-06-06T07:52Z [----] followers, [----] engagements
"@ScottPh77711570 If I don't residualize it to market this is what I get which I assume you should see something similar in the arithmetic returns space"
X Link 2024-06-06T08:52Z [----] followers, [----] engagements
"@Andrew30482844 curious where you interpreted the requirement for a consistent drift"
X Link 2024-06-06T16:59Z [----] followers, [--] engagements
"Ah okie that's fair. 1) I do not think all factors have positive premia all the time think they have pockets of predictability just like most signals. For the ones who do have long-term positive premia I do expect that they have some undesirable characteristics that allow them to persist. I should word that better. 2) I do believe however that factors are the results of the impact from persistent trading. E.g. That the covid factor is likely the impact of gross delevering across multiple strategies"
X Link 2024-06-06T17:14Z [----] followers, [---] engagements
"@weaponizedFOMO It's true that the bar to be competent for a somewhat competitive end-to-end system has gotten progressively higher :-( How do you deal with it"
X Link 2024-06-07T10:28Z [----] followers, [---] engagements
"@pedma7 Thank you Trend is a very sticky. It's relatively high vol low sharpe and has high drawdowns. BUT it is one of the very few things you can say for sure has positive premia"
X Link 2024-06-07T11:15Z [----] followers, [---] engagements
"@pedma7 you can try adding carry to your repertoire next It also has positive premia higher sharpe has a lower drawdown but requires you to manage margin"
X Link 2024-06-07T11:17Z [----] followers, [---] engagements
"Why are there so little interest in prediction markets Why have they traditionally failed to garner any significant interest"
X Link 2024-06-07T14:42Z [----] followers, [----] engagements
"Roaring Kitty's biggest flaw is unveiling the mystique he had built up over the [--] years and showing everyone that there was no magic after all. Only a man and his dream to be a billionaire through a meme"
X Link 2024-06-08T07:03Z [----] followers, [----] engagements
"If I were his game theorist I would not have recommended that move. He did not backtest live streaming before using it in production. Typically strategy drift and copium in production"
X Link 2024-06-08T07:15Z [----] followers, [----] engagements
"@vitrovic1 This is interesting we were watching too. And was a learning/point-of-interest to us. I don't know what dataset the "algos" were using we don't have anything that has such high fidelity. It's probably some raw live datafeed. https://x.com/systematicls/status/1799121040192242157 So interesting that the instant he said "lose it all" prices dropped by [---] bps. https://x.com/systematicls/status/1799121040192242157 So interesting that the instant he said "lose it all" prices dropped by [---] bps"
X Link 2024-06-08T07:28Z [----] followers, [---] engagements
"@vvsmoe Sorry do you mean people speculated that he had some plans Or that he announced somewhere that he had some plans"
X Link 2024-06-08T07:57Z [----] followers, [---] engagements
"@quant_spence @TheFlowHorse @0xfdf I'm just another dumbass larp"
X Link 2024-06-08T15:56Z [----] followers, [--] engagements
"You can bet on a global portfolio of [-----] stocks that stocks with increasing dividend yield will have higher relative returns while stocks with decreasing dividend yield will have lower relative returns. You would express such an idea as dividend_yield_delta - ts_zscore(dividend_yield 63) ptfl_mean - cs_mean(dividend_yield_delta market) cs_dividend_yield_signal - dividend_yield_delta - ptfl_mean"
X Link 2024-06-09T22:43Z [----] followers, [---] engagements
"You need to get the factor mimicking portfolio (the weights of the factor portfolio) for the characteristic portfolio approach you need to estimate the fmp. If you are do fama-french N-sorting you get the fmp as a result. Once you have the fmp and the factor returns you can regress to find your exposures and hedging is then -exposure*fmp"
X Link 2024-06-10T06:29Z [----] followers, [---] engagements
"@wiltonhotz Do note that how you construct your factors will affect your exposures your hedging and the return characteristics you see. @CliffordAsness's AQR has a good paper on this: https://www.aqr.com/Insights/Research/Journal-Article/Craftsmanship-Alpha-An-Application-to-Style-Investing https://www.aqr.com/Insights/Research/Journal-Article/Craftsmanship-Alpha-An-Application-to-Style-Investing"
X Link 2024-06-10T06:36Z [----] followers, [---] engagements
"@itsalexjune @__paleologo Talking about is the key phrase here"
X Link 2024-06-10T07:19Z [----] followers, [---] engagements
"@HPierre_ @__paleologo Infeasible area"
X Link 2024-06-10T19:18Z [----] followers, [----] engagements
"In the event that your corr(signal returns_t+1) is held constant across your universe this is true Which is fascinating because your lower bound will be held constant while your upper bound will increase basically giving you more opportunities to juice your returns. Very intuitive"
X Link 2024-06-10T20:14Z [----] followers, [----] engagements
"Fascinating read. A summary in a picture: https://www.bloomberg.com/news/features/2024-06-06/sam-altman-bent-the-world-to-his-will-long-before-chatgpt-s-rise https://www.bloomberg.com/news/features/2024-06-06/sam-altman-bent-the-world-to-his-will-long-before-chatgpt-s-rise https://www.bloomberg.com/news/features/2024-06-06/sam-altman-bent-the-world-to-his-will-long-before-chatgpt-s-rise https://www.bloomberg.com/news/features/2024-06-06/sam-altman-bent-the-world-to-his-will-long-before-chatgpt-s-rise"
X Link 2024-06-11T12:14Z [----] followers, [----] engagements
"@therobotjames years oops I meant this. He's a very niche very indie artist. More famous for playing games (Rust) than he is a musician but he uses his soundtracks in his videos. https://open.spotify.com/track/3ShromO1Sf75HXGFABMNlU https://open.spotify.com/track/3ShromO1Sf75HXGFABMNlU"
X Link 2024-06-12T05:10Z [----] followers, [---] engagements
"Unironically true for strategies in quant finance as well. Strategies get crowded after they do well as they become the new hot thing and then they eventually collapse under their weight and deflate. In the collapse the space clears up and a new pecking order emerges. If your opponents are opportunists one way to beat them is to outlast them. Opportunists almost by definition lack staying power. If your opponents are opportunists one way to beat them is to outlast them. Opportunists almost by definition lack staying power"
X Link 2024-06-15T21:50Z [----] followers, [----] engagements
"@__paleologo book Advanced Portfolio Management is my go-to recommendation for learning more about factors with a good balance between intuition and rigor. There is also material on hedging in the appendix. For a very light (and hopefully intuitive) introduction to factors you can refer to a recent article: https://x.com/systematicls/status/1799941636618867011 https://x.com/systematicls/status/1799941636618867011"
X Link 2024-06-17T13:11Z [----] followers, [----] engagements
"All examples here can be trivially reproduced. Data used here is "retail grade" data from Polygon but it shouldn't matter even if its plucked from yfinance because its OHCLV data for Sp500 constituents. The banks variable is actually dollar volume as in banks=close*volume. Lastly .shift(2) is used because I cannot assume being able to get my entire position at the close. Having an entire day to vwap into my position is more realistic for me. If you are retail trading less than 1m you can probably use .shift() for your signals"
X Link 2024-06-17T21:12Z [----] followers, [----] engagements
"Redbone by LAV8 on spotify is real musical alpha for coding"
X Link 2024-06-19T07:30Z [----] followers, [----] engagements
"@__paleologo The time series approach is useful for high-complexity factors where its not trivial to define a known characteristic for an instrument e.g. index-rebalness"
X Link 2024-06-19T08:37Z [----] followers, [----] engagements
"It is a factor. Everything that has been discussed is a factor. But these factors differ in construction. There are [--] ways to calculate factors: [--]. Time Series Where you have the factor's portfolio and therefore returns and you need to estimate the instrument's loadings on the factor. [--]. Fundamental Where you have each instrument's loading to the factor and you need estimate the factor's returns and from returns back out its portfolio. --- Then there are statistical factors where you are doing some kind of statistical method to back out latent factors from the returns e.g. PCA"
X Link 2024-06-19T15:01Z [----] followers, [---] engagements
"@macrocephalopod @0xfdf @SmityAndre82456 @__paleologo @palash_i_am This means you can only hedge by doing some kind of optimization on your original portfolio right - what gappy refers to as "internal hedging" where your factor exposures are a part of the constraint/objective"
X Link 2024-06-19T15:17Z [----] followers, [----] engagements
"@macrocephalopod @0xfdf @SmityAndre82456 @__paleologo @palash_i_am I guess this works I don't particularly like it though. I've found the extra constraints make the portfolio worse in testing"
X Link 2024-06-19T15:20Z [----] followers, [----] engagements
"The one sticky use-case for blockchains that have actual PMF is a decentralized marketplace of all shapes and sizes: amms orderbooks gambling prediction markets etc. I think it is only natural to imagine that the space will continue doubling down on it and trying to expand vertically and horizontally into the domain of TradFi. Liquidity also tends to consolidate there is a strong winner takes all phenomenon in exchanges/marketplaces. The first one who can crack RWAs at scale will likely have a significant advantage at venturing into more and more esoteric offerings"
X Link 2024-06-20T08:36Z [----] followers, [----] engagements
"Have just begun to read it and already a banger. Foresee this to be the new reference book to all the materials needed in the quant landscape wer alpha tho MAJOR DROP: "The Element of Quantitative Investing" latest book draft. Links at the end of the 🧵. Short introduction: I am writing a book on quantitative investing for Wiley (April 2025). It's the book I wanted to write all along but I strayed and wrote another book first. MAJOR DROP: "The Element of Quantitative Investing" latest book draft. Links at the end of the 🧵. Short introduction: I am writing a book on quantitative investing for"
X Link 2024-06-22T06:07Z [----] followers, [----] engagements
"@ScottPh77711570 You might be interested in such frameworks for crypto flow around special sits"
X Link 2024-06-22T06:44Z [----] followers, [----] engagements
"@ScottPh77711570 If you think about it even DEXes in new chains are subject to these binary events. They either make it have a winner's take all effect and have their valuations 100x or don't have [--] liquidity and eventually die"
X Link 2024-06-22T06:56Z [----] followers, [----] engagements
"Can construct a market-neutral long trump short everyone else to collapse the reality spread. If you're risk-averse long trump biden short everyone else to collapse the illusion of choice spread. Betfair US election probabilities - 58% Trump 17% Biden 19% Newsom Harris Obama or Clinton 5% Anyone else Kind of amazing. Betfair US election probabilities - 58% Trump 17% Biden 19% Newsom Harris Obama or Clinton 5% Anyone else Kind of amazing"
X Link 2024-06-28T08:36Z [----] followers, [----] engagements
"I remember in [--] the institutional equities guys were all laughing at the crypto guys for how manic the crypto markets were. Funny how no one escapes the meme mania"
X Link 2024-06-28T22:31Z [----] followers, [----] engagements
"Can't agree with this take. I'm fairly certain Ken Griffin can get into Citadel. The point is never to find someone perfect in CS math and finance. But to look for a genuine and deep interest in at least one of them as a signal that they have a strong desire for curiousity AND competence. One of the most outdated or mid curve things in quant is the talent selection process. No one can be perfect in CS math and finance. I am not sure if Ken Griffin can get into citadel now. One of the most outdated or mid curve things in quant is the talent selection process. No one can be perfect in CS math"
X Link 2024-06-29T01:19Z [----] followers, 58K engagements
"@blueprintsmb22 @JaredKubin understood apologies"
X Link 2024-06-30T10:59Z [----] followers, [--] engagements
"For those in the game what are popular institutional datasets in crypto Price is not an issue. E.g. Is ravenpack still the go-to news provider for crypto"
X Link 2024-06-30T14:15Z [----] followers, 22K engagements
"To make sure I get my message across. My original post: --- Imagine selling a course on "end-to-end infrastructure" for "complex alpha representations" so you have "all you need to begin quantitative trading". Only to run a funding-rate arb bot. What people sell tells you a lot about what they believe is profitable. (hint: they don't sell anything they believe is profitable) The fact he blocked me right after tells you a lot about the validity of my post. I assume most of my followers are smart rational people. Let's have a simple thought experiment. You discover an investment process that"
X Link 2024-07-02T09:53Z [----] followers, 45.7K engagements
"nothing wrong with teaching funding rate arbs His post was about his own funding rate arb implementation which I found funny because [--]. His entire content is teaching people to build a system for backtesting and alpha discovery [--]. In the same post he talked about moving off manual PnL recording into a .csv file. I'm not sure someone who found it more convenient to manually record PnL should be teaching courses on automation"
X Link 2024-07-02T14:46Z [----] followers, [----] engagements
"@TimMeggs @spxudi @quant_arb @TimMeggs Gave me a great breakdown of each of the popular vendors and their pros and cons. I think he has a strong understanding of what each competitor's strengths and weaknesses are and how they compare to his product"
X Link 2024-07-02T15:06Z [----] followers, [----] engagements
"@m1ha313 @quant_arb HFT Infra for ML How does that work (genuine question)"
X Link 2024-07-03T10:31Z [----] followers, [---] engagements
"@m1ha313 @quant_arb hahaha understood"
X Link 2024-07-03T10:35Z [----] followers, [--] engagements
"Let me tell you guys its shocking how many MFT equity statarb teams are up 3% on GMV YTD. The only question is. are you up 3% on GMV with [--] sharpe (underperformer) [--] sharpe (average) or [--] sharpe (best-in-class)"
X Link 2024-07-04T11:28Z [----] followers, 17.4K engagements
"@BearNecessiti15 I've always wondered what its like to play poker with weak currencies and huge denominations. All in. [--] million (they normally convert to cr) rupees Must feel good to say"
X Link 2024-07-23T08:00Z [----] followers, [----] engagements
"@rosswern @macrocephalopod Very interesting. I've not looked at KR before because its relatively thin compared to HK/IN. This is great thank you"
X Link 2024-07-23T08:52Z [----] followers, [---] engagements
"Perspective on running a global book seems to be interesting to you guys. Heres one more - there are two ways to run a global statarb book. You can run every country as a seperate portfolio and so all your signals are country-wide signals and you can't make bets on cross-country patterns and can't easily hedge global factors. Doing it this way is easy to manage and requires no further thought. However if you are a mega chadster the other way you can run a global book is to trade all [-----] stocks in a single portfolio. You can now make bets on global patterns and hedge out global asset-class"
X Link 2024-07-23T11:44Z [----] followers, 14.2K engagements
"Yes but then you don't have issues of your signals being misaligned cross-sectionally across countries. Lets try another rather than value lets imagine cross-sectional momentum instead. If you ranked your signal across the globe for cs-momentum it would be a substantially different portfolio from ranking your signals country-wide and concatening them together"
X Link 2024-07-23T12:18Z [----] followers, [---] engagements
"I think you're getting stuck on the trading aspect the salient point is the signal aspect. Are your signals global Or are they country-wide. If they are global then you have to deal with implementation shortfalls of your signals across the different countries. Not thinking about this is catastrophic. If they are country-wide then you can combine them into a single portfolio or trade them separately I don't think the difference is catastrophic"
X Link 2024-07-23T12:25Z [----] followers, [---] engagements
"@0xfdf @IronEconomist @macrocephalopod AFAIK there are many many competent statarb teams that have never created global signals in their lifetime"
X Link 2024-07-23T12:33Z [----] followers, [---] engagements
"@macrocephalopod hey maybe I furiously plant trees on the weekends you never know mate"
X Link 2024-07-25T07:29Z [----] followers, [----] engagements
"It seems like analytics and optimizations of portfolios and risks are the truest expressions of quant skills. I suppose that makes enterprise risk managers quintessential quants. There are so much of a portfolio manager's skillset that have nothing to do with expressing quant skills but I would consider to be integral in uncovering alpha (e.g. project management deciding the search path along the alpha search space)"
X Link 2024-07-27T08:16Z [----] followers, 10.4K engagements
"Because short + long index is a vastly inferior product to long/short. For the simple reason that you are only capturing short-side alpha and forgoing long-side alpha AND your long index cannot hedge out your factor exposures on the short side efficiently I dont understand why there are not more short only hedge funds Own index short get paid for alpha. Median stock does not create much value. This chart is exaggerated but upside is narrow Used to do generalist short selling & loved it. Wish that was still a thing I dont understand why there are not more short only hedge funds Own index short"
X Link 2024-07-28T05:06Z [----] followers, 18.3K engagements
"My job is to make the long-run EV calculation using a flawed model given asymmetric information right Yes I know you are passionate and would kill for a chance to be in XXX I know you would want to stay with me till the end of the time of course you would never leave. Of course you'll make the team 100M PnL in your first year. I'm going to assume you are going to come in for 2-3 years learn the know-how contribute fuck-all on average and leave. I'm going to put a hair-cut on everything you tell me. Your modal contribution is going to be [--] post tx cost. And so my modal answer is going to be"
X Link 2024-07-28T08:03Z [----] followers, 13.1K engagements
"You can tell someone's seniority by when they go to the gym. Before and after work Not in management. After lunch Likely senior/management but not top echelon. Comes in at 1pm in running shoes and changes straight into a suit CXO"
X Link 2024-07-29T08:54Z [----] followers, 19.8K engagements
"Quote: Practice. Thats all it is. If you practice enough you can sense where the open ice will be. Watching the three men interact I could not help but notice the banality of genius. It occurred to me that Carbon did not have any superpowers beyond its boring and total efficiency. People worked like machines: they made goals they accomplished them this was the genius of following through. There would be no deliberations no excuses. There was will followed by action. Mean it do it. There was no such thing as a slump. But I wondered if the genius here could also be the horror. You might through"
X Link 2024-08-01T17:33Z [----] followers, [--] engagements
"Gappy woke up today and chose violence. But +1 quants generally stay away from economists and more generally economics/business/finance graduates. If you are a student and want to be a quant math/physics/cs is the way to go @stevehouf Whether economists make great traders or make money in markets actually doesn't say anything about the correctness of the economic models. Youd expect that some economists would at least make good traders/PMs or risk managers. Where are they Geneakoplos Thaler Fama @stevehouf Whether economists make great traders or make money in markets actually doesn't say"
X Link 2024-08-05T07:57Z [----] followers, 21.7K engagements
"I think the highest density are engineers/physics/math/cs graduates"
X Link 2024-08-05T07:59Z [----] followers, [----] engagements
"@kelxyz_ I think having a MSc in a technical discipline is very advantageous to be a quant. It does have some signal. I can only speak for this industry and not for others"
X Link 2024-08-05T08:23Z [----] followers, [---] engagements
""Hedge funds have underperformed the market. Hedge funds have underperformed the market." John sobs quietly to himself as his global long-only ETF portfolio gets decimated. Meanwhile - how's just another Monday for all my L/S broskis - better check that size and momentum exposure and we should be good to go"
X Link 2024-08-05T15:02Z [----] followers, [----] engagements
"How does FinTwit feel about platforms like https://numer.ai/home https://numer.ai/home"
X Link 2024-08-11T06:04Z 10.3K followers, [----] engagements
"So the main reason I'm asking this is because I'm curious why none of these have achieved significant scale - some have come and gone like Quantopian. Some have come and never managed a significant amount of capital and most have fizzled out of the news"
X Link 2024-08-11T08:19Z 10.3K followers, [----] engagements
"If I wasn't a quant and if all whimsical jobs in an advanced society were equally financially rewarding so there was nothing to min-max one of the more interesting jobs is to go around the world and look for very interesting hero journeys (a la campbell)) or people with exceptional character and document all the features that make them the way they are. Eventually all features will either be endogenous or exogenous. Create a test to find endogenous traits and create a school to assert exogenous traits. Raise money to go to subsaharan africa and find all impoverished children who match the"
X Link 2024-08-12T23:11Z 10.3K followers, 15K engagements
"@PinguCapital @pleaseconsider4 @__paleologo Here's some plain speak: you are antiquated. The best discretionary PMs hire quants in their teams today precisely because they have learnt that a quant overlay to guide portfolio management decisions significantly improves performance"
X Link 2024-08-23T03:01Z 10.3K followers, [---] engagements
"Small timer wonders why he can't break into the big leagues. #Quants are evolving into religious proselytizers with their equations like a new gospel in finance. Theres nothing new about using maths when trading or investing. History shows theres been many extremely successful investors/traders before quants became mainstream. #Quants are evolving into religious proselytizers with their equations like a new gospel in finance. Theres nothing new about using maths when trading or investing. History shows theres been many extremely successful investors/traders before quants became mainstream"
X Link 2024-08-23T03:02Z 10.3K followers, [----] engagements
"@PinguCapital Yeah I know some guys who call their $10K PA a "family office" too"
X Link 2024-08-23T03:50Z 10.3K followers, [---] engagements
"If we are to carve Olympus into our image it will not be from one mighty blow but a million chips to the surface - each insignificant but in totality world-changing"
X Link 2024-08-27T21:36Z 10.3K followers, [----] engagements
"Can't speak for all careers obviously but at least for a traditional quant career the first few years are slow but normally you'll get to a point where PnL becomes positive and attributable and it's a much more comfortable ride from there on out. Your purchasing power should outpace everything short of financial armageddon"
X Link 2024-09-02T00:08Z 10.3K followers, 21.6K engagements
"How does one start building systematic strategies for these markets Where is a good place to get normalized data Being able to trade all the states of the world a little step at a time. https://t.co/OelD7Jg21j Being able to trade all the states of the world a little step at a time. https://t.co/OelD7Jg21j"
X Link 2024-09-10T05:46Z 10.3K followers, 11.9K engagements
"@saplash_i_am I think that depends on the driver right. If two bets are implicitly betting on let's say the health of joe biden then theoretically it should hold even at the smallest discrete timesteps - any deviation is a statistical arbitrage opportunity"
X Link 2024-09-10T06:38Z 10.3K followers, [---] engagements
"Taking a poll on behalf of a friend - A start-up fintech company has a novel idea at the end game it wants to crowdsource portfolio managers and allow thousands of people to participate in the upside of the financial markets while risking no personal capital. Here are the mechanisms: [--]. You first have to participate on their crowdsourcing platform to submit signals. These signals are very simple mathematical models that take about 1-60 minutes to create one. You are paid for each signal you submit. If the firm uses your signals in live trading you are paid a profit share on the PnL as well."
X Link 2024-09-14T06:00Z 10.3K followers, 13K engagements
"Every hopeful quant research candidate in every interview: "You don't bring any special skillsets to our team. Seems like a generic profile and we only want the best." Candidate: "Not me there's only one of me in all the world. I'm one in a krillion https://www.youtube.com/watchv=01Runa3neb0 https://www.youtube.com/watchv=01Runa3neb0"
X Link 2024-09-17T13:51Z 10.3K followers, [----] engagements
"@0xFaust12 @choffstein @0xfdf You sound like you work in Citadel coverage"
X Link 2024-09-19T05:03Z 10.3K followers, [---] engagements
"I'm curious to know what is the intersection of the demand for quant knowledge. Imagine I've stumbled onto an infinite article factory. Comment below on every quant finance article you wish you had access to and I might be able to make it happen"
X Link 2024-09-19T08:10Z 10.3K followers, 22.3K engagements
"@__paleologo @ScottPh77711570 @pedma7 @macrocephalopod @buffetsalpha @therobotjames Based on all my time on X that's basically Scott's game"
X Link 2024-09-20T10:03Z 10.3K followers, [---] engagements
"@ScottPh77711570 @__paleologo @pedma7 @macrocephalopod @buffetsalpha @therobotjames never better honestly thanks for asking. heard you were dishing out life saving advice to a certain oxbridge quant"
X Link 2024-09-20T10:07Z 10.3K followers, [---] engagements
"@pedma7 @__paleologo @ScottPh77711570 @macrocephalopod @buffetsalpha @therobotjames Is there still premium to harvest in ICOs I remember there was a study where if you went long anything on Binance Launchpad and exited [--] hours later it was like a 300% returns strategy. Was not true for other launchpads"
X Link 2024-09-20T10:11Z 10.3K followers, [---] engagements
"@pedma7 @ScottPh77711570 @__paleologo @macrocephalopod @buffetsalpha @therobotjames @__paleologo Imo this [--] tweets tell you everything you need to know for the retail quant crowd. At least the ones who aren't steeped in the institutional L/S stuff - my anecdotal experience is that most of them have become masochists and have learnt to embrace pain"
X Link 2024-09-20T10:15Z 10.3K followers, [---] engagements
"@ScottPh77711570 @pedma7 @__paleologo @macrocephalopod @buffetsalpha @therobotjames I don't even want to click on http://Dumpy.Fun http://Dumpy.Fun"
X Link 2024-09-20T10:18Z 10.3K followers, [---] engagements
"@friendscallmeap @ScottPh77711570 @pedma7 @__paleologo @macrocephalopod @buffetsalpha @therobotjames Have you guys checked out RWA perps vs CeFI exchanges @kaledora's @OstiumLabs is trying to put RWA perps on-chain. Might have some alpha there not sure what's the institutional participation like"
X Link 2024-09-20T10:21Z 10.3K followers, [---] engagements
"PM: "Our team moves fast we can put risk on your ideas very early if you demonstrate potential" QR: Waiting for risks to be put on his ideas after demonstrating potential"
X Link 2024-09-23T03:02Z 10.3K followers, [----] engagements
"It's a good read - at least worth thinking about for retail portfolios. For quants from my understanding I can summarize "return stacking": [--]. Portfolio optimisation exercise where you have at least one alpha generating strategy/manager lets call it A1. [--]. Hedge all factors from A1 you don't want. [--]. Add exposures to A1 that you do want. Use levered exposures for capital efficiency. They've combined [--] and [--] and called it "beta swapping". [--]. Combining [--] and [--] is "returns stacking". It doesn't seem like stacking anything here because we start from A1 but the article implicitly assumes starting"
X Link 2024-09-26T03:30Z 10.8K followers, [----] engagements
""Never give in. Never give in. Never never never neverin nothing great or small large or pettynever give in except to convictions of honour and good sense. Never yield to force. Never yield to the apparently overwhelming might of the enemy.""
X Link 2024-10-09T12:22Z 10.3K followers, [----] engagements
"On the mindless pursuit of power. https://youtu.be/w_MSFkZHNi4 https://youtu.be/w_MSFkZHNi4"
X Link 2024-10-23T15:04Z 13.9K followers, [----] engagements
"Learnt a ton from this podcast about running large models on GPUs. https://signalsandthreads.com/the-uncertain-art-of-accelerating-ml-models/ https://signalsandthreads.com/the-uncertain-art-of-accelerating-ml-models/"
X Link 2024-10-28T02:34Z 13.9K followers, [----] engagements
"In [----] if DL in finance still doesn't work as far as you know then you don't really know a lot @systematicls Well what you're saying makes sense in theory but it doesn't work in practice as far as I know. Some of the best firms have tried to make it work but the problem is that deep learning requires a lot of data to avoid overfitting. This problem exists even with high-frequency @systematicls Well what you're saying makes sense in theory but it doesn't work in practice as far as I know. Some of the best firms have tried to make it work but the problem is that deep learning requires a lot of"
X Link 2024-12-10T04:39Z 10.5K followers, 16.5K engagements
"Idiots on fintwit: no real firms uses DL in practice. Also firms:"
X Link 2024-12-11T14:21Z 10.7K followers, 23.4K engagements
"@arnaud710 @macrocephalopod @quant_arb If what you are saying is literal and true it makes what he's done even more impressive. And I'm not being kind on larps - I mean guys lets be objective here"
X Link 2024-12-12T15:56Z 10.7K followers, [----] engagements
"Just for the uninitiated. When convering inr to usd. Lakh 1k usd Crore 100k usd You're welcome. i can tolerate lakhs but i draw the line at crore. wtf is a crore just tell me in normal numbers i can tolerate lakhs but i draw the line at crore. wtf is a crore just tell me in normal numbers"
X Link 2024-12-20T12:11Z 10.7K followers, [----] engagements
"Playing POE2 and really liking the currency market especially as a market practitioner. In general every game that has a free market allows a patient speculator to profit either from trends or market making; but there is some nuance in every one of them. In POE2 the primary mode of exchange is a barter trade around valuable commodities like the one shown below. There is a currency market that facilitates this barter trade and it allows you to buy "5 exalts for [--] chaos". The most reliable method I've found so far to profit from the markets is to quote a spread around liquid markets (example"
X Link 2025-01-05T15:06Z 12.9K followers, 21.4K engagements
"Land of the guiseppes and it was beautiful and delicious It was mind blowing to see thousands of people flock to appreciate work done thousands of years ago"
X Link 2025-03-02T11:55Z 10.9K followers, [----] engagements
"An unironically clear guide on various arbitrages (mechanical arbitrages triangular arbitrages constituent arbitrages) through a game medium: https://maxroll.gg/poe2/resources/flipping-with-the-currency-exchange https://maxroll.gg/poe2/resources/flipping-with-the-currency-exchange"
X Link 2025-03-23T06:18Z 11.3K followers, [----] engagements
"Guessing from a portfolio standpoint: If you knew with certainty (P=1) how the stock price of a single instrument will move tomorrow the rational optimal thing to do is to borrow from every outlet you can and bet the farm house and tomorrow's farm and house on it. The more certain you are the less you want to dilute that alpha. Even in the context of having [--] stocks with P=1 you would just pick the one with higher abs returns; in cases of complete information there is no need for diversification. However when P is small diversification becomes the only free lunch. Noise gets averaged to 0"
X Link 2025-05-19T06:49Z 12.2K followers, 19.9K engagements
"Poetry Patriotism does not come naturally to meI have always been far too fond of the foreign. Aesthetically I yearn for the rich history and beautiful architecture of Europe; politically I dream of the competent governance and high state capacity of Singapore or Taiwan. I am a Patriotism does not come naturally to meI have always been far too fond of the foreign. Aesthetically I yearn for the rich history and beautiful architecture of Europe; politically I dream of the competent governance and high state capacity of Singapore or Taiwan. I am a"
X Link 2025-07-04T12:55Z 12.7K followers, [----] engagements
"Somehow I think the young white collars are underestimating how devasting AI is going to be in their job prospects. At least in the "short run" until we all become redundant. I foresee that the first tranche of jobs that are going to be impacted are the junior white collars that are traditionally the byproduct of being a median performer of a fairly technical degree (e.g. your 50th to 90th percentile CS grad). These median performers who would otherwise end up with jobs at SMEs with simple requirements for "run-of-the-mill computer jobs" will be the first to have their place in the economy"
X Link 2025-07-06T16:23Z 12.8K followers, 12.4K engagements
"PSA: Do not put skilled in excel or list excel as your first skill when applying for quant researcher roles. Its a negative signal that suggests you are probably going to be a bad swe"
X Link 2025-07-08T09:00Z 12.8K followers, [----] engagements
"@coastal_quant Post more on this please"
X Link 2025-08-17T22:10Z 12.9K followers, [---] engagements
"If you want to have a finger on how the median US statistical arbitrage pod is doing just take a look at some heuristic of w_mom * momentum factor + w_size * -size factor. Momentum is down AND size is up It's a quant quake"
X Link 2025-08-24T13:06Z 12.9K followers, 25.7K engagements
"Quality LinkedIn interactions as usual"
X Link 2025-09-04T06:49Z 13.7K followers, 11.9K engagements
"The world does not hand crowns to the idle nor mercy to the weak"
X Link 2025-10-03T09:47Z 14K followers, [----] engagements
"Wisdom is wasted on the old they best they can do is to try to give it away to oft unwilling receivers"
X Link 2025-10-05T03:10Z 14K followers, [----] engagements
Limited data mode. Full metrics available with subscription: lunarcrush.com/pricing