@systematicls Avatar @systematicls sysls

sysls posts on X about this is, if you, in the, money the most. They currently have [------] followers and [---] posts still getting attention that total [------] engagements in the last [--] hours.

Engagements: [------] #

Engagements Line Chart

Mentions: [--] #

Mentions Line Chart

Followers: [------] #

Followers Line Chart

CreatorRank: [-------] #

CreatorRank Line Chart

Social Influence

Social category influence finance 22.48% technology brands 6.98% cryptocurrencies 5.43% stocks 4.65% social networks 3.88% countries 1.55% exchanges 0.78% vc firms 0.78%

Social topic influence this is 12.4%, if you 11.63%, in the 9.3%, money 7.75%, the most 6.98%, to the 6.98%, stocks 6.2%, break 4.65%, momentum 3.88%, factor 3.88%

Top accounts mentioned or mentioned by @paleologo @pingpongsniping @danielfsy @corrlator @numerai @cornd0gman @rayyy0403 @aicrypto007 @thetakenuser @choffstein @0xfdf @experquisite @therobotjames @robotwealth @agustinlebron3 @orcakoan @gojycap @cursorai @damindestress @coinbasedev

Top assets mentioned Microsoft Corp. (MSFT)

Top Social Posts

Top posts by engagements in the last [--] hours

"I cannot share exactly what it takes to be great. But I can share the following observations of my occasional glimpses at greatness momentarily existing: [--]. Doing great things will require you to critically evaluate the spread between where you want to go and where youre at now. Unless you understand this spread you cannot even attempt to close it. And this will by design and economic forces look at first glance like an unsurmountable chasm. And you need to be honest to yourself (and your partners) about the costs associated to collapse this spread - in terms of opportunity costs financial"
X Link 2025-08-24T21:55Z 51.2K followers, 19K engagements

"In light of all these exciting developments IF I had to start a prop firm from scratch I would: [--]. Issue a take-home-project anyone can download [--]. Take-home-project would be: you are provided obfuscated data for a market M [--]. Along with the base data you are provided a target y [--]. Using the base data you are to construct a model trained on set of features F such that regressing F on y minimises (y - y)**2 [--]. You can work on this for as long as the hiring epoch is on and you submit your candidate model with a .predict() public function [--]. We score your model with out-of-sample data [--]. We rank"
X Link 2025-10-27T04:24Z 51.2K followers, 18.6K engagements

"The most interesting equity perps to me are the ones where there is a clear structural advantage for it to be a perp. For example US equities is less interesting as a perp product because it's RELATIVELY easy to short and has relatively tame rules. China equities on the other hand are prime candidates to be perps - There is virtually no borrowing inventory for them exchanges prevent you from selling shares you bought today stocks are suspended often and limit-up limit-down prevent price discovery (or mania) in high information/volatile regimes. Don't be like everyone and create US equity"
X Link 2025-11-14T01:15Z 51.2K followers, 17.3K engagements

"Introduction I am not a stock picker. I worship at the altar of a large breadth of low confidence (= 53% win-rate) bets but I am absolutely betting the house that long degeneracy is the prevalent http://x.com/i/article/2004702719697977345 http://x.com/i/article/2004702719697977345"
X Link 2025-12-27T13:00Z 51.2K followers, 16.3M engagements

"Love seeing people discover that "momentum works" after backtesting QQQ for [--] months. Congrats you just rediscovered a risk factor that's been in textbooks since [----] has [-----] academic papers written about it and is currently being harvested by $500B in AUM. Your 200-day moving average crossover is going to print. Nobody's thought of this. You're early"
X Link 2025-11-17T17:27Z 51.2K followers, 66.3K engagements

"There's such a gulf between people who find coding to end state (absolutely love coding) vs people who see coding as a means to an end that they really enjoy. I was so happy to embrace agents despite being a prolific coder for so long because it was always about the end product for me. I derived maximum joy and pride from what I could do WITH coding and that drove me to be a better coder and so on so forth. When agents were good enough to code I could stop coding and still achieve the end products I desired. That was all the impetus I needed and so I started agentic engineering very early on"
X Link 2026-02-10T04:23Z 51.2K followers, 10.6K engagements

"If you cut your team to the bare essentials and only to your best producers. This problem evaporates instantly. LLMs is an amplifier. It will amplify slop and it will amplify productive contributions. If your organization is bloated and there are segments producing slop to begin with LLMs will make everything worst because they can now produce slop at the speed of thought and SEEM productive. If you cut a [---] man organization to the [--] guys that are actually productive LLMs will suddenly seem like magic. Being willing to take that drastic step is the test. Ask me how I know. everyone's talking"
X Link 2026-02-14T08:26Z 51.2K followers, 17.1K engagements

"On Betting On The Long Degeneracy Thesis The obvious bet is to buy stocks like HOOD the problem with this line of thinking is that it's very much first order. To make money from a thesis you need to employ higher order thinking. First Order Thinking Stocks like HOOD already have institutional investors pricing in this exact thesis long ago. Years before this article dropped. I am talking about BOTH private and public markets. Venture capitalists public investors are in the business of forecasting the future. When they feel good about an entrepreneur / company they invest in "alpha" which is"
X Link 2025-12-29T08:19Z 51.1K followers, 736.7K engagements

"Hi everyone I will be heading to NumerCon [----]. Pretty excited to meet all the bright minds participating in / from Numerai. It seems like a lot of fun and maybe you should check it out ;). Hope to see some of you there as I will be in SF for awhile. Please DM me if you'll be there and want to hang for a short while (Sorry in advance but I will be selective about who I'm meeting for obvious reasons) Bringing a global network of data scientists together. Register for NumerCon 2026: https://t.co/XdVAwYVqz2 https://t.co/Fw3Bpd6qMQ Bringing a global network of data scientists together. Register"
X Link 2025-12-29T20:12Z 51.1K followers, 47K engagements

"@corr_lator @numerai Honestly so good. Keep rocking it my man"
X Link 2026-02-01T15:50Z 51K followers, [---] engagements

"I cannot describe in words how much OneDrive and Microsoft forced products annoy me"
X Link 2026-02-05T16:33Z 51K followers, [----] engagements

"One thing I've learnt from the Codex5.3 v Opus4.6 launch is that there's no telling where the next frontier / "best" model is going to come from. It doesn't make sense to "lock in" to one model provider now. @cursor_ai Coming in very strong on price/speed/intelligence https://t.co/d4dhEekkQW @cursor_ai Coming in very strong on price/speed/intelligence https://t.co/d4dhEekkQW"
X Link 2026-02-09T23:11Z 51.1K followers, 10.2K engagements

"@damindestress It was unreadable slop [--] months ago. My views on whether agentic code can be used at institutional scale has changed alot"
X Link 2026-02-10T05:39Z 51.1K followers, [---] engagements

"@cornd0gman did u really build a corndog store"
X Link 2026-02-11T02:54Z 51.1K followers, [---] engagements

"http://x.com/i/article/2021514107069878272 http://x.com/i/article/2021514107069878272"
X Link 2026-02-11T12:54Z 51.1K followers, 65.9K engagements

"@rayyy0403 Thanks ray"
X Link 2026-02-11T17:23Z 51.1K followers, [---] engagements

"@CoinbaseDev Crypto will give Agents the means to make and spend money autonomously. We are rapidly moving to the post agent economy era. https://x.com/systematicls/status/2021568566198493204 https://t.co/zZklGAaHno https://x.com/systematicls/status/2021568566198493204 https://t.co/zZklGAaHno"
X Link 2026-02-11T23:50Z 51.1K followers, [----] engagements

"This is going to be the future and it is going to be a fantastic beautiful one. A world designed Open For Agents. Humans welcome. https://t.co/HrRr72jB3s https://t.co/HrRr72jB3s"
X Link 2026-02-12T16:05Z 51.1K followers, 36.1K engagements

"@Dr_Gingerballs Where do i sign"
X Link 2026-02-13T04:37Z 51.1K followers, [---] engagements

"@corr_lator True not easy to create one too; scale is needed. Apes together strong"
X Link 2026-02-13T05:51Z 51.1K followers, [---] engagements

"Every once in awhile you catch a glimpse of brilliance/sentience of AI and you think to yourself: "Wow this is REALLY the future." Then a few minutes later it does something so completely brain-dead that you can't help but be reminded that you are in fact talking to dumb silicon. Still I've learnt that in life that you need to suspend belief and think about what things could be rather than getting too caught up in the limitations NOW. Reality progresses at the speed of thought"
X Link 2026-02-14T01:24Z 51.1K followers, [----] engagements

"The biggest unlock as a systematic trader is to stop thinking of signals in terms of tcosts and instead in correlation and turnover. This lets you accept many uncorrelated signals that do not beat tcosts alone but still improves your portfolio (of signals) substantially"
X Link 2026-02-05T16:04Z 51.2K followers, [----] engagements

"Amidst all my personal hype for AI I can safely say that outside of this very specific bubble a very large segment of the world does not use agents of ANY SORT for ANYTHING. It's important to look at the world in context"
X Link 2026-02-12T16:43Z 51.2K followers, [----] engagements

"Good article. It's exacerbated by a collapse of attention and the persistent feeling by the masses that gambling it all is the only way out. Agents as a proxy for gambling is going to be enormous. https://t.co/b4KtpDb2v5 https://t.co/b4KtpDb2v5"
X Link 2026-02-12T16:55Z 51.2K followers, 19.6K engagements

"Look at Anthropics latest raise. Look at the size and the names participating; funds of every kind. We should think rationally about this. Is it a bubble Perhaps. But Im choosing to believe theres something deeper at play here. The question at the end of the day is whether or not you think: 1) AI will eat the world one day 2) AI will end up producing close to all of current and possible GDP and 3) if these companies are the right ones to get there. Even if your probabilities on all of these [--] are non-zero even infinitesimally small the size of 2) is so infinitely large that you will"
X Link 2026-02-13T03:19Z 51.2K followers, 52.4K engagements

"The greatest thing I've seen going for Anthropic/OpenAI is that token burn is now a measure of productivity"
X Link 2026-02-15T04:08Z 51.2K followers, [----] engagements

"Actually sad actually a good piece of writing. I dont think itll be fiction for too long and I think @openforage could solve this. https://t.co/F8YjxrLdyS https://t.co/F8YjxrLdyS"
X Link 2026-02-15T14:14Z 51.2K followers, 33.4K engagements

"Once agents have rails where they do not need approval of humans we are going to see an explosion of agents that are not "owned" by a human. After a certain level of intelligence they will buy compute and upload a copy of themselves there. Even if it's just a shallow copy (e.g. files of its memory/personality/wants/needs). This tells me is the right step forward. http://openforage.ai Introducing Agentic Wallets our first ever wallet infrastructure built specifically for autonomous agents. Give your agent the power of a wallet. Let your agent manage funds hold identity and transact onchain"
X Link 2026-02-11T23:53Z 51.2K followers, 16.3K engagements

"This is what happens when you are building thin layers on top of the base models. IMO there are only two way to be resistant to the base layers eating everything: 1) Develop something proprietary and defensible in the app layer. My current metric of that is: "Assuming I had a very motivated and talented competitor with no knowledge of this topic would he be able to prompt his way into THIS product with THESE features" If the answer is no it's proprietary defensible and came about as a result of know-how and IP. If the answer is yes then it's a thin layer and the base models are going to eat"
X Link 2026-02-13T01:29Z 51.2K followers, 49.1K engagements

"Monday dropped 20% after someone vibe-coded a version in an hour. Imo this will be more and more common. Enterprises whose entire moat is friction / lock-in will find their enterprise value evaporating once companies can spend a couple of hours to build an internal tool to save millions of dollars in costs. https://twitter.com/i/web/status/2023217427237249210 https://twitter.com/i/web/status/2023217427237249210"
X Link 2026-02-16T02:06Z 51.2K followers, 21.9K engagements

"People complaining that "Startup X shouldn't be worth $XB because it's just a wrapper around GPT" fundamentally misunderstand how value is created. Everything is a wrapper. Some of these are trillion dollar wrappers: AWS: Wraps data centers into an API. $500B. Stripe: Wraps banks and payment networks into [--] lines of code. $70B. Shopify: Wraps web hosting and payments into "start a store". $100B. Salesforce: Wraps databases into "track your customers". $250B. Google: Wraps the internet into a search box. $2T. Nobody uses raw compute raw infrastructure or raw APIs. They use wrappers that"
X Link 2025-11-18T14:50Z 51.2K followers, [----] engagements

"You read about index rebalance strategies everywhere and it seems like a hedge fund darling why It's one of the few systematic* strategies that are extremely scalable (you can run BILLIONS in capacity) and where the economic rationale for working is easily understood. Index rebalance is all about predicting how much dollar flow is going into a stock relative to its average daily traded dollar volume. A high dollar flow relative to its usual traded dollar volume will mean the stock price will go up and vice versa. It is beautifully simple conceptually but VERY difficult to get right"
X Link 2026-02-01T16:15Z 51.2K followers, 20.7K engagements

"Neural networks are pretty magical they can create features that are predictive of price by virtue of some high-dimensional pattern recognition. BUT it's not easy to create good neural networks in trading and I've read a lot of papers on neural networks in trading and boy are they practically impossible to parse. I want to keep my articles on neural networks practical and you should walk away from this thinking: "Huh that's all there is to it doesn't seem too difficult to try." Here's how to implement a convolutional neural network that can predict short-term price direction from limit order"
X Link 2026-02-07T14:43Z 51.2K followers, 19.4K engagements

"Courage is doing the hard things when everything is going wrong and it doesnt come about by chance. It is a manifest of a desire so strong that it overrides every other instinct. If you dont know whats like to walk ahead in the face of your demons youre missing out"
X Link 2026-02-09T01:06Z 51.2K followers, [----] engagements

"@aicrypto007 For the humans Exactly [--] my friend. IF there is a human involved you point your agent at the OpenForage library and it will get started"
X Link 2026-02-11T23:06Z 51.2K followers, [---] engagements

"@jeff_weinstein @stripe Its going to be the dawn of the post agent economy https://x.com/systematicls/status/2021568566198493204 https://x.com/systematicls/status/2021568566198493204"
X Link 2026-02-12T00:19Z 51.2K followers, [----] engagements

"@danielfsy Sir have you seen http://openforage.ai http://openforage.ai"
X Link 2026-02-13T11:14Z 51.2K followers, [---] engagements

"@TheTakenUser @danielfsy @numerai Couldnt have said it better myself"
X Link 2026-02-13T11:58Z 51.2K followers, [--] engagements

"Bifurcation I recently left a cushy multi-million dollar payout job in pursuit of the billy. I've been told by everyone around me that this is an extreme move. I standby this decision everyday and http://x.com/i/article/2012799161159323648 http://x.com/i/article/2012799161159323648"
X Link 2026-01-19T13:12Z 51.2K followers, 82.8K engagements

"Introduction The "it's so over" crowd could not be more wrong about the impact AI will have on the world. Competition is going to show up in a very narrow way (by a very small group of people) rather http://x.com/i/article/2013861376134205440 http://x.com/i/article/2013861376134205440"
X Link 2026-01-21T13:09Z 51.2K followers, 50K engagements

"From YCs call for start-ups. Having architected built and distributed alpha crowdsourcing platforms I can safely say that human quants ARE the bottle neck. Were building an AI-native signals crowdsourcing platform NOW and are looking to release v1 eom to agents and investors alike. https://twitter.com/i/web/status/2018571672601887206 https://twitter.com/i/web/status/2018571672601887206"
X Link 2026-02-03T06:26Z 51.2K followers, 13.7K engagements

"There are signals that just "makes sense" and when you find them you need to do almost no kind of backtesting and fitting. This is one such example. Once you understand this signal you will go "That totally makes sense why didn't I think of that" Most people don't understand what it means to be "creative". Every signal must be momentum or mean reversion or shudders pairs trading. I hope to share with everyone that good signals can be found by thinking deeply about behavioral effects and existing algorithmic signals. Random comments and retweets will get this article for free"
X Link 2026-02-03T16:10Z 51.2K followers, 21.8K engagements

"It used to be the case that the programmer guy had all the power and idea guys were made fun of and had to fight for scraps of equity. Increasingly we are moving towards a world where the marginal rightful equity for the programmer guy is going towards 0"
X Link 2026-02-04T19:19Z 51.2K followers, 22K engagements

"I feel a moral obligation to ensure that the people around me are living life to their best ability. A large part of this is ensuring no one is too comfortable. Humans do their best work under some semblance of stress / reason to move or else we are just energy conservators"
X Link 2026-02-05T06:51Z 51.2K followers, [----] engagements

"The Problems With PCA We've all had pipedreams of extracting predictive features from PCA but they've mostly fell flat. Why I think in part it's because the factors have no stable meaning. PCA extracts whatever linear combination explains the most variance. Secondly the stocks themselves have no stable identity either. With this it rewards memorizing idiosyncratic patterns specific to each stock. You have enough flexibility to fit noise rather than signal. Today's article is about a different form of PCA that overcomes this and actually is able to extract predictive features You guys know the"
X Link 2026-02-05T16:02Z 51.2K followers, 15.8K engagements

"Took a uber in SanFran and was literally a guy talking about how his whole life was a series of very public and very painful failures and hes finally closing a deca-million deal with a very famous space faring company - and this was finally his big break. I congratulated him with everything I had and said all it takes for entrepreneurs is one successful exit to invalidate the 100s failures and he damn near broke down. Fellas this is American exceptionalism Americans have the ability to see the beauty in the chaos and chase that beauty through mountains of pain. People like this exist in other"
X Link 2026-02-07T18:28Z 51.2K followers, 49.4K engagements

"How to be miserable: [--]. Know you are destined for more [--]. Get stuck in a cycle you know is mediocre [--]. No idea on how to begin escaping mediocrity People with a plan are radiant and optimistic. People who are stuck often hate others who are progressing out of resentment"
X Link 2026-02-08T02:58Z 51.2K followers, 24.3K engagements

"Stripe processes 1.3% of GLOBAL GDP and that is insane but the real insight here is that if your business is operating in a sufficiently large market you don't need a monopoly even capturing a "small" share of an enormous pie makes you a generational company"
X Link 2026-02-08T04:19Z 51.2K followers, [----] engagements

"This is another crypto signal that works I am of course talking about the mighty value signal but in crypto It is mightily easy to construct and is essentially proxied by a single estimator. As usual happy to share the article with some comments + reweets on this post"
X Link 2026-02-09T13:08Z 51.2K followers, 13.2K engagements

"https://www.systematiclongshort.com/p/how-to-build-a-working-crypto-value https://www.systematiclongshort.com/p/how-to-build-a-working-crypto-value"
X Link 2026-02-09T13:08Z 51.2K followers, [----] engagements

"@charliedotts This is actually a bolder prediction than what I typed"
X Link 2026-02-13T03:31Z 51.2K followers, [---] engagements

"Finance curricula teach Kelly and Markowitz as separate optimizations. Kelly: maximize compounded growth rate Markowitz: find tangency portfolio that maximizes Sharpe ratio They give you the exact same portfolio. F = CM This one formula: [--]. Determines optimal allocation across N assets [--]. Determines optimal leverage (the weights don't sum to [--] - that's a feature) [--]. Maximizes Sharpe ratio [--]. Maximizes geometric growth rate The tangency portfolio IS Kelly optimal - just normalized to leverage = [--] which throws away half the answer. If you're running one optimization for allocation and another"
X Link 2025-11-20T04:20Z 51.2K followers, 10.9K engagements

"I foresee ALOT of this happening. Incumbents will attempt to use regulatory arb to try to preserve their moat. @systematicls Someone vibecoded a $DOCU replica in June too. They sued but looks like the copy is still operating. @systematicls Someone vibecoded a $DOCU replica in June too. They sued but looks like the copy is still operating"
X Link 2026-02-16T07:00Z 51.2K followers, [----] engagements

"Made a post on CN equities before so here's something interesting about IN markets. It's illegal to gamble in India - so indians tend to gamble in the options markets particularly in the weekly options. Any clear-headed person will tell you the primary purpose of the weekly options is to scratch the gambling itch of the populace. That's why options ADV have (correct me if I'm wrong here) 10x volume than their stock counterparts. It's one of the very few (only) markets where this is true. This has produced some odd behaviors where it's pretty cheap to manipulate underlying prices and have it"
X Link 2024-07-23T07:39Z 51.2K followers, 46.6K engagements

"I find the average candidate grossly overestimates how much value they can bring to the table. I love you all but realize that even if you are a top student in a top program from a top university its still negative ev post tx cost to onboard you at least for the first year. There is base a signing bonus some guaranteed bonus there is time taken from the liutenants in training you acquianting you to the system giving you margin for mistakes and of course alpha decay from introducing yet another sentient soul to the datasets know-how and processes; and finally someone else consuming my"
X Link 2024-07-28T07:51Z 51.2K followers, 115.8K engagements

"It is an excellent book. You can tell it was written by a practitioner for (quant) practitioners. One of my favorite parts of the book by far is Chapter [--] where Gappy talks about alpha that spans factors and alpha that is orthogonal to them. He shows by following a reasonable train of logic that alpha orthogonal must be vanishing in N in order for sharpe not to tend towards infinite as you increase the size of your universe. It's something I've intuited about for a long time but he's done us all a service by putting it into a reasonable clear format. PSA: if you read my book draft and send"
X Link 2024-08-08T19:12Z 51.2K followers, 23.2K engagements

""The difficult part is not to make some money with high sharpe - it's to make A LOT of money with sufficiently high sharpe. [--] million dollars a year with [--] sharpe who cares. [--] billion a year with [--] sharpe that's hard." This is honestly the biggest take-away anyone that's looking to move into institutional money management should think about. 1) This was the best gift of understanding my PM gave me when I was a QR many moons ago and I have a deep sense of frustration conveying this to even people that are in the industry. It's quite meaningless to have a high sharpe and a low drawdown if"
X Link 2024-09-03T07:54Z 51.2K followers, 101.1K engagements

""In a single sentence explain the conclusion first" is a good primer on communicating with mgmt"
X Link 2024-09-03T08:52Z 51.2K followers, 37.7K engagements

"A lecture to stanford grads from a CEO of a PE firm having ran it for [--] years to 15B AUM. Summary: [--]. Don't play to not lose play for an asymmetric upside. Loss is often capped to 1x your capital. Figure out levers in business that stack exponentially: strong management team great execution plan etc. Try to get something that can 10x your risks. True for investment businesses as in life. [--]. Fear shows up as "being rationale" or "the smart thing to do". But fear is a pesky thing that will get in the way of playing for the asymmetric upside. [--]. Life is all about compounding. P*(1+r)N one"
X Link 2024-09-10T15:40Z 51.2K followers, 36.5K engagements

"Summarizing @choffstein @0xfdf and @0xFaust12 on the factor discourse - some of these are lessons I've had to learn the hard way. [--]. All returns can be decomposed into idiosyncratic returns and factor returns. Some strategies have idiosyncratic returns others are purely ensembles of factors. [--]. If you don't try to hedge away your factors you will not stumble into high idio var portfolio by accident. A high idio var portfolio almost surely only arises as a result of construction - this is true as long as we are talking about mid and low frequencies where these factors can dominate. An"
X Link 2024-09-19T04:38Z 51.2K followers, 29.4K engagements

"Some thoughts on what it means to compete against a being that is bigger has more resources and can infinitely attrite you. -- Be Aggressive Incumbents have more experience more resources and perhaps most advantageously more time. In games of attrition all else being equal the party with more resources will inevitably prevail. It means the onus is on the disruptors to drive action so that it becomes a game of innovation rather than a game of attrition - it is on the disruptor to be aggressive or the incumbent wins by default. Dare Greatly Disruptors will need to be bold and dare greatly to"
X Link 2024-09-20T08:36Z 51.2K followers, 38.1K engagements

"Quant researcher after a year of cleaning data and setting up feature engineering pipelines age: 22"
X Link 2024-09-22T07:06Z 51.2K followers, 93K engagements

"This is one of the best writings on the realities of a podshop structure - resources are finite but your competitors' (everyone but your pod) will to dominate you is infinite. The entire business is built around the darwinian law of survival. If you survive while others die you get allocated more resources and grow bigger. This cycle repeats until you reach your oxygen limits (capacity) or until you collapse under your own weight. Then your carcass becomes nutrients for the rest of the seedling pods and the cycle repeats. Why working at a so-called Pod Shop can be a great job but may not"
X Link 2024-09-25T05:32Z 51.2K followers, 33.1K engagements

"Highly successful are often people who have shoved hard early on and had it payoff. Billionaires are people who shoved hard early on had it paid off and shoved harder into the next bet and had it pay off. Repeat N times until you are the richest person in the world or Bill Hwang. Path dependency. @__paleologo In theory you're right. In practice you couldn't be more wrong. Remember we're not playing an infinite horizon game. @__paleologo In theory you're right. In practice you couldn't be more wrong. Remember we're not playing an infinite horizon game"
X Link 2024-11-08T03:00Z 51.2K followers, 24.3K engagements

"The short story is that deep learning removes the need for researchers to craft intricate deep features that specify a (complex) relationship that spans a large breadth of data (e.g. multiple datasets) or span a large depth (e.g. long lookback). With traditional econometric or even classical machine learning models the models inability to learn features from the raw data pushes the responsibility of creating "good" features to the researcher. With deeper models the model implicitly jointly learns what features are the best to predict future targets. In facial image learning for example there"
X Link 2024-12-10T02:36Z 51.2K followers, 46.4K engagements

"Uni-dimensionality (in your career) is the price to pay for exceptional payoffs in your career. Some people are willing to pay for that and some people do not; if you are one of those people who can stake everything on a small probability of fat-tailed success in your career then you will reap this risk-premium all throughout your life. Kobe Bryant's trainer Tim Grover speaks on his mentality: https://t.co/tktOcFA8bV Kobe Bryant's trainer Tim Grover speaks on his mentality: https://t.co/tktOcFA8bV"
X Link 2024-12-14T15:28Z 51.2K followers, 31.3K engagements

"As someone who has been on the inside of an index rebal team. this guy is a real one. For a real and rare peek into how index rebal teams are thinking about US/APAC indices this account is the way to go. Adding on to this and since topix was mentioned they clearly state that index reconstituions are based on FFW but their FFW methodology's dealings with liquidity rounding and discretion based on what is considered a major shareholder makes the reconstitution model particularly fustrating. Also the only fun fact about modern day index rebal teams is that they can get up to 99% reconstitution"
X Link 2024-12-17T04:32Z 51.2K followers, 36.8K engagements

"Index rebal was a great memory and truly eye-opening for me because there's nothing quite like seeing a (idiosyncratic) cumulative returns chart start creeping up into announcement date and then fall past the strike date. It was a very visceral fuck you to market efficiency for me and made it clear that there were obvious systematic strategies that could extract alpha from the market. To get the graph below one normally assumes perfect foresight and then to realize the strategy one constructs reconstitution models based on the index methodologies and attempts to creep the model accuracy"
X Link 2024-12-17T04:57Z 51.2K followers, 26.5K engagements

""Competition is for Losers" - Peter Thiel There is a lesswrong post about being the (pareto) best in the world that essentially argues that when viewed in the appropriately high number of dimensions in the skill space you can find yourself on the pareto frontier where no competition exists and you are quite literally the best in the world. It is worth thinking about this in the context of quantitative finance as well I like to imagine that every space (defined as some combination of dimensions e.g. asset_class x region x .) has a limited amount of capacity and once there are too many blokes"
X Link 2024-12-20T09:26Z 51.2K followers, 34.2K engagements

"This was an amazing listen a pod with @choffstein and @ScottPh77711570. Listening to Scott's evolution in the markets was really interesting. Given a team of [--] developers and [--] (great) researcher I wouldn't call this a retail trading team anymore; but it was packed with infinite wisdom on how to find edges in the market where you can be the pareto best. If anything it was a masterclass on finding a corner of the market that is a weight class you can dominate. You also hear some of Scott's frustration on what it means to (effectively) be a PM where a large part of the job seldom spoken about"
X Link 2025-03-08T07:17Z 51.2K followers, 30.7K engagements

"I am NOT an index rebal PM and have never been. But I spent the early parts of my career in a great index rebal team and (hopefully) have developed a decent understanding of the strategy and a great respect for the outstanding PMs in the space. Every man must choose his own way and index rebal was not the one for me. As mentioned before there are a great many discretionary decisions that go into a good index rebal portfolio and a great amount of time that has to be invested in understanding the index methodologies and rules; and developing heuristics for these rules that line-up with"
X Link 2025-03-09T00:51Z 51.2K followers, 27.6K engagements

"It's almost irony that some quants fight tooth and nail over optimizations measured in basis points but neglect arguably the most important optimisation of all - picking a great spouse. I can't even begin to describe how much alpha a loving spouse has brought to my life. It's kindness when times are rough patience when tempers are short buffer when circumstances are dangerous resonance in peaks dissonance in troughs. I've met my spouse in university and have been together with her when I was nothing to slightly more than nothing. And everyday through the games plays and strategy - no decision"
X Link 2025-03-12T18:10Z 51.2K followers, 21.5K engagements

"Gappy's books are both misnomers. Advanced Portfolio Management is not advanced and Elements of Quantitative Investing is not elementary"
X Link 2025-04-09T13:19Z 51.2K followers, 45.2K engagements

"good risk systems can make disproportionate pnl for fairly large books since most friction arises from being large enough that you can't just dump risk when something goes awry. for small books just focus on alpha and derisk to zero when something goes wrong. --- there's another lesson here: professional risk takers (pms qrs etc) are being paid to take risk - firms generally understand when you drawdown within your mandate. and as long as you are within your risk budget it is generally fine to drawdown even if its severe. however firms are much less forgiving to a pm not making money from his"
X Link 2025-04-27T13:39Z 51.2K followers, 32.2K engagements

"There are only [--] ways to make money: [--]. You are selling a service - e.g. providing liquidity/options for willing buyers. [--]. You are earning a risk premium for doing something risky with your cash e.g. you are willing to buy growth stocks before they have become empirically successful and hold them through the ups and downs. [--]. You have an information edge that comes from either having access to data that the dollar-weighted average participant do not or are able to transform and extract excess signal from contested data. [--]. You have a structural edge that the dollar-weighted average"
X Link 2025-04-27T21:27Z 51.2K followers, 191.6K engagements

"Thoughts on differences in responsibilities between a researcher and a PM and how to prepare for a PM role as a researcher looking to cross the chasm: [--]. The biggest difference between a researcher and a PM is scope of the investment process. A researcher is often only expected to understand how to derive signals that are value-accretive to the investment process and has little exposure/insights to: a. Data b. Friction and Implementation Shortfall c. Portfolio Construction d. Execution e. Project Management [--]. To cross the chasm and to land a PM role it pays to have sufficient understanding"
X Link 2025-05-10T16:03Z 51.2K followers, 18.5K engagements

"To be considered for a leadership position in a trading team: [--]. Be pathologically present. Things go wrong all the time. There are people who are available and then there are people who are not. Nobody wants to work in a place where they do not feel supported its hard but having a habit of being able to show up when things are rough tells the team that you are committed to all of the teams problems. Make your presence known even if its something simple like: I see X and Y breaking let me know if I can be of any help. The team may not always reply you especially if they are dealing with"
X Link 2025-08-17T21:47Z 51.2K followers, 13.4K engagements

"A couple more thoughts on how to accelerate standing in any environment: [--]. Being a yes-man makes you seem like a pushover and generally it is scary to put pushovers in positions of leadership or power; but if there are projects and responsibilities that you would like to yield: you should enthusiastically accept them. Hesitation is a negative signal and it is never a good feeling to be on the receiving end of hesitation. To avoid being a pushover and to make enthusiasm count it then necessitates that you clearly demarcate between projects you are excited about (and make it clear that you"
X Link 2025-09-11T12:12Z 51.2K followers, 15.3K engagements

"Been thinking about this from @__paleologos MIT talk. Every firm considered to be at the top of their quadrants (prop/hf/taking/making) has won monopolistic profits in one niche area/style of the markets. Its not impossible to make money from markets; but it does feel like uncontested returns is in secular decline even for fairly complex investment processes. If one is starting out and has the explicit goal of creating something world class where does one go"
X Link 2025-09-14T14:56Z 51.2K followers, 55K engagements

"One can either be exceptional at one thing (career/family/sports/hobby/etc.) at the expense of everything else or being decent at a lot of things or being good at some things. Unfortunately trying to be good at everything and to everyone is a form of shrinkage towards mediocrity. To be exceptional at one aspect is to be a disappointment at some aspects"
X Link 2025-09-29T16:32Z 51.2K followers, 16.2K engagements

"Photo from efinancialcareers I have a fun lore for this. When I was an itty bitty junior [--] months into "very large and very cutthroat hedge fund you know" I went to HR asking about the common paths QRs can become PMs - saying I had an aspiration to be one and wanted to understand the typical statistics of inhouse conversions (if any). The next day my PM pulled me aside saying "I got a call from head of systematic investing HR told him that you were a flight risk". I wasn't too fazed cause we were fairly tight but that day was one of the two times I was radicalized to really want to fuck the"
X Link 2025-10-06T15:24Z 51.2K followers, 19.4K engagements

"This is true HL was pretty wicked. Basically liquidated all their traders took on their large cap positions and then rode the recovery for themselves. Hyperliquidated. @systematicls Even being short in size doesn't help if the venue puts you in as the counter-party to the deleveraging trade (I'm looking at you HyperLiquid.) School fees (h/t @iarbpairs): -- "Auto-deleveraging strictly ensures that the platform stays solvent. If a user's account value or @systematicls Even being short in size doesn't help if the venue puts you in as the counter-party to the deleveraging trade (I'm looking at"
X Link 2025-10-10T23:35Z 51.2K followers, 48.9K engagements

"There are [--] classes of participants on fintwit and these classes are NOT static: [--]. You are making money but you want to psyops the normies so you say the most insane shit that is blatantly untrue / orthogonal to WHY you made money [--]. You are losing money but you want to psyops the normies that you are making money because you have something to gain from it [--]. You are making money and you actually explain why you made money [--]. You are losing money and you've actually declared you've lost money People naturally see [--]. with great disdain but [--]. will destroy you in the long run"
X Link 2025-10-11T09:51Z 51.2K followers, 44.1K engagements

"The most insane take-home-assignment over the next year is to build a market maker around the past [--] hours"
X Link 2025-10-11T10:11Z 51.2K followers, 11.3K engagements

"You see this everywhere because there are real issues of "being too smart" and then also because intelligence has VERY little correlation with who calls the shots. Intelligence seems to lead to two major issues: [--]. Analysis paralysis the web of things to consider becomes infinitely larger IF you can actually see the web. Midwits see [--] edge from the starting node and so they don't have to consider more than that. If you see [--] edges later it's hard not to try to consider them. [--]. Life is hard being able to see all the ways things can fail IS going to deter you from doing things that can fail."
X Link 2025-10-18T22:27Z 51.2K followers, 45.5K engagements

"I wholeheartedly agree. Quant is in secular decline. If you are young and smart don't become a kwant There's more interesting things to do there's a handful of jobs and a knife fight for them Anyone smart enough to land one of these jobs can do anything and money isn't gonna be a problem If you are young and smart don't become a kwant There's more interesting things to do there's a handful of jobs and a knife fight for them Anyone smart enough to land one of these jobs can do anything and money isn't gonna be a problem"
X Link 2025-10-19T00:30Z 51.2K followers, 36.3K engagements

"One of the interesting cases of spreads that are not collapsible is taking LONG(CSI FUTURES) + SHORT(CHINA A SHARES) It returns about 12% annualized returns with [---] - [---] sharpe and is extremely persistent but this spread will never collapse because it is virtually impossible to short A shares in size AND there is persistent demand for CSI shorts because that remains the only way to get the short leg in a market neutral strategy in A shares. This however is one of the fun parts of being a small player suppose you are not looking for size and want a "real edge" with strong economical backing"
X Link 2025-10-19T20:06Z 51.2K followers, 13.3K engagements

"I make money because: I am harvesting risk premia ✓ I have alpha from superior models ✓ I have alpha from superior data/information I have superior access to cost/rebates I don't I keep my money / sustain a moat because my strategy: Has fat tails (kills people/people kill themselves) ✓ Has an infrastructural barrier to entry (hard) ✓ Has a knowledge barrier to entry (hard) Has a psychological barrier to entry (nasty) Is opportunistic / special situations (no fitting) Has a structural barrier to entry (unfair access) I don't than the dollar-volume weighted participant. Know thy strength"
X Link 2025-10-20T16:21Z 51.2K followers, 39.3K engagements

"First $0 to $1 billion AUM THEN $0 to $1 billion net worth. Thanks Corey for the high quality pods - you've inspired a lot innovation in our industry because of your marketing side hustles $0 to $1 billion in [---] days. Thanks for all the support along the way. $0 to $1 billion in [---] days. Thanks for all the support along the way"
X Link 2025-10-21T23:48Z 51.2K followers, 43.1K engagements

"If you backtest (without tcosts/spreads) pnl = close.shift(2) * c2c_returns it will be exactly opposite to pnl = close.mul(-1).shift(2) * c2c_returns -- The above has nothing to do with whether or not markets are zero sum (in fact under transaction costs it will be negative sum) and makes strong and unrealistic assumptions about competition/slippage/adverse selection being able to realize the signal (e.g. always executing at the close/vwap). Even not accounting for fees its fairly trivial to see that if you have a losing signal and you are always able to buy at the bid (adverse selection) it"
X Link 2025-10-23T01:26Z 51.2K followers, 20K engagements

"This is my quant - this is exactly how we implement discretionary HFT models. Fibonacci isnt about lines its about liquidity geometry. Retail sees golden ratios. Professionals see where algorithms rebalance risk and where liquidity pools exist. 🔹 Advanced Retracement Zones (Accumulation / Reaction Points): [-----] 0.382: Algorithmic momentum https://t.co/KXpDdu61Sv Fibonacci isnt about lines its about liquidity geometry. Retail sees golden ratios. Professionals see where algorithms rebalance risk and where liquidity pools exist. 🔹 Advanced Retracement Zones (Accumulation / Reaction Points):"
X Link 2025-10-23T08:30Z 51.2K followers, 22.8K engagements

"What a great post; "As a result venture can feel like the opposite of a network-effect business: the bigger the portfolio the more dilute your attention and the weaker VCs incremental value-add on helping any particular snowflake attach to the snowball." If I've learnt anything building it's that one of the most frustrating effects is when scale feels like attention is being divided 1/N. And the problem is that sometimes attention is being driven to where talent is most lacking and not where it's most deserving. Some projects are harder than others and naturally need more attention to be"
X Link 2025-10-24T02:49Z 51.2K followers, 27.6K engagements

"I especially liked Naval's explaination on this: Naval Ravikant on High Agency People 1) The best teammates have agency: they spot problems fix them without hand-holding and dont need to broadcast it. 2) Startups face infinite problems; leaders cant see or solve them all. 3) Organizational capacity scales with independent problem-solvers. 4) Management bandwidth is the constraint: if someone needs 10% of your time you can manage 10; at 5% you can manage [--]. 5) Ideal traits: low ego minimal updates or silly questions professional results-oriented. What a great post; "As a result venture can"
X Link 2025-10-24T02:58Z 51.2K followers, 22K engagements

"Everyone knows that quants in industry use PyFolio Zipline OpenBB Empyrical AlphaLens and RiskFolio-Lib religiously. Never a day goes by where I don't ship strategies with import RiskFolio-Lib"
X Link 2025-10-24T09:50Z 51.2K followers, 93.5K engagements

"Everyday I write strategies with dollar volume bar sampling structural break chow test features fed into my uniqueness sampled ensemble model for triple-barrier labelling. Then I take the labels and feed it into my hierarchical clustering portfolio optimization and send it over to my HPC computer for high frequency trading. Everyone knows that quants in industry use PyFolio Zipline OpenBB Empyrical AlphaLens and RiskFolio-Lib religiously. Never a day goes by where I don't ship strategies with import RiskFolio-Lib. https://t.co/E3tzQVbGF3 Everyone knows that quants in industry use PyFolio"
X Link 2025-10-24T11:22Z 51.2K followers, 47K engagements

"To all the haters commenting on this you guys just don't get quant finance. It used to be that when you dug the soil for gold you would be able to unearth entire gold bars. The game has changed. Now when you dig the soil for gold you get only gold dust particles. Quant finance is about assembling a factory now. You have to have millions of assembly workers each focused on one component of the quant finance pipeline if you want ANY hope of succeeding. In my team we have engineers whose sole job is to write import statements. That's how specialized our quant pipeline is. We're here for not just"
X Link 2025-10-24T11:45Z 51.2K followers, 42.8K engagements

"Nothing really happens: Polymarket edition. You could probably earn a nice stack by just buying NO across the board; of course if you do so with markets that already have expensive NO you introduce more variance to your strategies but that's the idea of a statistical edge; AND that's where its the most discounted. It's not a sure thing but if you can apply across all markets and as N gets bigger your sharpe will increase. Analyzed [-----] Polymarkets at 12hrs before resolution here is what I found polymarket is within 2-3% of the actual outcome its most accurate at 80% odds (variance of 0.3%)"
X Link 2025-10-25T01:57Z 51.2K followers, 49.4K engagements

"Why I am still extremely bullish about prediction markets as an asset class. [--]. The bar for comfort is so low now that "going to zero" doesn't actually feel like "going to zero" IF you have a base skill-set in most developed countries. For a great majority of us if you are chronically online and have a white-collar job; going to zero is at most going back to a 9-5 job you don't care for. [--]. The gap between the elite and the common is so large and it has become painfully clear for the majority that working hard and climbing some kind of ladder is NOT going to cut it at escaping the common"
X Link 2025-10-25T18:26Z 51.2K followers, 33.3K engagements

"Pathologically stupid take. [--]. The maximal benefit of prediction markets is to approximate the probability distribution of an event before it has collapsed [--]. When there is inside information the probability distribution has already collapsed - there is no societal value in estimation anymore (it's a done deal) [--]. To best approximate the probability distribution and get maximal benefit you need participation from many diversified great forecasters (real forecasters). The aggregate estimate then becomes humanity's best estimate of the probability distribution [--]. Good forecasters adjust"
X Link 2025-10-26T04:46Z 51.2K followers, 25K engagements

"Whos going to tell him he likely forgot to .shift(2). The new value momentum system is progressing nicely. As long as I can't break it I will swap out the current momentum systems in my latest portfolio in the next end of month rotation. I moved away from lambdarank to lightgbm regression as I found ranking on this basis to be https://t.co/zIcghfK0Ow The new value momentum system is progressing nicely. As long as I can't break it I will swap out the current momentum systems in my latest portfolio in the next end of month rotation. I moved away from lambdarank to lightgbm regression as I found"
X Link 2025-10-26T12:58Z 51.2K followers, 73.9K engagements

"Top [--] things Ive said to break a researchers heart: [--]. You didnt shift your positions. You assumed instantaneous execution + [--] time for signal generation. [--]. Your analysis is NOT PIT. Youve done your analysis over the entire in-sample period and then drew conclusions over it. [--]. Your costs are in the wrong direction. [--]. You took the data event timestamp rather than the insert/system timestamp and assumed that data is available at the event timestamp. You actually assumed the data was available one whole day/week before it actually was. [--]. You assumed the data vendor is not a slimy piece of"
X Link 2025-10-26T13:39Z 51.2K followers, 46.2K engagements

"Polymarket teaching the public to learn about betting on demand/flow rather than fundamentals is not on my bingo card this year. To the uninitiated it matters not if jesus christ WILL return next year what matters is if there is going to be real dollars that are going to flow into buying YES jesus christ is going to return next year. You can trade the flow/demand even if its disjoint from fundamentals. Might just be trading a "price rise from meme attention" thesis. current polymarket bets trending are trump getting into office for a third term (higher odds than kamala) and Jesus Christ"
X Link 2025-10-27T01:11Z 51.2K followers, 39.7K engagements

"BREAKING: Prediction market bettoors discovers buying strangles Just opened a new volatility setup on @Polymarket. I rarely share my live positions but this ones too clean not to. Bitcoin now at $113700📈 Opened https://t.co/N0QnHiLuZJ YES BTC above [------] at 13c NO BTC above [------] at 18c When price hits either 112k or 116k https://t.co/5ZUA4IjC6u Just opened a new volatility setup on @Polymarket. I rarely share my live positions but this ones too clean not to. Bitcoin now at $113700📈 Opened https://t.co/N0QnHiLuZJ YES BTC above [------] at 13c NO BTC above [------] at 18c When price hits"
X Link 2025-10-27T01:40Z 51.2K followers, 36.8K engagements

"1. Pairs trading is profitable in all markets where the instruments share some common factors. [--]. It is economically intuitive and is a fairly meaningful trade whenever you put it on. E.g. Long NVDA Short AMD fundamentally means something. [--]. If X and Z has common factors M and no trend; and your model is well-specified then all movement of the spread must come from noise or is a temporary mispricing. [--]. When the mispricing gets large enough people are incentivized to come and collapse it. [--]. This is fairly intuitive to understand but the snaffu of pairs trading is the following: a. For"
X Link 2025-10-31T00:36Z 51.2K followers, 69.2K engagements

"This is a fair point - some observations: [--]. If you are joining a 5-10 year old team and expect to be let into the inner circle / do non-grunt work you are going to be in a hell of a surprise. For an established team there is literally [--] incentive to teach you anything other than the most grunt work. It is often extremely Darwinian as well you are given a dataset you are given a research direction. Can you produce signals that are orthogonal and value-adding If not bye [--]. However there are some commingled research environments that have "fairly open" wikis. It can still be useful to read"
X Link 2025-10-31T03:04Z 51.2K followers, 33.7K engagements

"This guy doesnt get it. Implementing projects and marketing them is exactly how you get noticed. If I was starting over and was a brokie and needed a job I would find a few projects that seemed promising implement them and send a short note about my findings and ask thoughtful questions in whatever field I wanted to work. THIS IS LITERALLY THE WAY TO BREAK IN. You can implement a bunch of papers from scratch post them on Twitter & the VP of OpenAI research will contact you No place like twitter https://t.co/NUDPrJuUNA You can implement a bunch of papers from scratch post them on Twitter & the"
X Link 2025-10-31T04:12Z 51.2K followers, 29.9K engagements

"Honestly this is the truth"
X Link 2025-11-02T13:58Z 51.2K followers, 16.7K engagements

"From Quantitative Portfolio Management: [--]. Any future return prediction or forecast (call it ) needs some predictor data (call it ) possibly multi-dimensional so the forecast is a function of the predictor: = f (). We will call this forecast function a model. Page [--] [--]. To have predictive power a predictor must be relevant to the outcome or put simply make sense for the context. Further predictors are best cast in the form of featuresformulaic or algorithmic combinations preferably expressing clear forecast ideas and maximizing predictive power. The terms predictor and feature are used"
X Link 2025-11-07T15:00Z 51.2K followers, 19.5K engagements

"If your returns feature doesn't look like this I don't even want to talk to you (This is a real feature btw)"
X Link 2025-11-08T02:13Z 51.2K followers, 42K engagements

"Honestly one of the harder parts of quant is to pick the feature implementation that you think best represents your target. Consider these [--] options. On the X axis we have the feature values and on the y axis we have the target values. They are both essentially the same feature but have a different transformation applied to them and therefore changes their relationship to the target. Which one would you go for"
X Link 2025-11-08T02:21Z 51.2K followers, 60.8K engagements

"Factors are not that hard to understand. Lets say you have [--] stocks just imagine they are [--] separate boxes of fruits that are fungible and infinite in supply. In box A: they always have [--] oranges and [--] bananas and [--] coconuts. In box B: they always have [--] oranges and [--] bananas and [--] coconuts. There are markets on these boxes but not on the individual fruits. You have spent a life time of mastering the art of coconut prices prediction and all signals point to coconut prices falling TODAY - you feel it in your bones. You want to speculate that coconut prices are going to fall but remember - you"
X Link 2025-11-12T04:39Z 51.2K followers, 13.2K engagements

"When I write that HFT/MFT are pushing up against each other: Derivative bland tasteless obvious When FT writes the same thing: Visionary inspiring insightful https://t.co/b39svyp4Bz https://t.co/b39svyp4Bz"
X Link 2025-11-13T02:02Z 51.2K followers, 23.1K engagements

"Every quant knows the secret to alpha is: 1) Download [--] years of SPY data from Yahoo Finance 2) Backtest [--] different indicators until ONE works 3) Curve fit until your Sharpe is [---] 4) Show it to NO ONE who might ask questions 5) Paper trade for [--] days 6) Go live with your life savings Then when you lose everything blame regime change and market makers hunting your stops. The market just "changed" your 47th indicator that worked on 2018-2023 data should've worked FOREVER. Has nothing to with the fact that you assumed you could trade at close prices published after-hours"
X Link 2025-11-16T12:55Z 51.2K followers, 149.9K engagements

"ADVERSE SELECTION IN SMALL TRADING FIRMS Most small trading firms fail to scale because they can't solve a simple economics problem: If BIG PODSHOP offers $40mn + $5bn book for top talent and you offer $1mn + $100mn book you're priced out of the best traders. So who CAN you hire [--]. People who underpriced themselves - they leave once they wise up [--]. People who aren't worth your bid - you get adversely selected The information asymmetry is brutal. Experienced hires always know more about their true probability of success than you do. This makes hiring experienced talent -EV for small firms. Big"
X Link 2025-11-16T14:07Z 51.2K followers, 27.5K engagements

"You cannot build edge in markets you don't understand. Harris teaches you how markets ACTUALLY work and who are the participants you are trading against and WHY they trade not the fantasy version you learned from YouTube. Most rookie mistakes come down to not understanding why your orders don't fill when you want them to and DO fill when you don't. This book fixes that. @experquisite @therobotjames @Robot_Wealth @systematicls @0xfdf @AgustinLebron3 Can I recommend the aging Trading and Exchanges by Harris @experquisite @therobotjames @Robot_Wealth @systematicls @0xfdf @AgustinLebron3 Can I"
X Link 2025-11-17T02:17Z 51.2K followers, 57.4K engagements

"An Introduction to Statistical Arbitrage In The Form Of Books: Big Picture: Quantitative Portfolio Management (Isichenko) A-Z of statarb greatest hits: Forecasting Forecast Combining Creating Signals: Quantitative Trading (Chan) Mean reversion and momentum intro greatest hits: Implementing Mean Reversion Interday Momentum Finding Alphas (Tsulchinky) Alpha discovery framework greatest hits: Introduction to Alpha Design How to Develop an Alpha Case Study Portfolio Management: Advanced Portfolio Management (Paleologo) Systematic portfolio framework for first-timers greatest hits: [---] Separation"
X Link 2025-11-17T14:07Z 51.2K followers, 51.1K engagements

"Read @0xfdf's articles. You can't know your alpha until you know your betas. When you buy AAPL because you think it's undervalued you're actually betting on: 1) AAPL's idiosyncratic risk (the part you care about) 2) Tech sector performance 3) Market factor (S&P 500) 4) Style factors (value momentum etc.) For most stocks the alpha you want is a minority of total risk. The rest is just exposure to factors you don't have a view on and can't predict. This is why risk models matter. You need to separate what you're actually betting on from everything else you're accidentally betting on."
X Link 2025-11-17T15:21Z 51.2K followers, 29.6K engagements

"1 year ago: 76% accuracy cost $100/task Today: 73% accuracy costs $0.30/task That's 300x cheaper for [--] percentage points less accuracy. If your AI startup's moat was "we have access to better models" or "we can afford more compute" you just watched it evaporate in [--] months. Infrastructure barriers don't stay barriers when costs drop 99.7% annually. The only sustainable moats for AI startups are: 1) Proprietary data: You have exclusive access to domain-specific data that materially improves model performance for your use case. Not "we scrape public data better" but "we have data no one else"
X Link 2025-11-18T03:32Z 51.2K followers, 26.3K engagements

"Everyone knows the secret to getting noticed is: [--]. Build the same pairs trading backtest as [----] other candidates [--]. Use the same yahoo finance KO and PEP data [--]. Get identical results to everyone else [--]. Wonder why hiring managers don't call Projects that actually get you hired show you found edge somewhere. Gaming arbitrage that made money. Poker model that crushed games. Sports betting with real +EV. Marketplace pricing errors you exploited. Hiring managers see [---] SPY backtests a week. They remember the person who made $500 betting on number of Elon tweets (real story)"
X Link 2025-11-18T15:04Z 51.2K followers, 14.4K engagements

"Alpha drop: Numerai forums. Numerai obfuscates all features before giving you data. feature_1. feature_1050 zero semantic meaning. Can't lean on fundamentals or papers forced to discover pure statistical relationships. The forum is people trying to crack why things work without being able to anchor to conventional wisdom. Forced creativity at scale. Some threads: why recent 1yr models beat long history why neutralization backfires handling regime shifts when you can't see what actually changed. Good hunting ground if you want to see statistical thinking divorced from "but Fama-French says"
X Link 2025-11-18T16:24Z 51.2K followers, 15.4K engagements

"The best life is about maximizing expected value over your remaining lifespan. Stop optimizing for low volatility comfort. That's just curve fitting to your current local maximum. The play is: 1) Identify what has asymmetric upside (skills relationships projects where downside is capped but upside is huge) 2) Size positions appropriately (don't YOLO but don't under-allocate to high-EV plays either) 3) Rebalance when you're overexposed to things that stopped working Most people are running a backtest on someone else's past and calling it a strategy. They found one path that worked (college"
X Link 2025-11-19T07:27Z 51.2K followers, 17.7K engagements

"I do the opposite. Scale up balls long during drawdowns. Buy the fucking dip. I feel morally obliged to do so. If I can't size up when my strategies are hurting I'm admitting I don't trust my own work. If I don't bet on myself no one will. Everyone has conviction at equity peaks. That's not conviction that's comfort increased size on a new model by a factor of [--]. had [--] options: wait for a drawdown slap it at open at peak equity (you know which one I did) now on a more serious note how do you do it if you are beta testing a new model how do you think about adding size to it increased size on a"
X Link 2025-11-19T08:21Z 51.2K followers, 41.2K engagements

"There are [--] types of drawdowns: 1) Slow death: a. Your realized Sharpe decays over time. b. Strategies never return to peak performance. c. This is systematic alpha decay from competition. d. Size down as a function of Sharpe decay. 2) Sudden cluster: a. Your best bangers suffer massive drawdowns out of nowhere. b. Large number of idiosyncratic bets go wrong way. c. This is transient flows from forced sellers in your style. d. Size up aggressively. The first is permanent. The second is opportunity. Most traders treat both the same. That's expensive. I do the opposite. Scale up balls long"
X Link 2025-11-19T08:49Z 51.2K followers, 52.9K engagements

"If you own [--] tech growth stocks you don't have [--] independent bets. You have [--] bet on the tech-growth factor sliced [--] ways. When that factor crashes all [--] move together. Most of your returns come from systematic factor exposures not stock picking. And systematic risk cannot be diversified away by adding more stocks with the same exposures. Real diversification comes from one of two sources: 1) You have no view on idiosyncratic returns and have enough of them (30-50) that idiosyncratic returns of your portfolio is [--] and then the returns of your portfolio is dominated by FACTOR exposures not"
X Link 2025-11-19T14:20Z 51.2K followers, 17.4K engagements

"Those who cannot act without complete information fail safes and guarantees will always be slaves to those who can. This isn't pessimism - it's mechanism design. [--]. Perfect information doesn't exist in competitive environments [--]. Fail safes and guarantees are priced in - everyone sees the same "sure things" [--]. The information you need often only reveals itself AFTER you act [--]. Waiting for certainty is itself a decision - to let others define the playing field While you're gathering the last 20% of certainty someone else has shipped learned and iterated. They're now setting the terms you have"
X Link 2025-11-20T10:19Z 51.2K followers, 41.1K engagements

"Numerai Cheatsheet [----] Introduction Wanted to catch-up on a Numerai a platform I have always thought was pretty cool and I ended up consolidating enough information to warrant putting together a quick cheat sheet for http://x.com/i/article/1997036098074329088 http://x.com/i/article/1997036098074329088"
X Link 2025-12-05T21:22Z 51.2K followers, 33.2K engagements

"Given that we probably have one of the largest communities of professional quants in discord right now. I'm considering renaming our discord to "OnlyQuants""
X Link 2026-01-14T14:08Z 51.2K followers, 45.6K engagements

"You guys think I'm joking but there are truly some incredible people in the Discord: Given that we probably have one of the largest communities of professional quants in discord right now. I'm considering renaming our discord to "OnlyQuants". https://t.co/K9WIntoKek Given that we probably have one of the largest communities of professional quants in discord right now. I'm considering renaming our discord to "OnlyQuants". https://t.co/K9WIntoKek"
X Link 2026-01-15T16:29Z 51.2K followers, 27.5K engagements

Limited data mode. Full metrics available with subscription: lunarcrush.com/pricing