Happy Anniversary. 40% bagged + beating Blackrock's BDMIX by a mile😇 69% weekly win rate (that's the trade duration, weekly rebalancing Tuesdays)

Today’s Daily Watchlist comes from our exclusive Support/Resistance dataset, every level backtested 12 different ways, then filtered to the single best historical hold rate weighted by reward:risk. Standouts: $CCO at pivot support with a 92.9% 7D hold rate and 15.19 R:R; $APGE showing strong support quality at 86.7% / 14.55; $HCMA is the cleanest short-side resistance profile with a 98% hold rate. Data-driven levels, not guesses. Not financial advice.

Can LLM-based trading systems increase market volatility? The idea is quite simple: 1. LLMs tend to act as "belief homogenizer" - producing similar outcomes based on similar trusted sources 2. This creates clustering of trades 3. The more crowded clusters become, the higher the resonance Came across several interesting studies that support this idea. 1. SSRN paper “Strategic Homogeneity in Trading: Amplifying Market Volatility and the Emergence of Mini-Crashes” argues that convergence of trading strategies can amplify market volatility 2. "Algorithmic Trading and Liquidity Commonality" paper finds that increased algorithmic trading contributed to stronger liquidity co-movement: 5-minute liquidity co-movement rose by 30–50%, and daily liquidity co-movement was 90% higher. In plain English, automation did not just change individual trades; it made liquidity conditions move together more 3. While the previous 2 explore fundamental processes and are not exactly referring to LLMs, this third one actually supports the common ground. "When ChatGPT Stops Talking: GenAI-induced Retail Herding and Systematic Risk" by Xiaoxue He. The finding is that ChatGPT acts as a "belief homogenizer," synchronizing retail beliefs and trading decisions Basically, the more AI systems are in place, the sharper the swings. And while the systems might not be wrong at all, amplitude may cause risk management systems to break down and trigger response from stop loss positions and market makers. 2 quick options here: 1. Use short-based volatility based position sizing. 6-12 months max 2. Play around with portfolio composition. Market neutral framework has done wonders for us lately

LLMs might accelerate alpha decay in trading Here's how it works in a nutshell: - free basic data - conventional "top of mind" indicators and strategies - something LLMs know and trust because of multiple sources - very similar backtest fitting All of these combined lead to generating signals that use very similar logic. This crowding effect will cause 2 things: alpha decay and volatility spike. I'll expand on vol next time, alpha decay is a big enough topic One recent paper in The Journal of Finance shows that anomaly returns are concentrated in the first month after information release and then decay soon after. In other words, even before LLMs, the monetization window for many signals was already narrow. Now add LLMs to the research stack. A 2025 paper on LLM-driven alpha mining makes the risk explicit: LLM-based approaches can rely too heavily on existing knowledge, generating homogeneous factors that worsen crowding and accelerate decay. That is the part I think the market is underestimating. If many teams are drawing from overlapping public data, familiar factor logic, and similar backtesting workflows, LLMs help discover signals faster. But they also make it easier to produce the same kinds of signals faster. My bet: LLMs will not kill all alpha. But they will compress the shelf life of copyable alpha. So what can traders do about it? Option A: accelerate strategies launch. Have a pipeline, be ready to halt underperformers. Expensive, time consuming, capital heavy Option B: add alternative data, create/buy prop indicators, work on portfolio composition

Claude and ChatGPT enter a bar. 2 pints later Claude says: I wish I could cut my costs by 90% ChatGPT says: let me explain.. Claude: I can explain it too, but I just want my costs cut (ancient joke about consultants)

Are parrots actually smart? A global survey of 3,750 households owning a parrot found that 54% of family members ignored their parrot’s opinion when planning their household chores.  Another 33% haven't even asked the parrot anything. Combined, eight in ten households are avoiding or rejecting the bird that cost their household an average of $54 (million) feeding this year. The problem is in the core technology of the parrot. It mimics, repeats patterns and most certainly doesn’t have any realistic idea of it’s own limitations and capabilities. So at the end of the day, it’s all about the input. If you teach it to swear and make poor jokes, the guests might be offended. And no kids parties should be allowed. But it can serve as a decent alarm clock, fully automating a workflow of waking up when the sun rises. This obviously was about the LLMs (sorry, can’t force myself to write AI – nothing intelligent as of now). Only 9% of workers trust AI for complex business-critical decisions compared to 61% of executives. Workers lose the equivalent of 51 working days per year to technology friction, up 42% from last year, almost exactly equal to the 40-60 minutes per day Goldman Sachs says AI saves workers who use it correctly. This gap is frightening. LLMs repeating generic fluff it found somewhere all while boosting executive’s egos shows a very scary pattern. A tool that’s only good enough to boost productivity does not do that. Instead, it supports decision making of the highest order, all while no critical thinking, data validation or experienced-based insights are part of it. So if you use it to make your investment decisions, make sure the data processes are set up correctly, risk management is intact and there’s a “switch off” button. inspired by @HedgieMarkets and @tomfgoodwin

5 reasons LLMs might have actually peaked the way we know them Like any new tech that actually works - internet, blockchain, tulips (just to be a bit cheeky) - the excitement seems slightly exaggerated. Here are a couple things I noticed and my takeaway. 1. Robots.txt. Basically, platforms tend to protect their data. Even those relying on generating LLM traffic like reddit and LinkedIn limit "scraping" requests. The likes of X.com have fully blocked any scraping and push users to X API with extortionate prices. 2. Paywalls. Anything remotely valuable is hidden behind paywalls. Authentic thoughts that used to be blog posts are now substack subscriptions. And I didn't even start talking about data 3. "Poisoned pills". The amount of "It's not Y, it's Z" copy on the internet makes me want to become violent. New gen of general LLMs eats what old gen of LLMs has produced 4. Cost. Ok, the search market is conquered. Now losing money doesn't really make sense. Anthropic already hit hard with it's OpenClaw decision. I've seen reports of people going up to 50x on actual cost. Because guess what: AI requires compute power. And it's expensive. Especially in the geopolitical energy crisis. 5. Ancient "automate what works" problem. I remember our first CRM integration in a bank I used to work at in 2008. We didn't really have a defined sales process. And I mean - not even remote. No meeting reports, no performance metrics, just end of month revenue assessment. And then suddenly everyone had to fill in a form with 20 inputs after each call. Obviously, didn't fly. And that's what is happening now. Processes that don't work (and people have very vague understanding of how they should) suddenly get an AI wrapper. And that's all it is. A wrapper. LLMs don't create value. They help cut time. That's how they should be referred to. A time saving tool. They don't invent. They don't create. They use the available input to generate an output that requires checking and cleaning. When we launched Alfred, our AI assistant, the effort to prevent hallucinations was probably 80% of the work. My key takeaway: I strongly believe all those AI agents will generate evidence that will lead to the only possible conclusion: LLMs are a wrapper. And the hype will go down. Real adoption will bring real people in. Policies and standards will be created. Vertical clear use cases generating 1 outcome at a time. Just like a good software. Security, reliability and compliance will take over again (they always do when any process settles down). And the winners? Data and infrastructure holders.

End of March was tough and revealed what we need to work on next: market regimes We had a very similar problem back in October when daily tweets created a whipsaw. Our data has a natural 3-5 day analysis period which works amazingly well in-trend and at major reversals. Where we fall behind is 3 contradicting signals in 3 days. At the moment we treat is as "it is what it is" kind of risk. What can help though is a deeper research of market regimes: VX*name, trend strength, market breadth in options flows. On it.

Looking for 10 partners to boost each other into a couple billions of revenue then IPO and retire. DMs open (just kidding obv)

Quick hello to all "AI hedge fund" vibe coders. You forgot about data, lads. Here's how I see it (and the process we actually followed at Visual Sectors)

We called our AI agent Alfred because we don't want it to jump around in Batmobile. It's a Librarian. It has to know where the data is, be able to pull the right "book" and explain what a specific metric means. And only use our documentation for it. The nightmare with LLMs is: they can't say "I don't know". They start hallucinating, making stuff up. And I don't mean explainers, I mean making numbers up. The toughest part was to actually limit it to only returning what's in our data and docs. Otherwise it's garbage in - garbage out. People need to talk about underlying data quality more. So feel free to test it. It helps identify entry points around S/R levels, find stop loss prices - all based on human designed and performed backtests. Give it a go for free (visualsectors com)

Funny. I actually wrote this 4 months ago. SPY is up 1.14%. Breaking even today.

Interactive Brokers just published our Market Neutral strategy case study. No stone unturned interactivebrokers.com/campu…

4.9 Sharpe over the last 30 days, 78% of opened last week positions are jn the green and up 2% today. Here is how we did it: Buy 7 stocks with high conviction in the options market at their delta adj support levels Short 7 weak stocks using same logic Rebalance weekly (30 mins)

Market Neutral Strategy January results (live trading account): +4.82% return +335bps of Alpha 3.03 Sharpe ratio Very boring stuff: ✅ goes up when markets grow ✅ goes up when markets fall ✅ 30 minutes a week to rebalance Yawn. Right?:)

The beauty of Market Neutral. Our public portfolio is green on a 1.5% red market

Last February we launched our Market Neutral strategy. Very consistent and convincing performance The idea is quite simple: we Buy 7 stocks with positive options flows at Support and Short 7 stocks with negative options flow at Resistance Now we want to boost our execution of those Support/Resistance levels. Comment Support if you want your opinion/pains/ideas engineered in our new and updated tools

Price action of the last couple of months is a clear indication the trend is broken. And it's not some "cool down pullback", it's a stop to redefine the narrative. And this is exactly why market neutral strategies should be one of the tools at your disposal. There is no directional trade in the market now, so sit this one out or go market neutral

Options data can reveal hidden market signals and strengthen your trading 📈 Full episode now live: piped.video/watch?v=eliXiUgS…

There are 2 reasons for low liquidity in the financial system: trust issues and bad assets. The spread of overnight REPO/reserves rate looks scary. Pretty sure, the Fed will bail out any problems but what is the underlying story? Small banks strughling with new deposits or an overall quality of assets dropping? Credit card delinquencies are on the rise but bad debt is nowhere near the levels that can affect liquidity ratios. Can this be a hidden real estate (especially commercial) restructuring effect? Anyway, the amount of risks in the system pushed me towards Market Neutral this year and I'm surely not going all in on beta any time soon