# Don't count strategies, *count independent bets* Author: Jiří Fabšic · Jul 14, 2026 · Category: Methodology Canonical URL: https://www.binaryfintech.com/en/blog/how-many-strategies/ > Diversification has a ceiling: each added strategy smooths less (√N) and in a crisis correlations rise toward +1. Where you stop competing with the market and start competing with yourself — and why to count independent bets, not strategies. ## TL;DR - **Diversification has a ceiling.** Each additional uncorrelated strategy smooths less — volatility falls with the *square root* of the count (√N), not the count — while the cost (capital, liquidity, attention) rises. The optimum is not a number; it is the point where the next strategy stops paying its rent. - **There are always fewer independent bets than strategies.** Strategies share hidden risk factors, and in a crisis their correlation rises toward +1. What matters is the *number of genuinely independent bets*, not how many you count on paper. - **Past the ceiling you compete with yourself.** The twentieth strategy eats the capital and liquidity of your best one. So for each new strategy, ask: does it bring a new source of return, or is it just one more mouth at the same trough? Adding one more strategy to a portfolio looks like free money. Always. The equity is smoother, you sleep better — and as we showed in [running strategies concurrently](https://www.binaryfintech.com/en/blog/strategy-portfolios/), uncorrelated sources of return genuinely reduce volatility without discounting expected return. So why not add forever? Because diversification has a ceiling. And past it the whole game quietly flips: **you stop competing with the market and start competing with yourself.** ## Does diversification have a ceiling? It does, held in place by two mechanisms. The first is pure arithmetic: the volatility of an equally weighted portfolio of *N* uncorrelated strategies falls with the **square root** of *N*, not with *N* itself. The first strategy you add cuts the most volatility, the second a little less, the tenth almost nothing. The benefit crumbles — the first bet is a punch, the twentieth is a rounding error. In equities this is long mapped out: after roughly twenty to thirty well-chosen names, the further reduction in unsystematic risk is negligible. But strategies are not stocks in an index. They share **one** pool of money and fight over the same liquidity, so their ceiling arrives differently — and usually sooner. ## How many independent bets do you actually have? The second mechanism is sneakier. Counting twenty strategies does not mean you hold twenty independent bets. Strategies share hidden risk factors — five trend logics on five markets are still mostly one bet on the idea that markets trend. Real diversification is not measured by the number of rows in a table but by the **number of genuinely uncorrelated bets**, and that number tends to be far lower. On top of that, correlation is not a constant: low in calm markets, it rises toward +1 in a panic — exactly as [concurrency](https://www.binaryfintech.com/en/blog/strategy-portfolios/) and [black swans](https://www.binaryfintech.com/en/blog/black-swans-fat-tails/) warn. Right when you need diversification most, it evaporates in your hands. So the first question is not “how many strategies do I have?” but “how many of them are genuinely independent?” ## When you start competing with yourself This is where the game flips. As long as you add independent sources of return, you play against the market and win — the portfolio is calmer than any of its parts. But every additional strategy has a second side too, and it grows with the count. It sits on that same pool of money: the twentieth strategy locks capital your best one could have put to work, because [capital is one pool](https://www.binaryfintech.com/en/blog/strategy-portfolios/). It fights for the same liquidity — when several of your strategies want into the same market at the same moment, they [push the price against each other](https://www.binaryfintech.com/en/blog/order-execution-mechanics/). And the more positions you hold, the thinner the ration your strongest idea gets; Peter Lynch has a word for it, **diworsification** — you keep adding until you dilute the very thing that was meant to earn. Overhead rises with it: more code, more monitoring, more things to break at the worst possible moment. The opponent has quietly moved from the market to you. Fat Tony, from our article on [black swans](https://www.binaryfintech.com/en/blog/black-swans-fat-tails/), wouldn't ask how much your twentieth strategy *adds*. He'd ask what it **eats**. *Figure: A schematic, not data: the marginal benefit of each further bet falls (√N), the marginal cost — capital, liquidity, attention — rises. Where they cross is the ceiling; past it, adding only dilutes yourself.* ## Don't count strategies, count independent bets So the optimum is not a number I could write into a table — it is different for stocks than for strategies, and different for your portfolio than for your neighbour's. It is the point where the **marginal benefit** of another independent bet (falling) meets the **marginal cost** of capital, liquidity and attention (rising). The practical rule is simple: for each new strategy, ask one question — does it bring a new, genuinely uncorrelated source of return, or is it just one more mouth at the same trough? If the latter, you are not improving the portfolio, only diluting it. And because you can't spot the ceiling from your desk, you have to see it in the data. That is exactly why [BXF](https://bxf.binaryfintech.com) tests a portfolio as a single whole and surfaces what is [worth watching](https://www.binaryfintech.com/en/blog/strategy-portfolios/) in a concurrency — correlation of returns, exposure overlap over time, portfolio drawdown against the drawdowns of its parts, and risk concentration. The moment an added strategy raises risk concentration without lowering portfolio drawdown, you have just crossed the ceiling. Because the best portfolio is not the one with the most strategies. It is the one where **every bet still pays its rent.** **Reading:** Related articles: [Strategy portfolios: one pool of money](https://www.binaryfintech.com/en/blog/strategy-portfolios/) · [Overfitting](https://www.binaryfintech.com/en/blog/overfitting/) · [Black swans and fat tails](https://www.binaryfintech.com/en/blog/black-swans-fat-tails/) · [Order execution mechanics](https://www.binaryfintech.com/en/blog/order-execution-mechanics/).