#夏日创作营 To understand the logic of how Simons makes money, remember one line: he never predicts the future—he only looks for mathematically “statistical certainty.”

Unlike Buffett’s “buy and hold,” Simons’ profit model can be broken down into these four steps:

· Step 1: Ditch intuition—only trust data. The people he hires are not finance elites, but top mathematicians, physicists, and cryptographers. Their job isn’t to study company financial statements, but to mine short-lived patterns that are statistically significant and repeatable from decades of historical data (such as prices, trading volume, weather, and more).
· Step 2: Catch the moment of “mispricing.” The core of the Renaissance Technologies’ flagship strategy is short-term statistical arbitrage, with an average holding period of only 2 days. The model continuously scans thousands of stocks; once it finds that the price relationship between them deviates slightly (for example, A-shares and B-shares have long moved together up and down, but today suddenly diverge), it immediately executes a basket of buy-and-sell trades, betting that the deviation will snap back to normal like a rubber band. The money comes from the return-to-fair-value of the pricing gap, not from caring whether the stocks themselves are good or bad.
· Step 3: Use leverage to amplify tiny profits. The spread in this arbitrage is usually extremely small (possibly only a few cents), but the win rate is very high (reportedly over 50%). Simons uses high leverage to scale up the trade size—multiplying a few cents of profit by massive capital—then repeats this kind of trade millions of times, turning small wins into big wins. It’s like a casino that relies on a rule where the “win rate is only slightly above 50%.” As long as you play enough, the law of large numbers will stand on your side.
· Step 4: A stringent risk-control “capital-cutting machine.” Once the model detects that losses on open positions exceed a preset threshold (for example, 2%), the computer automatically and unconditionally liquidates the positions, with no room for human intervention. This eliminates emotional interference, ensuring losses are strictly capped, while profits keep the model running.

---

So, to sum it up, Simons’ ultimate secret is: use math to find the market’s short-term “inefficiencies,” and use the law of large numbers from high-frequency trading to make money.

This completely bypasses the logic of traditional value investing. But it also leads to a consequence: the amount of capital this strategy can absorb is limited. Once the scale becomes too large, the model ends up eating its own profits. That’s also why the Renaissance Technologies fund closed the channel for external investors as early as 1993 and traded using only its own money—one of the key reasons its performance could keep the myth alive.