Home › Stories › Legends

Jim Simons: the codebreaker who let the data decide

Jim Simons started out cracking codes for the US government and won a top prize in geometry. Then he spent decades trying to find patterns in market prices. The fund he built, Medallion, has a record that most experts call the best ever reported.

Story · 7-minute read · Updated 2026-10-21

Who they were

Jim Simons was born in 1938 near Boston. He studied math at MIT and earned a PhD from the University of California, Berkeley, at 23.

In the 1960s he worked as a codebreaker at the Institute for Defense Analyses, a research group that worked for the US government. He was let go in 1968 after he publicly spoke out against the Vietnam War.

He then led the math department at Stony Brook University in New York. In 1974, with the mathematician Shiing-Shen Chern, he published work in geometry now known as Chern–Simons theory. It is still used in physics today.

What happened

From math to markets

In 1978, Simons left the university to trade. He started a firm that became Renaissance Technologies in 1982.

At first he traded partly on instinct and judgment, like most people. He found it stressful. Some days he was sure, and wrong. Over time he pushed the firm toward something different: rules found in data, run by computers.

Hiring scientists, not traders

Renaissance became famous for whom it hired. Mathematicians, physicists, astronomers, computer scientists. Few came from Wall Street.

Their job was to look at huge amounts of price data and find small, repeating patterns. Then they tested each pattern carefully before trading it.

The Medallion Fund

The firm's main fund, Medallion, started in 1988. In The Man Who Solved the Market (2019), journalist Gregory Zuckerman reported that from 1988 to 2018 it averaged about 66% a year before fees, and about 39% after fees.

Those fees were very high. And since the mid-2000s, the fund has been open only to the firm's own employees. Outsiders could not buy in.

Simons stepped down as CEO in 2009. He spent much of his later life on science and education through the Simons Foundation. He died in May 2024, at 86.

The moment everyone remembers

The famous part is the number. Nobody else has reported a record like Medallion's for so long.

But the idea traders study most is how small each edge was. In Zuckerman's book, Renaissance executive Robert Mercer is quoted as saying the firm was right only about 50.75% of the time.

That's barely better than a coin flip. The point: a tiny edge, repeated over a huge number of trades, with every loss kept small, can add up. No single trade matters much.

What it teaches

1. Win rate isn't everything

Many beginners chase a high Win rate. Renaissance shows why that's the wrong focus. What counts is the average result across many trades, called Expectancy.

Here's a math example in units of risk. Say you risk 1 unit per trade, win 45% of the time, and wins average 2 units:

• 45 wins × 2 units = 90 units
• 55 losses × 1 unit = 55 units
• Over 100 trades: about +35 units

That's a math example, not a promise. Real results include losing streaks, costs and mistakes. Try your own numbers in the expectancy calculator.

2. Test before you trust

Renaissance tested ideas on past data before risking money. You can do a simple version for free: practice your rules on past charts with replay, and never peek ahead. See how to backtest a trading strategy.

3. Write it down, then look at the record

Data only helps if you collect it. For a beginner, that's a trading journal: every trade, the setup, the result in units of risk, and how you felt. After 20 or 50 trades, patterns show up that you'd never notice day to day.

4. Rules protect you from your own mood

Simons moved away from gut calls because they were stressful and hard to trust. A written plan and a pre-trade checklist do the same job on a small scale: the rules decide, not the feeling of the moment.

What it doesn't mean

It doesn't mean anyone can copy Medallion. Renaissance had hundreds of PhDs, enormous computing power, decades of cleaned data and very fast trading. Its methods are secret. No book, indicator or course can hand that to you, and anyone who claims to is selling something.

It doesn't mean computer trading always works. Many quant funds have had bad years, and some have closed. Even Renaissance's funds open to outside investors had much more ordinary results than Medallion.

And a "51% edge" is not something a beginner can measure on a few trades. It took Renaissance millions of trades to trust its edges. What carries over is the habit: test, record, review, and keep every loss small.

Timeline

Questions