Skip to content
← The Scroll
Canonical · Paper

The Kelly Criterion & Position Sizing

John L. Kelly / Edward Thorp · 1956

"Kelly derived the fraction of bankroll to bet to maximize long-run logarithmic growth — bet too little and you compound slowly, bet too much and a single bad run wipes out the geometric advantage even with a positive edge."

The idea

Kelly asked Bell Labs' question in 1956: if you have an edge, what fraction should you actually bet? The answer — edge divided by odds — is the fastest sustainable compounding path, and most investors bet far past it.

Why it works

With a 60/40 edge and even payoff, Kelly says bet 20% of capital; bet 40% and you grow slower, bet 100% and a short streak ruins you despite the edge. The formula maximizes log wealth, which is geometric growth, not expected value alone. Practitioners use half-Kelly because estimates are noisy — volatility is the tax on misestimating your own edge, so position size is inseparable from uncertainty about the edge.

The takeaway — recall it first
Check your understanding

Why does the Kelly Criterion recommend betting less than full expected-value maximization would suggest?

Further reading

Read more about the topic

The explanation above is written with AI assistance. These are the originals — go to them to check it.

  • A New Interpretation of Information Rate — the Kelly Criterion (1956)J. L. Kelly / Wikipedia
Up NextSuggested: Continues the theme of Finance Theory

The Capital Asset Pricing Model (CAPM)

"CAPM isolates the price of systematic risk — expected return rises linearly with beta to the market, so only non-diversifiable risk earns a premium and alpha is excess over that line."

William Sharpe · PaperContinue→
Listen
0 / 5