Kelly Criterion: The Math Behind Optimal Position Sizing

📅 May 2026 | ⏱️ 6 min read | 🏷️ Risk Management, Math

What Is the Kelly Criterion?

The Kelly Criterion is a mathematical formula developed by John L. Kelly Jr. in 1956 for Bell Labs. Originally used for signal noise problems, it was quickly adopted by gamblers and investors as the optimal bet-sizing formula.

Warren Buffett uses it. Ed Thorp beat Vegas with it. We use it for both trading and value betting.

The Formula:

f* = (bp – q) / b

f* = fraction of capital to risk
b = net odds (profit per unit risked, e.g., 2.0 for 2:1 RR)
p = probability of winning
q = probability of losing (1 – p)

Real Example — FTMO Trade

Your RSI+MA20 system has a 60% win rate and a 1:2 Risk:Reward ratio.

Applying Kelly:
b = 2 (you win $2 for every $1 risked)
p = 0.60, q = 0.40

f* = (2 × 0.60 – 0.40) / 2 = (1.20 – 0.40) / 2 = 0.40 = 40%

Kelly says bet 40% of your capital per trade. But that’s aggressive — most professionals use Half Kelly (20%) or even Quarter Kelly (10%) for psychological stability.

Why We Use 1% (Not 40%)

On FTMO, pure Kelly would destroy us — not because the math is wrong, but because the rules don’t allow it. With a 5% daily loss limit, a few correlated losses at 40% Kelly would immediately breach the limit.

Our solution: use Kelly to verify we have an edge, then cap risk at 1% per trade for FTMO compliance.

Kelly Fraction Risk Level Used by
Full Kelly (40%) Extreme Theoretically optimal, practically dangerous
Half Kelly (20%) Aggressive Professional gamblers, hedge funds
Quarter Kelly (10%) Moderate Conservative investors
1% (FTMO) Conservative Our system — FTMO compliance priority

Kelly for Value Betting

The same formula works for sports betting. If a bookmaker offers odds of 2.5 (decimal) on a match you estimate has a 50% probability of occurring:

b = 1.5 (net profit per unit), p = 0.50, q = 0.50
f* = (1.5 × 0.50 – 0.50) / 1.5 = 0.25/1.5 = 16.7%

Our bet scanner uses this calculation on every detected value bet to size positions automatically.

⚠️ Kelly Criterion assumes accurate probability estimates. Overconfidence in your win rate is the #1 cause of Kelly-based blowups. Always verify with at least 100 trades before applying.

Comments

Leave a Reply

Your email address will not be published. Required fields are marked *