Kelly Criterion Calculator for Stock Investors
How much of your portfolio one idea deserves: full, half and quarter Kelly.
How often theses like this one work out for you
Full Kelly: the growth-maximizing position size
25.0%
A full Kelly this large tolerates brutal drawdowns and assumes perfect estimates. Most investors run half or quarter Kelly, because estimation error compounds faster than returns.
- Half Kelly (common in practice)
- 12.5% · $12,500
- Quarter Kelly (conservative)
- 6.3% · $6,250
- Payoff ratio (R)
- 1.50
How to size a position with Kelly
- Estimate your win probability honestly: how often have theses like this one worked for you?
- Enter the average gain when you're right and the average loss when you're wrong.
- Read full Kelly as a ceiling rather than a target. It assumes your estimates are exact.
- Size at half or quarter Kelly; add your portfolio value to see the dollar amounts.
- A negative result means no position. The formula never says "just size it smaller."
How this was calculated
f* = W − (1 − W) / R
W is the probability your thesis works out and R is the payoff ratio: average gain when right divided by average loss when wrong. f* is the fraction of capital that maximizes long-run compound growth for those inputs. Half and quarter Kelly scale it down, because with estimated rather than known probabilities, over-betting costs more than under-betting, so most practitioners size fractionally. A negative f* means the edge is negative and the correct position is zero. These are educational estimates, not investment advice.
How does your stock score?
MonkScore™ distills 149 fundamental ratios into one 0–100 score across five pillars. The scores live inside MonkStreet.
- Growth (value available with a MonkStreet trial)
- Profitability (value available with a MonkStreet trial)
- Quality (value available with a MonkStreet trial)
- Conviction (value available with a MonkStreet trial)
- Safety (value available with a MonkStreet trial)
Frequently asked questions
The Kelly Criterion, from John L. Kelly's 1956 paper, computes the fraction of your capital that maximizes long-run compound growth when you have an edge: f* = W − (1 − W) / R, where W is your probability of being right and R is the ratio of average gain to average loss. Bet more when your edge or payoff is bigger; bet nothing without an edge.
Yes. It originated in information theory and was popularized in gambling, but investors including Ed Thorp applied it to markets. For stocks, W is the probability your thesis plays out and R the ratio of expected upside to downside. Because those inputs are estimates rather than known odds, investors almost always use fractional Kelly.
Full Kelly maximizes growth but the drawdowns are violent (a 50% loss of capital is routine), and it assumes your probability estimates are exactly right. Half Kelly delivers about 75% of the growth with half the volatility, which is why most practitioners size at half or quarter Kelly. Overestimating your edge at full Kelly means systematic over-betting.
It means the numbers you entered describe a losing proposition: your win probability is too low for the payoff on offer. The formula's answer is to take no position at all, not a smaller one.
In sports betting you can compare your estimate against bookmaker odds; in investing there is no posted line, so W comes from your own base rates: how often theses like this one have worked for you historically. Track your decisions over time and be conservative. Uncertain probabilities argue for fractional Kelly rather than precision you don't have.
No. It optimizes how much to risk given an edge; it cannot create one. With honest inputs it maximizes long-run growth and mathematically avoids total ruin (you never stake everything), but if your true edge is zero or negative, Kelly sizing just controls how fast you lose.
Fixed-percentage sizing (say 5% per position) ignores how attractive each opportunity is; Kelly scales the position to your edge and payoff, so a high-conviction, asymmetric idea gets more capital than a marginal one. The cost is sensitivity to estimation error, which fractional Kelly mitigates.
The MonkStreet letter
One weekly email on what the fundamentals say, without the hype.
Subscribe freeData updated: July 2026