Kelly Criterion Calculator

A kelly criterion calculator turns two numbers you already track, your win rate and your win to loss ratio, into a suggested position size, meaning the fraction of your account to risk on a single trade. The Kelly criterion is a formula from 1956 that finds the bet size that grows a bankroll the fastest over the long run when you truly know your edge. The calculator above applies that formula to US markets, so you can see what full Kelly, half Kelly, and quarter Kelly would suggest for a trade in AAPL stock, an SPY option, or an ES futures contract. This page explains every input and output, shows the math with real numbers, and stresses why almost no disciplined trader ever risks the full Kelly amount.

What the Kelly Criterion Is

The Kelly criterion answers one question: given a repeatable bet where you know your chance of winning and how much you win versus lose, what fraction of your money should you stake each time to grow the fastest without going broke? It was published by John Kelly, a scientist at Bell Labs, and was later adopted by gamblers and then by investors. In trading terms, one bet is one trade with a defined risk. If you buy 100 shares of AAPL at 230 and set a stop at 220, your risk per share is 10 dollars and your loss is defined. Kelly then tells you how large that defined-risk position should be relative to your whole account.

The key word is edge. Edge means a genuine, tested advantage where your wins and win rate are large enough to beat your losses over many trades. Kelly only makes sense when you have a real edge measured from a real sample of trades. If your numbers come from a hunch or from ten lucky trades, the calculator will still produce a confident-looking percentage, and that false confidence is where accounts get hurt.

How the Calculator Works: Inputs and Outputs

The calculator above takes two inputs and returns four outputs. Each is explained below so you know exactly what you are typing and reading.

The Two Inputs

• Win Rate (percent): the share of your trades that end in a profit. If 55 of your last 100 closed trades made money, your win rate is 55 percent. The calculator converts this to the decimal W = 0.55 inside the formula.
• Win to Loss Ratio (R): your average winning trade divided by your average losing trade, both measured in dollars or in R multiples. If your average win is 300 dollars and your average loss is 200 dollars, your ratio is 1.5. This is a payoff ratio, not your win rate, and the two are easy to mix up.

The Four Outputs

• Full Kelly: the raw fraction the formula produces. If it shows 25 percent, full Kelly says risk 25 percent of your account on one trade, which is extreme for real trading.
• Half Kelly: full Kelly divided by two. This is the most common practical setting because it keeps roughly three quarters of the growth with far smaller account swings.
• Quarter Kelly: full Kelly divided by four, a conservative choice for uncertain or new strategies where your win rate estimate could easily be wrong.
• Assessment: a plain-language read of the result, such as a strong edge, a thin edge, or no edge (a negative number), telling you whether the trade is even worth sizing.

The Formula and Mechanics

The core formula is f = W - (1 - W) / R. Here f is the fraction of your account to risk, W is your win rate as a decimal, and R is your win to loss ratio. Read it in plain English: start with your win rate, then subtract your loss rate scaled by how small your losses are relative to your wins. When your payoff ratio R is large, the term you subtract shrinks, so Kelly grows. When your win rate is low or your losses are as big as your wins, the subtracted term dominates and Kelly can turn negative.

Three Worked Examples With Real Numbers

Example 1: A Swing Trader in AAPL

Suppose you swing trade AAPL stock and, over 100 closed trades, you won 55 (W = 0.55) with an average win of 300 dollars and an average loss of 200 dollars, so R = 1.5. Full Kelly = 0.55 - (1 - 0.55) / 1.5 = 0.55 - 0.45 / 1.5 = 0.55 - 0.30 = 0.25, or 25 percent. Half Kelly is 12.5 percent and quarter Kelly is 6.25 percent. On a 40,000 dollar account, quarter Kelly means risking about 2,500 dollars per trade, which for a 10 dollar stop is 250 shares. Full Kelly would risk 10,000 dollars on one trade, an amount that could halve your account after just a few bad trades in a row.

Example 2: An SPY Options Buyer

Now say you buy SPY call options, a directional bet where the whole premium can go to zero. Long options tend to have a lower win rate but a bigger payoff. Suppose W = 0.40 and your average win is three times your average loss, so R = 3. Full Kelly = 0.40 - (1 - 0.40) / 3 = 0.40 - 0.60 / 3 = 0.40 - 0.20 = 0.20, or 20 percent. Half Kelly is 10 percent and quarter Kelly is 5 percent. Because an option can expire worthless and lose 100 percent of the premium, most disciplined option buyers stay at quarter Kelly or lower, so a wrong win-rate guess does not blow up the account.

Example 3: An ES Futures Scalper With No Edge

Imagine you scalp the ES (E-mini S&P 500 futures) and your honest numbers are W = 0.45 with an average win equal to your average loss, so R = 1. Full Kelly = 0.45 - (1 - 0.45) / 1 = 0.45 - 0.55 = -0.10, or negative 10 percent. The Assessment reads no edge. Kelly is telling you that at these odds you lose over time, and the only correct size is zero. Futures are leveraged and can lose more than your deposit, so forcing a trade here is exactly how disciplined accounts get wrecked.

Kelly Fraction Reference Table

Notice the 50 percent win rate with R = 1 row: Kelly is exactly zero. Winning half the time while your wins and losses are the same size is a break-even system before costs, and after US commissions and slippage it loses. This is why win rate alone tells you nothing without the payoff ratio.

Why Full Kelly Is Too Aggressive for Trading

Full Kelly is optimal only if your win rate and payoff ratio are exactly correct and never change. In markets, neither is true. Your edge is estimated from a limited sample, market conditions shift, and a good strategy can hit a losing streak of eight or ten trades by pure chance. At full Kelly, a normal losing streak can cut your account in half, and recovering from a 50 percent drawdown requires a 100 percent gain. This is why professionals use fractional Kelly.

1. Measure your real win rate and payoff ratio from at least 50 to 100 closed trades, not a handful.
2. Enter those into the calculator above and read the Full Kelly output.
3. Divide it yourself to be conservative, or read the Half Kelly and Quarter Kelly the calculator already shows.
4. Choose Half Kelly for a well-tested strategy or Quarter Kelly for anything new or uncertain.
5. Cap the result at a sane ceiling, such as 1 to 2 percent risk per trade, if Kelly suggests more than you can stomach.
6. Recompute every month as new trades update your true win rate and ratio.

Half Kelly keeps roughly 75 percent of full Kelly's long-run growth while cutting the size of your account swings by about half. Quarter Kelly gives up a little more growth for much smoother equity. For most traders, smoother equity is what keeps them following the plan instead of panicking, so the slightly slower theoretical growth is worth it.

Tips for Using Kelly Well

• Feed the calculator honest, sampled numbers, not optimistic guesses. Overestimating win rate by even 5 points can double your suggested size.
• Treat the output as a ceiling, not a target. It is safer to risk less than Kelly than more.
• Recalculate as your data grows. Early win rates are noisy and often too high.
• Never scale up after a hot streak by raising your win-rate input to match recent luck.
• Remember costs. US commissions, bid-ask spread, and slippage lower your real payoff ratio below the raw price numbers.
• Use quarter Kelly or a fixed 1 percent rule for leveraged products like ES or NQ futures, where losses can exceed your margin.

Common Mistakes to Avoid

• Confusing win rate with win to loss ratio. Win rate is how often you win; the ratio is how much you win versus lose. They are different inputs.
• Using full Kelly in live trading. This is the fastest route to a giant drawdown and is almost never used by professionals.
• Ignoring a negative result. A negative Kelly is a clear signal that the system loses, not an invitation to trade smaller.
• Sizing off a tiny sample. Ten trades cannot reveal your true edge, and Kelly on ten trades is a guess dressed up as math.
• Forgetting that options can expire worthless and naked options carry large or even unlimited risk, which Kelly's simple two-outcome model does not capture.

How Kelly Supports Disciplined Journaling

Kelly is only as good as the win rate and payoff ratio you feed it, and those numbers come from journaling. When you log every trade with its entry, exit, and result, you build the honest sample the formula needs. A trading journal turns Kelly from a one-time guess into a living measurement that updates as your real edge changes. It also protects you from the biggest danger, which is overestimating your win rate, because your journal shows the truth instead of your memory of the good trades. Discipline first means sizing from measured reality, taking no trade when Kelly is negative, and staying at half or quarter Kelly even when a streak tempts you to push.

The Kelly criterion rewards traders who know their real numbers and punishes those who guess. The most reliable way to know your true win rate and payoff ratio is to log every trade honestly, then let those figures drive your sizing. Start journaling your US stock, options, and futures trades on OneTradeJournal, watch your real edge take shape over time, and let the calculator above size from facts instead of hope.