Position Sizing for Your Personal Account: Kelly Criterion and Practical Rules

Affiliate disclosure: TraderVerdict earns commissions from some firm links. Scores are assigned before any commercial relationship and are unaffected by affiliate status. Learn more

TraderVerdict is reader-supported. Some links in our reviews are affiliate links. We only recommend products we've personally tested.

You have an edge. Your strategy wins more than it loses, and the wins are bigger than the losses. But you're still bleeding money because you're sizing wrong. Too big on the losers, too small on the winners, or too inconsistent across trades. Position sizing kelly criterion math gives you a framework for optimal sizing, but applying it to live futures trading requires modifications that the textbooks skip.

Why Position Sizing Matters More Than Your Entry Signal

Two traders run the same strategy. Same entries, same exits, same instruments. Trader A risks a fixed dollar amount per trade regardless of setup quality. Trader B adjusts size based on edge strength and account equity. Over a hundred trades, Trader B's equity curve looks dramatically different. The entries were identical. The sizing made all the difference.

Position sizing determines how much of your edge you capture. An excellent strategy with poor sizing underperforms a mediocre strategy with optimal sizing. This is counterintuitive for most traders because the industry obsesses over entry signals and ignores the math that actually compounds capital.

For personal accounts (as distinct from prop firm accounts), position sizing is entirely your decision. No daily loss limits imposed externally. No trailing drawdown someone else defined. That freedom is dangerous because the only guardrails are the ones you build yourself. The Kelly Criterion provides a mathematical starting point for those guardrails.

The Kelly Criterion Explained for Traders

The Kelly Criterion was developed for information theory and later applied to gambling and investing. The formula tells you what fraction of your bankroll to bet on each wager to maximize long-term growth rate. For trading, it translates to: what percentage of your account equity should you risk on each trade?

The basic formula: Kelly % = W - (L / R), where W is your win rate as a decimal, L is your loss rate (1 - W), and R is your average win divided by your average loss (the payoff ratio).

A practical example: your strategy wins 55% of the time and your average win is 1.5 times your average loss. Kelly % = 0.55 - (0.45 / 1.5) = 0.55 - 0.30 = 0.25, or 25% of your account per trade.

If that number seems enormous, it should. Full Kelly sizing is extremely aggressive. It maximizes the long-term growth rate but produces massive drawdowns along the way. The equity curve at full Kelly is a roller coaster that most traders can't psychologically survive, even if the math works out over the long run.

The important insight from Kelly isn't the specific number. It's the concept that optimal sizing exists and it's a function of your edge (win rate and payoff ratio). Size too small, and you leave money on the table. Size too large, and you risk ruin even with a positive edge. Kelly tells you where that sweet spot is in theory.

Why Full Kelly Is Dangerous for Live Trading

The Kelly Criterion assumes you know your exact edge. In trading, you don't. Your win rate and payoff ratio are estimates based on historical data. Markets change. Your recent sample might overestimate or underestimate your true edge. Full Kelly sizing based on an overestimated edge leads to oversizing, and oversizing with uncertain edge leads to account destruction.

The formula also assumes infinite time horizon and no psychological constraints. A full Kelly bettor will experience drawdowns of 50% or more regularly. Mathematically, they recover. Psychologically, most traders abandon their strategy long before the recovery happens. They cut size after the drawdown (the worst time to cut) and add size after a winning streak (the worst time to add). This behavior turns optimal sizing into a losing strategy in practice.

Full Kelly also ignores the correlation between trades. If your losses tend to cluster (which they do in trading, because losing streaks often coincide with unfavorable market regimes), the drawdowns are deeper and longer than the formula predicts. The independence assumption that Kelly requires is violated in real markets.

For these reasons, no serious trader or fund uses full Kelly. The concept is the foundation. The application requires significant modification.

Fractional Kelly: The Practical Approach

The standard modification is fractional Kelly. Instead of risking the full Kelly percentage, you risk a fraction of it. Half Kelly is the most common starting point. Quarter Kelly is common for traders with shorter track records or higher uncertainty about their edge.

Using the earlier example: full Kelly was 25%. Half Kelly = 12.5%. Quarter Kelly = 6.25%. These are still aggressive by most standards, but they dramatically reduce the drawdown severity while only modestly reducing the growth rate.

The math behind this tradeoff is powerful. Half Kelly produces roughly 75% of the growth rate of full Kelly but cuts maximum drawdown roughly in half. That's an excellent tradeoff for anyone who values sleep. Quarter Kelly produces roughly 50-60% of the growth rate while cutting drawdowns further. You're sacrificing some compound growth for significantly better survivability.

For futures traders on personal accounts, we generally recommend starting at quarter Kelly or even lower until you have at least a hundred trades of live data confirming your edge estimates. The confidence in your edge estimate matters as much as the estimate itself. A trader with a confirmed 55% win rate over three hundred live trades can justify higher Kelly fractions than a trader with the same win rate over thirty trades.

The Kelly vs Fixed Percentage Debate

This is the advanced-reader debate that divides quantitative traders. Should you use Kelly-based dynamic sizing or a simple fixed percentage of equity per trade?

The Kelly camp argues that fixed percentage sizing ignores edge quality. Risking 2% per trade whether you have a 60% win rate or a 52% win rate means you're oversizing the weak-edge trades and undersizing the strong-edge trades. Kelly adjusts to the edge, which is mathematically optimal.

The fixed percentage camp argues that edge estimates are too noisy in practice to make Kelly adjustments meaningful. The estimation error in your win rate and payoff ratio swamps the theoretical benefit of dynamic sizing. A simple 1-2% per trade, applied consistently, produces acceptable results without the complexity and estimation risk of Kelly-based sizing.

There's also a hybrid argument: use fixed percentage as your baseline but adjust size based on setup quality using a simplified Kelly-inspired framework. A-grade setups (highest confidence, best risk-reward) get full size. B-grade setups get half size. C-grade setups get skipped. This captures some of the Kelly intuition without requiring precise edge estimates for each setup type.

Our view: for most personal account traders, the hybrid approach is the best balance of simplicity and optimization. Pure Kelly requires data collection and statistical analysis that most discretionary traders won't maintain. Pure fixed percentage leaves edge-quality information on the table. The tiered approach captures the most important insight (size up when conditions favor you) without demanding precision that doesn't exist.

Practical Position Sizing Rules for Personal Futures Accounts

Here's what we actually do on our personal accounts, informed by Kelly but adapted for the realities of discretionary futures trading.

Base risk per trade: we start with a percentage of account equity that falls within our estimated quarter-to-half Kelly range. For most strategies with moderate edges, this works out to a few percent of equity per trade. The exact number depends on your specific win rate and payoff ratio. Calculate it from your journal data.

Tiered sizing by setup quality: our highest-conviction setups (strong confluence, ideal location, order flow confirmation) get our full base size. Standard setups get roughly half that. We don't take setups below that threshold.

Equity-adjusted sizing: we recalculate our base risk amount regularly as account equity changes. After a drawdown, the dollar risk per trade naturally decreases because the percentage is applied to a smaller equity base. After a run-up, it increases. This Kelly-inspired feature prevents catastrophic losses during drawdowns and scales up appropriately during good periods.

Maximum daily risk: regardless of individual trade sizing, we cap our total daily risk exposure. If we've lost our maximum daily amount, we stop. This isn't Kelly. It's a practical circuit breaker that prevents the emotional escalation that turns a bad day into an account-ending day.

We track every trade in a journal and recalculate our edge estimates quarterly. If the numbers shift, the sizing shifts. Position sizing kelly criterion math only works if you feed it real data, not assumptions.

How This Differs from Prop Firm Sizing

On prop firm accounts, the firm sets your constraints. Daily loss limits, trailing drawdowns, and consistency rules define your sizing ceiling. You're optimizing within someone else's parameters.

On a personal account, you set all the parameters. That's more freedom and more responsibility. The Kelly framework helps you think about sizing as an optimization problem rather than a gut feel decision. Even if you never calculate a formal Kelly percentage, understanding the concepts changes how you think about risk.

The core lesson: position sizing isn't about how much you can afford to lose on a single trade. It's about how much you should risk to maximize long-term equity growth while keeping drawdowns within your psychological and financial tolerance. Kelly gives you the math. Your trading journal gives you the inputs. The rest is discipline.