Kelly Criterion for Prop Firm Traders: Why Full Kelly Will Blow Your Account

Kelly criterion gives you the mathematically optimal fraction of your bankroll to bet on each trade to maximize geometric growth rate. It's real math, rigorously derived, widely cited in professional trading literature. And applied directly to prop firm accounts, full Kelly will blow them. Reliably. Because the mathematical assumptions that make Kelly optimal don't hold in the prop firm environment.

Understanding why — and what to use instead — is worth the five minutes it takes to work through it.

Kelly in One Paragraph

The Kelly formula: f = (bp - q) / b

Where:

• f = fraction of bankroll to bet
• b = net odds received per unit bet (reward-to-risk ratio)
• p = probability of winning
• q = probability of losing (1 - p)

For a trader with a 55% win rate and 1.5:1 reward-to-risk ratio: f = (1.5 × 0.55 - 0.45) / 1.5 = (0.825 - 0.45) / 1.5 = 0.375 / 1.5 = 25% of bankroll per trade.

Kelly says bet 25% of your bankroll on every trade with these parameters. If your bankroll is $3,000 (your prop account cushion), that's $750 per trade.

Why Full Kelly Destroys Prop Accounts

Kelly's optimality relies on two assumptions that don't hold for prop firm trading:

Assumption 1: Infinite time horizon. Kelly maximizes geometric growth over an infinitely long sequence of bets. In the long run, Kelly produces the highest possible growth rate — but in the short run, it produces extreme variance. A losing sequence at full Kelly can draw down the bankroll by 50, 60, even 80% before recovery. For a personal account with infinite time, that's fine — the math guarantees recovery if you have enough bankroll and enough time. For a prop account with a $3,000 cushion that closes the account if it reaches zero: the short-run variance kills you before the long run has time to work.

Assumption 2: Recoverable losses. Kelly implicitly assumes you can always increase your bet size back up when the bankroll grows. But the trailing floor mechanic means prop account losses are partially non-recoverable — the floor moved up, the cushion is permanently smaller, and there's no "growing back to your original position." You can grow above the current floor, but you can't restore the original floor distance. Kelly's math doesn't account for this.

At 25% Kelly on a $3,000 cushion, three consecutive losses reduce the cushion to approximately $3,000 × 0.75^3 = $1,266. On a trailing drawdown account where those losses raised the floor, the actual remaining cushion is even smaller. Four consecutive losses at full Kelly: $949 cushion. Five consecutive losses: $712. A 65% win rate strategy will hit 5 consecutive losses in a 200-session career approximately 2 times. Both times, full Kelly leaves you trading with 24% of your original cushion — which barely allows any position size at all.

Fractional Kelly: The Right Approach

Fractional Kelly uses a fraction of the full Kelly size — typically 25-50% of the Kelly recommendation. This reduces expected growth rate slightly while dramatically reducing variance and the probability of ruin.

At 25% Kelly (one-quarter Kelly) for the same strategy: 25% × 25% of cushion = 6.25% of cushion per trade. On a $3,000 cushion: $187.50 per trade. Five consecutive losses: $3,000 × (1-0.0625)^5 = $3,000 × 0.725 = $2,175 cushion remaining — 73% of the original, still functional.

The tradeoff: slower growth rate. At one-quarter Kelly, you're leaving significant expected return on the table. But the survival probability over a bounded evaluation period at one-quarter Kelly is substantially higher than at full Kelly. For prop firm accounts with limited cushion and non-recoverable losses, survival probability over the evaluation period matters more than maximizing expected growth rate.

The Practical Translation

Most prop firm traders don't need to calculate Kelly explicitly. The framework translates to a risk percentage guidance:

• Full Kelly at a 55% win rate / 1.5 RRR: approximately 25% of risk capital per trade — too aggressive for prop accounts
• Half Kelly: approximately 12.5% — still aggressive for trailing drawdown accounts
• Quarter Kelly: approximately 6.25% — viable, conservative
• Eighth Kelly: approximately 3% — conservative, matches the 2-3% risk percentage commonly recommended

The 1-2% risk percentage from our position sizing guide corresponds roughly to one-eighth to one-quarter Kelly for most common trading parameter ranges. This isn't a coincidence — the 1-2% recommendation was derived empirically from what keeps accounts alive through realistic variance sequences, and it roughly corresponds to the Kelly fraction that produces acceptable ruin probability over prop evaluation timescales.

When to Adjust Toward More Aggressive Kelly Fractions

One scenario where moving toward higher Kelly fractions makes sense: a static drawdown account (MFFU Core/Scale) with substantial accumulated equity above the static floor. If an MFFU trader has built $15,000 of equity above a static floor over six months, the effective cushion is $15,000 — not the original $3,000. Kelly sizing against $15,000 at 25% = $3,750 per trade is now reasonable because the cushion is genuinely large and the floor is genuinely static.

The accumulated cushion advantage of static drawdown, compounding over months of consistent trading, is one of the structural reasons MFFU's floor mechanic favors long-term equity building more than traders initially appreciate. The further you get above the static floor, the more Kelly allows — because the ruin probability from the floor decreases as the distance grows.