A trader who has been consistently profitable in a $5,000 personal account signs up for a $50,000 prop evaluation. They use the same strategy. They use the same setup criteria. They blow the account in 8 days.
What happened? Their personal $5,000 account was hot-trading 1 MNQ contract on each setup. Risk per trade: roughly $50-$80. Approximately 1-1.5% of account.
Same trader, on the $50K prop account, scales up to 3-5 MNQ contracts because “the account is bigger.” Risk per trade: $250-$400. Approximately 0.5-0.8% of account… but 10-20% of remaining drawdown buffer.
That sentence is the entire article: prop firm accounts are sized by buffer, not by equity. Almost no trader does this correctly. Almost everyone blows accounts because of it.
The Variable Most People Use Wrong
When retail trading literature talks about position sizing, the standard formula is:
Risk per trade = X% of account equity, where X is typically 1-2%.
On a $50,000 account, 1% is $500. So you can risk $500 per trade. Easy.
This formula is wrong for prop firm accounts. Here’s why.
On your personal $50K account, your “drawdown limit” is implicit. You decide it. If you lose $5,000 you might stop trading and reassess. If you lose $10,000 you definitely will. There’s no hard floor — just discomfort thresholds.
On a prop firm $50K account with a $2,500 trailing drawdown, the drawdown is explicit and binding. You hit the floor and the account is dead. Forever. No reassessment. No “I’ll trade smaller for a while.” Dead.
If your buffer is $2,500 and you’re risking $500 per trade (1% of equity), you survive 5 consecutive maximum-loss outcomes. With a 55% win rate strategy, the probability of 5 consecutive losses is (0.45)^5 = 1.8%. Sounds low. Across 100 trades, the probability of any 5-loss streak is roughly 30%. You’re not going to make it 100 trades. You’re going to make it maybe 20-40 trades before encountering a 4-5 loss streak and dying.
The correct framing is not “what percent of equity do I risk per trade” but “how many consecutive losing trades does my position size let me survive.”
The Buffer-Based Sizing Framework
Here is the framework that survives prop firm math:
Step 1: Identify your true buffer. Current buffer = current balance − current floor (where floor accounts for trailing). On a fresh $50K account with $2,500 trailing drawdown, buffer is $2,500.
Step 2: Decide how many consecutive losses you need to survive. For a strategy with 55% win rate, you should plan to survive 8-10 consecutive losses. (Probability of 8 straight losses at 45% loss rate is 0.45^8 = 0.17%, but across 200 trades the probability of an 8-loss streak somewhere is roughly 30%. Across a multi-month account, you will see one.)
Step 3: Compute max risk per trade. Max risk = Buffer / (consecutive losses to survive). On $2,500 buffer with 8-loss survival, max risk per trade is $312.
Step 4: Compute contract count. Contract count = floor(Max risk / (stop loss in ticks × tick value)). For MNQ at $0.50/tick with a 30-tick stop, that’s $15 per contract. $312 / $15 = 20 contracts max. But you’ll never trade 20 MNQ. Other constraints (firm contract limits, position size rules) will cap you. Let’s say firm cap is 10. Use min(20, 10) = 10.
Step 5: Reduce by confidence factor. This is where most traders go off the rails. Even on your A+ setup, do not size up to your max. The max is the cap; your default is 50-70% of max. Reserve the full max for situations where you have multiple confirmations, clean market structure, and tight risk.
For a default trade, you’d trade 5-7 MNQ in this example, risking $75-$105 per trade. On 1% of $50K equity, that’s well under the “standard” 1%. The buffer math forces you to trade smaller than retail conventions would suggest.
R-Multiple Sanity Check
Most prop traders compute R as “how much I make divided by how much I risked.” A 1.5R winner means you made $1.50 for every $1 you risked. That’s fine for evaluating individual trades.
But for evaluating whether your strategy survives at all, the relevant question is: across a sequence of trades, does my expected return exceed my variance?
The simplified math:
- Expected R per trade = (win rate × avg winner R) − (loss rate × avg loser R)
- For a 55% win rate, 1.5R avg winner, 1R avg loser: 0.55 × 1.5 − 0.45 × 1 = 0.825 − 0.45 = +0.375R per trade
Sounds good. But variance per trade is roughly the standard deviation of outcomes — which for that distribution is around 1.2R. Your edge is 0.375R per trade against 1.2R per-trade noise. You need a lot of trades for the edge to dominate noise.
Statistically, you need approximately (variance / edge)² × small factor ≈ (1.2/0.375)² × 4 ≈ 40-50 trades for the realized P&L to be reliably positive with 80%+ confidence.
Now go back to your buffer. If you’re risking 1R = $500 per trade and have $2,500 buffer, you can survive maybe 5R of drawdown before death. Your strategy needs to NOT lose 5R in any 50-trade sequence. With variance of 1.2R per trade, that’s not guaranteed — and you usually only get to find out by failing.
The fix: size such that 1R = a much smaller fraction of buffer. If 1R = $150 (0.3% of equity, 6% of buffer), then 5R of drawdown costs $750, leaving $1,750 buffer. You can survive a brutal stretch and recover.
The framework requires sizing dramatically smaller than retail trading literature recommends. That’s correct. Retail literature is sized for unlimited drawdown patience. Prop firm accounts have explicit, lethal drawdown limits.
The “But I Need to Hit the Target” Counter-Argument
The most common objection: “If I size that small I’ll never hit the $3,000 profit target.”
Math check. At 1R = $150 and an average +0.375R per trade, you make $56 per trade in expected value. The $3,000 target requires 53 expected-value trades. At 4-6 trades per day, that’s 9-13 trading days. With variance, real range is 12-25 trading days.
That is the actual time it takes to pass an evaluation if you’re trading the math correctly. Most evaluations have no time limit, so this is fine. If your firm has a 30-day evaluation window, you’re still well within it.
The traders who blow accounts in 3-5 days are trying to compress 12-25 days of expected-value work into 3-5 days of higher-variance work. They’re not failing because the strategy doesn’t work. They’re failing because the sizing math turns their 0.375R edge into a 4-7 loss streak survival problem.
The Tactical Daily Routine
For traders willing to actually apply this:
Pre-session (5 minutes before open):
- Check current buffer: Current balance − current floor.
- Compute max risk per trade: Buffer × 0.10 (so 10% of buffer per trade for safe sizing).
- Compute max contracts: Max risk / (your stop in ticks × tick value).
- Default contracts = 50-70% of max. Reserve max for A+ confluence only.
- Compute daily stop: Max risk × 2.5. If you hit this on the day, you’re done.
Intraday:
- Before entering each trade, re-confirm contract count based on current buffer. If you’ve taken a loss, buffer is smaller, max contracts are smaller.
- If unrealized equity touches starting balance + $1,200 in any session, walk away — buffer is ratcheting nicely, don’t give it back.
Post-session:
- Log session P&L, intraday peak, current floor.
- Update tomorrow’s max risk based on new buffer.
This is the actual job. Not “find the perfect setup” or “develop conviction.” The setup is whatever you already know how to trade. The job is correctly sizing it relative to the buffer math.
Bottom Line
Position sizing failures are the most common cause of “psychology” failures in prop trading. The trader feels tilted because they keep losing — but the structural cause is that their sizing math was wrong from the first trade. The losses they experience are normal variance; the account death is the sizing’s fault.
Fix the sizing math and most psychological symptoms disappear. You stop tilting because you stop having existential losses. You stop revenge trading because no single trade can produce existential damage. You stop overtrading because each trade isn’t a desperate attempt to dig out of a hole.
The math is unsexy. It’s also the entire game.
Next read: The 5 Reasons You Keep Losing Money Trading (Data-Driven) — the pillar piece.
Find firms where the sizing math works: Verified Prop Firms Shortlist.
