The exact math behind every PropFirmEV ranking

For each prop firm × account size combination, we estimate three numbers:

• Pass probability — the % of simulated evaluations your trade-stat distribution clears.
• Expected payout — the average $ withdrawn from one successfully-funded attempt, capped by the funded-account drawdown.
• Expected value (EV) per attempt — what you net on average across many attempts.

The headline EV formula is:

EV = pass_prob × payout − pass_prob × activation_fee − eval_cost

The eval cost is paid every attempt (whether you pass or fail). The activation fee is paid only when you pass. Both come out of your pocket; both are in the formula.

We treat your trade history as a probability distribution and simulate 1,000 independent evaluation attempts for every firm × size. Each attempt is a sequence of up to 400 trades with these rules:

• One trade per day (so EOD and intraday drawdowns converge under our assumption; we still surface the difference for awareness).
• Each trade is a win with probability = your historical win rate. Wins return your average win; losses return your average loss (negative).
• We update an EOD trailing drawdown floor after every trade. If balance ever drops below the floor, the attempt fails.
• If the daily loss limit (where firms have one) is breached, the attempt fails.
• If the balance hits the firm's profit target before the drawdown is hit, the attempt passes.

The seed is deterministic for a given input, so re-running with the same stats produces the same EV. We chose 1,000 runs because the EV confidence interval is narrow enough that increasing it doesn't meaningfully change the ranking.

Once an attempt passes, we run a second Monte Carlo (2,000 runs) on the funded phase. The whole funded-account model collapses to a single number:

expected_payout = P(reach +$2,000 profit) × $2,000

The reasoning, step by step:

• On every funded account the trader has $2,000 of EOD trailing drawdown. We simulate them trading 1 trade/day until either (a) cumulative balance reaches +$2,000, or (b) they bust on the trailing floor.
• If they reach +$2,000: the trader withdraws 50% of profit — a flat $1,000 payout. The trailing drawdown then locks (stops moving up). After 5 more winning days they can take another $1,000 payout — and we assume they do, since they've already cleared the high bar of reaching +$2,000 in the first place. Total: $2,000.
• If they bust before +$2,000: zero payout. Funded-account fail.

This is intentionally a clean, conservative model. It ignores unbounded additional payouts for elite traders (real life can be much higher) but reflects the typical first-payout cycle most active futures traders run on these products.

On /compare, we surface a different number from the calculator's EV: the average $ a marginal trader spends before they have a funded account.

cost_to_funded = eval ÷ pass_rate + activation_fee

For the compare page we assume a fixed 25% pass rate (a realistic baseline for the typical trader — industry-published pass rates sit in the 10–30% band). On the personalised calculator we use your simulated pass probability instead.

The intuition: at a 25% pass rate, a $50 eval effectively costs $200 (4 attempts on average). Add the one-time activation fee, and you have the realistic out-of-pocket to actually start trading funded.

• One trade per day. EOD and intraday drawdowns are mathematically equivalent under this assumption. If you scalp 5–10 times a day, your real intraday drawdown exposure is meaningfully higher than our model shows.
• Stationary distribution. We assume your win rate and average win/loss are stable. If your trading has improved or degraded recently, only the most recent slice will produce an accurate EV.
• No consistency-rule modelling. Several firms have a "biggest day ≤ X% of total profit" rule. We don't simulate that, which slightly over-estimates pass rates for traders with one outsized win in their history.
• Universal funded-account model. We use a single $2k MLL / $2k payout-per-pass heuristic (50% of +$2k profit, then another $1k after 5 more winning days). This makes EV comparable across firms but ignores firm-specific scaling, payout-ladder, and live-account bonuses.
• No fees for data, platform, or commissions. Brokerage and platform fees vary too much by broker; modelling them at the firm level would mislead.
• Promo pricing assumed. Where we list a discounted eval cost (e.g. JJ-promo prices), it's because that's the price the typical reader of this site can access. Rack pricing is documented in each firm's notes field.

A few of the choices in the model could reasonably go the other way. Here's why we picked what we picked:

• 25% pass-rate baseline (compare page): Published industry stats sit in 10–30%. 25% protects the marginal trader from anchoring on marketing claims; better traders will naturally see a lower personal cost-to-funded on the calculator.
• $2,000 payout per pass: Once a trader reaches +$2,000 on a funded account, they almost always take the first $1,000 payout (locking the drawdown) and most clear the additional 5 winning days for the second $1,000. We bake in both for simplicity — but ignore unbounded additional payouts for traders who keep running winners.
• EOD trailing DD default: Most firms either use EOD trailing or settle to it on funded. Apex's intraday variant is flagged separately on the compare page.
• No commission modelling: Commissions are stable across firms (they're a broker / data feed cost), so they don't affect the ranking. They absolutely affect the absolute EV — model accordingly.

• Consistency-rule modelling (per-firm best-day-percent caps).
• Firm-specific funded-account scaling and payout-ladder models.
• A "what if" slider so you can stress-test changes in your win rate or R:R.
• Sentry / Datadog runtime error reporting so we catch bugs in production faster.