We've published how we test, what we killed, and why most backtests lie. This post answers the question everyone actually opens with: how much money does the thing make?
Two rules for this page. First, no point estimates — every forward number is a distribution, because a single figure is the oldest trick in strategy marketing. Second, we publish the number that hurts: the prop-firm ladder math at the end is the least flattering thing on this website.
The certified champion: a 30-minute opening-range breakout on micro Nasdaq futures, long only, stop at the far side of the range, flat by 15:55. Seven and a half years of real CME data at 1-minute resolution, commissions and slippage included.
| trade-level statistic | value |
|---|---|
| trades taken | 1,324 (≈176/year) |
| days with no trade at all | 30.2% |
| average winner / average loser | $175.66 / −$158.94 |
| payoff ratio | 1.11 |
| expectancy per trade | $12.40 |
| biggest single win / loss | $2,143 / −$783 |
| longest winning / losing streak | 7 / 8 trades |
| how trades end | 64% end-of-day · 34% stopped · 2% hit target |
That last row is the strategy's whole personality. It almost never reaches its profit target — it takes what the session gives and goes home. Anyone selling you a breakout system with a high target-hit rate is selling you a fill model, not a strategy (see the finale).
"$2,181 per year" is an average of a very wide spread. Here's the actual range of one-year outcomes — every real 252-session window in our history, plus 20,000 block-bootstrap simulations:
| one-year outcome, 1 micro | bootstrap | real windows |
|---|---|---|
| 5th percentile (bad year) | −$2,407 | −$1,124 |
| 25th percentile | $270 | $1,956 |
| median | $2,191 | $2,945 |
| 75th percentile | $4,098 | $3,967 |
| 95th percentile (great year) | $6,910 | $5,643 |
| chance the year is negative | 21.8% | 12.8% |
And the month-to-month texture, which is what you'd actually live through:
Backtest returns are not income. To trade this with someone else's capital you have to pass an evaluation, then survive a funded account's rules long enough to withdraw. We simulated both using the strategy's real daily P&L distribution — not a smooth win-rate model — with the rules of Tradeify's Growth 150K product, including the trailing drawdown that keeps rising underneath you (a detail we got wrong once and corrected).
| Growth 150K evaluation | 2 micros | 3 micros | 5 micros |
|---|---|---|---|
| pass probability per attempt | 28.0% | 37.6% | 34.3% |
| median days to pass | 166 | 109 | 48 |
| expected fees per funded account | $614 | $527 | $551 |
Roughly one attempt in three succeeds, and each funded account costs about $550 in fees to obtain. Fine. Then the funded account has to actually pay you:
Read that chart carefully, because it's the crux. Between a third and a half of funded accounts ever generate one payout. The median account earns nothing: the trailing drawdown ends it before the balance clears the payout buffer. Bigger size passes the eval faster but kills the funded account faster too — median account life falls from 267 days at 2 micros to 48 days at 5.
This is the scenario we've been asked about most — start with $2,000, buy evals, compound the payouts into more accounts. Simulated 1,500 times per size, one year, using this strategy's real distribution and consistent Growth 150K rules throughout:
| $2,000 → one year | 2 micros | 3 micros | 5 micros |
|---|---|---|---|
| funded accounts earned | 0.67 | 1.55 | 3.33 |
| chance of at least one | 54.9% | 84.1% | 98.5% |
| median end-of-year cash | $893 | $608 | $899 |
| mean end-of-year cash | $996 | $1,231 | $2,023 |
| 90th percentile | $1,262 | $2,858 | $5,799 |
| chance you end below $2,000 | 97.0% | 82.6% | 68.4% |
Why? An edge of $12.40 per trade is real but thin, and the ladder taxes it three ways: eval fees on every attempt, a rising trailing drawdown that ends most accounts before payout, and a payout buffer that requires a run of profit the strategy produces only occasionally. The firm's rules aren't unfair — they're simply priced above this particular edge.
Trading it yourself, one to three micros, in your own account: a median of roughly $2,200–2,900 a year per contract, one year in five red, spending 90% of your life in a drawdown, for about 176 trades. That is a real edge and an honest, small business. It will not change your life and it is not supposed to.
Farming prop evaluations with it: the math doesn't work at $2,000 on the product we modeled. Either the edge needs to be bigger, the sizing smarter, or the product cheaper. A cheaper evaluation with a smaller profit target changes this materially — the fee and the target are the two variables that dominate everything.
And the live caveat, stated plainly: every number here is a backtest. Our own 2026 is negative so far — the widest volatility on record is squeezing a range-based stop — and the published research on strategy decay says newly-deployed edges typically lose a quarter to a half of their backtested Sharpe on contact with the future. Take the medians above, haircut them, and then decide.
scripts/empirical_prop_sim.py. Backtested results, not live results. Nothing here is investment advice; futures trading carries substantial risk of loss.