SIMULATION · FTMO 2-STEP RULES

Can a 40% win-rate bot pass a prop firm challenge?

Short answer

Usually, if the 40% is real and the risk is small. We simulated FTMO's 2-Step rules for a bot that wins 40% of its trades at 1:2. At 1% risk per trade it passed both phases 72.8% of the time; at 2% only 26.3%, mostly on the 5% daily limit. A bot with no edge at all passed 31.8% of the time, so a pass proves little about the bot.

The rules we simulated

One firm, read on its own site today: the FTMO 2-Step challenge, from FTMO’s trading objectives page. The limits are a share of the initial capital, so the answer does not depend on the account size.

RulePhase 1 (FTMO Challenge)Phase 2 (Verification)
Profit target10%5%
Maximum daily loss5%, measured on equity (open trades count) from the balance at midnight CE(S)T, Prague timeSame
Maximum loss10% below the initial capital, staticSame
Minimum trading days44
Time limitNone listedNone listed

For US traders: FTMO offers its simulated challenges to US residents through a partnership with OANDA, announced on 26 August 2025, and the FTMO x OANDA site lists the same 2-Step objectives. Several other prop firms say they do not offer MT5 to US residents; our prop firm page names them. FTMO’s 1-Step challenge has a 3% daily limit and a trailing maximum loss, so none of the numbers below apply to it.

The model bot

  • Wins 40% of its trades and each winner pays twice the risk (1:2).
  • Risks a fixed 1% of the starting balance per trade, 3 trades a day.
  • Pays one spread per trade. The median gold spread in our 1:2 risk-reward test was $0.20, which on its $5 stop is 0.04R. A win nets 1.96R, a loss costs 1.04R, and the average trade makes +0.16R.
  • Each trade is independent of the last, and stops fill exactly where they sit.

This is a simulation of the rules, not a record of real challenges. Instead of drawing random paths, the script carries the exact probability of every balance through every trade until less than one part in a trillion is still trading. A 100,000-path Monte Carlo run with a fixed seed agreed with it to within 0.2 points.

What the simulation found

At 1% risk the 40% bot passed Phase 1 83.4% of the time, Phase 2 87.3%, and both 72.8%. The other 27.2% of attempts ended on the 10% maximum loss. None ended on the daily limit: three losses in one day cost 3.12%. The passes took a median 57 trades, about 20 trading days.

Win rate at 1:2Average tradePhase 1Phase 2Both phasesMedian trades to pass
30%-0.14R17.9%37.7%6.7%58
34.7% (no edge)0.00R49.5%64.3%31.8%67
40%+0.16R83.4%87.3%72.8%57
45%+0.31R95.7%96.3%92.2%42
50%+0.46R99.0%99.1%98.1%31

Read the no-edge row. A bot whose winners only just pay for its losers and the spread passed both phases 31.8% of the time. Phase 2 is the soft part: a 5% target with 10% of room below it was reached 64.3% of the time by a bot with no edge at all. A losing bot that won 30% still passed 6.7%. The same luck that makes a short track record prove nothing passes challenges.

Risk per trade decides most of it

Risk per trade40% bot passesMedian trades (days)Ended on daily limitNo-edge bot passes30% bot passes
0.25%99.6%347 (116)0.0%32.9%0.0%
0.5%92.6%154 (52)0.0%32.6%1.1%
1%72.8%57 (20)0.0%31.8%6.7%
2%26.3%12 (8)73.7%14.1%6.8%
3%23.0%6 (8)77.0%14.9%9.3%

Halving the risk to 0.5% lifted the 40% bot to 92.6%; at 0.25% it reached 99.6%, but the median pass took 347 trades. Doubling it to 2% dropped the pass rate to 26.3%, and 73.7% of attempts died on the daily limit.

The reason is plain arithmetic. At 2% risk, three losses cost 6.24%, past the 5% daily limit. With three trades a day, the chance that a given day is loss, loss, loss is 0.6 × 0.6 × 0.6 = 21.6%, and every failure in that row came from it. Spreading the same trades out helps: at 2% risk, one trade a day passed 52.7% with no daily-limit failures, against 26.3% at three a day and 28.5% at six.

The losing bot shows the other side. Its pass rate rose with risk, from 0.0% at 0.25% to 9.3% at 3%. Big risk is how a bot without an edge gets lucky, and it is also how a bot with one fails.

Small stops pay more spread

StopSpread as a share of riskBreak-even win rate40% bot passes (1% risk)
No cost0.000R33.3%82.6%
$100.020R34%76.5%
$50.040R34.7%72.8%
$30.067R35.6%66.9%

Same bot, same $0.20 spread. With no cost it passed 82.6%; on a $3 stop, where the spread is 0.067R, it passed 66.9%. A firm that also charges commission moves every row down again.

What to set if your bot is taking a challenge

  • Risk 0.5% to 1% of the starting balance per trade, not 2%. FTMO lists no time limit, so the smaller size costs time, not pass rate.
  • Cap trades per day so a full losing day stays under 5%. At 1% risk that is 4 trades: 4 losses cost 4.16%, 5 cost 5.20%. Count the spread in that sum. We replayed how often gold hands a bot that full losing day, and how far equity dips on the way on a year of Exness prices.
  • Be flat at the daily reset. FTMO's day turns over at midnight Prague time: 6 pm ET (5 pm CT) [22:00-23:00 GMT, by season] on 337 days of 2026, and 7 pm ET (6 pm CT) from 8–28 March and 25–31 October, when Europe and the US change clocks on different dates. At 6 pm ET gold is just reopening after its daily break, and its spread is wider than in the daytime. A trade held across the reset carries its open loss into a day whose limit was set from a balance that never booked it.
  • Prove the win rate before you pay for the challenge, on trades the bot was not tuned on. A pass is not that proof.
  • The tables assume every trade is independent of the last. If your bot’s losses come in bunches, it breaks a daily limit more often than the tables show. Our losing-streak study counts how long gold’s runs of down days have lasted, which is not the same thing as a bot’s losing trades.

If you run an EA on an evaluation, the prop firm drawdown calculator turns a limit into a lot size, and how trailing drawdown works covers firms whose maximum loss moves.

What this does not cover

  • One firm, one program. FTMO 2-Step only. Firms with a trailing maximum loss, a consistency rule or a time limit will give different numbers.
  • No slippage and no gaps. A stop that fills beyond its level, or a weekend gap, costs more than 1R, and FTMO counts it in equity.
  • Fixed risk on the starting balance. A bot that sizes from current equity shrinks after losses and grows after wins, which changes both tails.
  • The funded account is a separate stage with its own rules. This page stops at passing Phase 2.

Common questions

What win rate do you need to pass a prop firm challenge?

It depends on the reward-to-risk ratio and the risk per trade. At 1:2, a bot breaks even at about 34.7% once a $0.20 gold spread on a $5 stop is paid. In our simulation of FTMO's 2-Step rules at 1% risk per trade, a 45% win rate passed both phases 92.2% of the time, 40% passed 72.8%, and a break-even bot still passed 31.8%.

What risk per trade is safest for an FTMO challenge?

Lower than most people use. In the simulation, a 40% win-rate bot at 1:2 passed 92.6% of the time at 0.5% risk, 72.8% at 1% and 26.3% at 2%. FTMO's objectives page lists no time limit, so lower risk costs time (a median 154 trades at 0.5%) rather than pass rate.

Does the FTMO daily loss limit include open trades?

Yes. FTMO measures equity, which it defines as balance plus open profit or loss, swaps and commissions. Equity may not drop below the balance recorded at 00:00 CE(S)T minus 5% of the initial capital. Its own example: a $100,000 account with a balance of $102,000 at midnight has a limit of $97,000 for the new day. Midnight in Prague is 6 pm ET (5 pm CT) on 337 days of 2026 and 7 pm ET (6 pm CT) on the other 28.

Can a bot with no edge pass a prop firm challenge?

Often. In our simulation a bot whose wins exactly paid for its losses and spread passed both FTMO phases 31.8% of the time at 1% risk. A losing bot that won 30% at 1:2 passed 6.7% at 1% and 9.3% at 3%. A pass on its own says little about whether a bot has an edge.

How long does it take a bot to pass a two-phase challenge?

At 1% risk and three trades a day, the passes in our simulation took a median 57 trades across both phases, or 20 trading days. About one pass in ten took more than 122 trades (41 days). FTMO requires at least 4 trading days in each phase.

Further reading

Prop firm drawdown calculatorHow many trades prove an edge

Educational information only, not financial advice. Trading leveraged products carries substantial risk of loss. Last updated 2026-11-14.