Your Expert Advisor finishes a backtest in profit. The report's maximum drawdown is 7%, and the prop firm allows a 10% maximum loss. So the EA would have passed the challenge?
Not necessarily. A challenge can end on one afternoon's floating loss, even if the trade recovers and the run finishes far above its profit target. Prop firm rules are not judged on the final result or on one drawdown figure. They are judged on the path the account took, day by day, against limits that move.
This guide shows how those limits are calculated, works through two breaches that an MT5 report hides, and then shows how to run the same check on a finished backtest in Atlaas.
Quick answer: how do you check a backtest against prop firm rules?
Replay the account's equity against the daily and overall loss limits for every day of the test, then check the profit target and the other objectives.
- Write down the exact rules for the firm, program and phase you plan to buy.
- Rebuild each day's starting balance and lowest equity, including open positions.
- Work out each day's daily loss floor from the balance at the firm's reset time.
- Work out the overall loss floor, static or trailing.
- Find the first day equity fell below either floor. That is where the challenge ended.
- If no floor broke, check the profit target, minimum trading days and any consistency rule.
The MT5 strategy tester report does not do steps 2 to 5 for you. Its drawdown figures are measured from the run's own peaks, not from a firm's daily reset or its fixed floors.
Why a profitable backtest can still fail
Three gaps separate a good backtest from a passed challenge.
Balance is not equity. Balance moves when a position closes. Equity also counts the profit or loss on positions that are still open. Most firms test their loss limits against equity, so a position that falls deep into loss and later closes at break-even barely touches the balance curve but can breach the rule while it is open.
Drawdown is measured from the wrong place. The tester measures how far the account fell from its own previous high. A daily loss rule measures from the balance at the start of each day. A static maximum loss rule measures from the initial deposit. A trailing rule measures from a high-water mark that only moves up. The same equity curve can pass one and fail another.
Time matters. A daily limit resets at a set time in a set time zone. A loss that is fine before the reset can be a breach after it, as the second example below shows.
Write the rules down first
Two programs from the same firm can calculate loss differently, so get the rules from the firm's own page for the exact account you plan to buy, and note the date you read them.
| Rule | What to pin down |
|---|---|
| Daily loss | The percentage, what it is a percentage of (usually the initial deposit), and the reference it is subtracted from (the day's starting balance, equity, or the higher of the two). |
| Daily reset | The time zone, the reset time and how daylight saving moves it. |
| Maximum loss | Static or trailing. If trailing: what it trails (end-of-day balance, intraday balance or equity) and whether it ever stops rising. |
| Breach condition | Whether touching the floor fails the account or equity must go below it. |
| Profit target | The target for each phase, and whether positions must be closed for it to count. |
| Trading days | How many are needed and what counts as one. |
| Consistency | The largest share of profit one day may make, and what that share is divided by. |
As a reference point, FTMO's trading objectives, read on 1 November 2026, set these for its two challenges:
| FTMO 2-Step | FTMO 1-Step | |
|---|---|---|
| Profit target | 10%, then 5% in Verification | 10% |
| Daily loss | 5% of initial capital below the 00:00 CE(S)T balance, tested on equity | 3% of initial capital below the 00:00 CE(S)T balance, tested on equity |
| Maximum loss | Static: 10% below initial capital | Trailing: 10% of initial capital below the highest 00:00 CE(S)T balance, never moves down |
| Minimum trading days | 4 per phase | None |
| Consistency | None | Best day at most 50% of total closed profit (orders from 12 October 2026) |
The same page lists other products with different rules. Its Instant account, for example, trails its maximum loss on intraday balance highs, not midnight balances. Do not carry these numbers over to another FTMO product or another firm.
The rest of a firm's agreement, such as news trading restrictions, lot limits, EA and copy trading terms, and inactivity rules, is outside any drawdown check. Read it separately.
Example 1: a daily breach the closed trades hide
These are illustrative numbers, not a backtest result. Take a $100,000 challenge with a daily loss limit of 5% of the initial deposit, measured from the day's starting balance and tested on equity.
The daily allowance is $5,000. The day starts at a balance of $102,000:
| Step | Amount |
|---|---|
| Balance at the daily reset | $102,000 |
| Daily allowance: 5% of $100,000 | $5,000 |
| Daily floor: $102,000 - $5,000 | $97,000 |
| Trades closed today, net of costs | -$2,000 |
| Balance now | $100,000 |
| Loss on the position still open | -$3,500 |
| Equity now | $96,500 |
Equity is $500 below the floor. The account has failed, even though the day's closed loss is only $2,000 and a trade log would show nothing alarming.
If price turns and the open position closes at a profit an hour later, the result does not change. The breach happened when equity crossed the floor, and a replay should stop there.
Example 2: the reset that moves the floor under an open trade
Same account, same $5,000 allowance. A day starts at $100,000, so its floor is $95,000. During the day the EA closes winners worth $4,000, lifting the balance to $104,000, while another position sits $6,000 in loss. Equity is $98,000, comfortably above $95,000.
Then the daily reset arrives. The new day's reference is the $104,000 balance, so the new floor is $99,000. Equity is still $98,000 (ignoring swap and price movement), and it is now $1,000 below the floor. The account fails at the reset without price moving at all.
Two lessons follow. Open losses carried across a reset are measured against a floor built from the balance, not the equity. And the reset time decides which side of the line a trade falls on, so a firm that resets at midnight CE(S)T and a broker server that rolls over at another hour can give different days.
Static and trailing maximum loss
A static floor is fixed to the initial deposit. On a $100,000 account with a 10% maximum loss, it sits at $90,000 for the whole challenge, however much the account makes.
A trailing floor follows a high-water mark up and never comes back down. Take a rule that trails the highest day-start balance by $10,000:
| Day-start balance high | Trailing floor | Static floor |
|---|---|---|
| $100,000 | $90,000 | $90,000 |
| $103,000 | $93,000 | $90,000 |
| $106,000 | $96,000 | $90,000 |
Once a day has started at $106,000, the floor stays at $96,000. A later drop to $95,500 fails the trailing account and leaves the static one with $5,500 to spare. An EA that gives back a large share of its gains before going higher again is much harder to run under a trailing rule.
What the mark is built from matters as much. A rule that trails intraday balance or equity moves sooner and higher than one that only updates at the daily reset. Choosing "trailing" in a checker without matching that detail can produce the wrong verdict.
Consistency: what is the best day divided by?
Some programs cap how much of the profit one day may make. Read the denominator carefully. Suppose a phase has four trading days with these net closed results: +$5,000, +$2,500, +$2,500 and -$2,000.
| Formula | Calculation | Share | Against a 50% cap |
|---|---|---|---|
| Best day / total net closed profit | $5,000 / $8,000 | 62.5% | Over the cap |
| Best day / sum of profitable days | $5,000 / $10,000 | 50% | Within the cap |
Same days, opposite answers. FTMO's current 1-Step rule divides by total profit; its legacy Best Day rule, for accounts ordered before 12 October 2026, divides by the profitable days. At FTMO, going over the cap is not a loss breach: the account keeps trading until the share falls back under it.
Make the backtest worth checking
A rule check is only as good as the trades it replays. Before checking anything:
- Match the account. Use the deposit, account currency, leverage and EA inputs you would trade on the challenge, on the broker's symbols where you can.
- Count the costs. Spread, commission and swap all move both balance and equity.
- Use real tick data where it matters. MetaQuotes notes that with real ticks the spread can change within a minute bar, while generated ticks use the bar's fixed spread. Where a minute has no real ticks, the tester generates them, so the mode name is not a guarantee. Check the history you actually have.
- Keep an unseen period. If you tune inputs until a run passes, you have fitted them to that history. Hold some dates back and check them once, at the end.
How to check a backtest against prop firm rules on Atlaas
Atlaas runs the replay above on a backtest that finished on Atlaas, so the first step is the backtest itself. It does not import reports from a desktop MT5 terminal.
1. Run the backtest
Open Backtests and click New backtest, keeping the Single tab. Pick the Expert Advisor, its Version and the Terminal to test against, then set the Symbol, Timeframe, Tick mode and dates. Under Advanced, set the deposit, currency, leverage and execution delay to match the challenge account. Click Run backtest.
The test uses that terminal's broker server and login, so symbols and history come from your broker, and your terminals keep trading while it runs. To keep a period unseen, switch on Hold back an out-of-sample period: those results stay hidden until you click Reveal out-of-sample. The backtesting guide covers the rest of the form.
2. Pick the rules
Open the finished run and choose Prop firm, the last entry in the section bar. In the Prop firm check card, pick a challenge under Rules: FTMO 2-Step Challenge, FTMO 1-Step Challenge or FundedNext Stellar 2-Step. The check runs as soon as you pick. Under the result, a link goes to the firm's published rules with the date Atlaas last checked them. Compare them with your account before you rely on the preset.
3. Change the numbers if your program differs
Choose Custom rules, or click Edit rules to start from a preset. For each phase you can set:
- Profit target %
- Daily loss %
- Max loss %, with Max loss from set to Static or Trailing
- Min trading days
- Best day cap %, left empty for no cap
Click Check these rules. Every percentage is of the initial deposit. Daily loss is subtracted from each day's starting balance; Trailing follows the highest balance a day has started on; the best day cap divides the best day's net closed profit by the phase's total net closed profit. Those definitions are fixed. If your firm uses a different reset time, trails intraday equity or divides by profitable days only, changing the numbers will not make the check match it.
4. Read the verdict
The headline gives one of three answers:
- Would fail: equity went below a loss floor. It names the phase, the limit (daily loss or max loss) and the day, and a line under the table shows how low equity fell, when, and the floor it broke.
- Would pass: every phase met its target and objectives, and it says how many days that took.
- Not passed: the run ended before a phase was complete, with how close it got. Nothing broke; the target, trading days or best day cap was not met in time.
Below it, each phase has a row with Profit target, Worst day, Max loss used, Trading days and Best day, and reads Passed, Failed, Not reached or Not started.
A few rules of the replay are worth knowing. A trading day is a day at least one position opened. The target only counts at a close that leaves no position open. When a phase passes, the next one starts the following day at the initial deposit: Atlaas shifts the same run's later results down to that balance, rather than running the EA again from scratch.
What the estimate can miss
The card is labelled Estimate, and the note under the result says why. Read it before trusting the headline.
- Days are broker server days. FTMO counts in CE(S)T and other firms use their own clocks. Trades held across one reset but not the other can land on different days.
- Equity is rebuilt from bars. Each open position is valued at the worst point of each bar. For one symbol that finds the low. Across several symbols, the worst points of different bars may not happen at the same moment, so the rebuilt low can be lower than the account really went: cautious, but not exact.
- No bars, no floating losses. If a run exported no bars, the note says only closed trades were checked. Losses on open positions are then missing entirely, and a pass says nothing about an equity-based limit.
- Costs land at the close. Swap and commission are counted when a position closes, not while it is open.
- Only these objectives are checked. News rules, lot caps, EA permissions and the rest of the agreement are not.
After the verdict
If it fails, go to the breach day first. Was it one oversized position, several correlated ones open together, a loss carried over the reset, or a trailing floor that had already moved up? Fix the cause in the EA's sizing or logic and run the backtest again. Loosening the rules in the checker until it passes fixes nothing.
If it passes, look at how close it came. Worst day and Max loss used show the deepest loss as a percentage of the deposit, next to each limit. A worst day of 4.8% against a 5% limit leaves almost no room for a slower fill or a wider spread. Then test other start dates and the period you held back, since a challenge can start in any month. Finally, run the same EA version and inputs on a demo account before paying for a challenge.
Frequently asked questions
Can an EA be profitable and still fail a prop firm challenge?
Yes. A breach ends the challenge when it happens, and later profit does not undo it. A run can also finish in profit without reaching the target in time or meeting the trading day and consistency rules.
Can I check prop firm rules from a trade list?
Only the closed-trade side. A trade list shows each position's result but not how far equity fell while it was open, so it cannot show whether an equity-based daily or maximum loss limit held.
Does "Would pass" mean the EA will pass a real challenge?
No. It says one historical run would have met the rules you picked, under the assumptions above. It does not predict future results or confirm compliance with everything in the firm's agreement.
Start from a finished run in Backtests, pick your challenge under Prop firm and read the worst day before the headline profit. The prop firm check guide is the short reference to keep open beside it.