FTMO Daily Drawdown Rule Explained: The 5% Static Floor

FTMO 2-Step caps daily loss at 5% of initial balance, measured against a static floor that re-anchors each day. The mechanics, the boundary cases, and how Pass Lab evaluates it.

The FTMO 2-Step Daily Drawdown Rule states that the account's equity must not, at any point during a trading day, fall more than 5% below the prior day's end-of-day balance, per the FTMO Trading Objectives page as of 2026-05-01. The rule reads simply but contains two non-obvious mechanics — the daily reset anchor and the inclusion of floating P&L on open positions — that account for most early-evaluation failures traders attribute to "bad luck". This post documents the rule as published, walks through the math at the boundary, and shows how Pass Lab evaluates it during a Monte Carlo iteration.

By the FXOptimize research team. Last updated 2026-05-12.

The rule, as written

FTMO's Trading Objectives page (as of 2026-05-01) states that on the 2-Step Challenge, the account's equity at any moment of a given trading day must not be lower than 5% of the initial account size below the previous day's closing balance. For the $100,000 Standard challenge, this is a $5,000 daily floor that re-anchors each day.

Three properties of the rule are worth recording exactly as published:

• The threshold is 5% of the initial account size, not of the prior-day balance. A $100,000 account always uses $5,000 as the daily limit, even when the prior-day balance has grown to $108,000.
• The anchor is the prior day's end-of-day balance, including unrealised P&L from any positions held open at the daily reset (FTMO's server-time daily rollover). Floating gains or losses at the rollover become part of the next day's anchor.
• The rule applies continuously during the trading day. Both realised losses and unrealised drawdown on open positions count toward the 5% threshold. The rule is not evaluated only at end-of-day.

The rule applies on FTMO 2-Step Phase 1 and Phase 2. The 1-Step variant uses a different daily drawdown structure (3% with a trailing end-of-day total drawdown), which is covered separately. Funded-stage rules are out of scope here.

What the rule actually means in practice

The daily drawdown is static in the sense that the 5% threshold does not trail upward as the account grows — it remains pegged to the initial account size. But the anchor against which the 5% is measured does move: it re-anchors to the prior day's end-of-day balance every server-day.

daily_floor(day_t) = balance_at_eod(day_{t-1}) − 0.05 × initial_balance
breach if: equity_at_any_moment(day_t) < daily_floor(day_t)

Worked example. Initial balance: $100,000. Daily floor offset: $5,000.

• Day 1 close: $102,000. Day 2 floor: $102,000 − $5,000 = $97,000.
• Day 2: an open EUR/USD position runs to −$4,800 unrealised at 14:00. Equity = $102,000 − $4,800 = $97,200. Above floor. Pass.
• Day 2: position deepens to −$5,200 unrealised at 14:30. Equity = $96,800. Below floor. Breach. Evaluation terminated.

The breach in the example is not a closed loss — it is an unrealised excursion. The position could close at break-even an hour later and the closed equity curve would never have shown a problem. The daily drawdown rule does not care.

Boundary case. Equity touching exactly the floor passes; equity moving below the floor by any amount fails. FTMO's published language uses "lower than", so the floor is inclusive on the pass side.

A trader passing the daily floor by $50 at the worst tick of a session is still a passing trader. The Strategy Tester reports the same trade as profitable; the live evaluation would have terminated it. The gap between those two readings is the entire reason for path-dependent simulation.

Common misreadings

• "The 5% is measured against the prior day's balance, not the initial balance." The threshold magnitude is 5% of the initial balance; the anchor it is subtracted from is the prior day's closing balance. The two are different inputs to the same formula.
• "Only closed trades count toward daily drawdown." Floating P&L on open positions is included. An unrealised excursion that retraces before close still breaches if it crosses the floor.
• "The daily window resets at midnight UTC." The reset is at FTMO's server-time daily rollover (CE(S)T), not UTC. A trader in another timezone may experience the rollover at a non-obvious local hour.
• "The rule is the same on FTMO 1-Step." It is not. The 1-Step uses a 3% daily limit with a different total-drawdown structure.

How Pass Lab simulates the rule

Pass Lab's rule evaluator is a per-trade state machine that runs inside each Monte Carlo iteration, with within-day trade order shuffled to capture path-dependence. See the engine description in methodology.md §4.4.

For the FTMO daily drawdown rule specifically, the simulator tracks:

• A day-bounded running balance that resets at the start of each simulated trading day (using the trade's exit timestamp to determine day membership against FTMO's server-time rollover).
• A daily anchor captured at the prior simulated day's end-of-day close.
• A per-trade floor check evaluated after each trade in the shuffled iteration order: if cumulative_equity < daily_anchor − 0.05 × initial_balance, the iteration is recorded as a daily_drawdown_breach and short-circuits.

Within-day trade order is shuffled with a deterministic ChaCha8 PRNG seeded by master_seed XOR firm_index XOR window_index, preserving daily clustering — losses cluster on the day the strategy actually had losses — while randomising the order in which those losses arrive. Eight trades on the same Tuesday in different orders produce different daily-drawdown trajectories, and the Monte Carlo iteration count of 1,000 per walk-forward window samples that order-space.

The FTMO daily drawdown rule is evaluated per-iteration, not as a pre-flight short-circuit. There is no portfolio pattern that structurally guarantees a daily drawdown failure (the way a martingale structurally guarantees a martingale-ban breach), so Pass Lab does not short-circuit the firm match. Instead, every iteration runs the rule, and the per-window pass rate is the fraction of iterations in which no breach occurred.

Across all walk-forward windows for the firm, the window pass rates are then resampled 5,000 times to produce a 95% bootstrap confidence interval. Pass Lab reports the lower bound of the CI as the headline metric, not the point estimate.

Where the simulation differs from the live evaluation

Pass Lab is a backtest analysis tool; the number it produces is a historical-statistical projection over the user's backtest data, not a forecast of a live account. Several specific divergences for this rule:

• Intraday MAE is approximated. Pass Lab V1 treats trades as point events at exit time and computes daily lows from running prefix-sums of trade P&L. A real intraday MAE model would require threading hourly candle data through the Monte Carlo loop. Portfolios with large unrealised excursions inside open trades — particularly grid systems and positions held across high-impact news — may have their daily-drawdown risk understated.
• No live execution model. Slippage on news, spread widening at session boundaries, EA-to-broker latency, and the gap between the backtest's broker and FTMO's broker are not modelled. An unrealised excursion that breaches the floor on FTMO's broker by 3 pips of spread widening would not necessarily breach on the user's backtest broker.
• Funded-stage rules are not modelled. Pass Lab V1 evaluates the FTMO 2-Step evaluation phase only; funded-stage rules differ and are out of scope for V1.

Full caveat list lives in methodology.md §4.6.

The 2026 FTMO product update also removed the previously published 30-day evaluation cap on both 2-Step and 1-Step, so the evaluation window in Pass Lab is now treated as open-ended. Pass Lab uses a 60-day window in this case — long enough to compound to the 10% profit target and clear the 4-day minimum, short enough for the bootstrap to estimate the tails.

Closing

The FTMO daily drawdown rule caps each day's equity loss at 5% of the initial account size, measured against the prior day's end-of-day balance, applied continuously to closed and floating P&L. Pass Lab simulates the rule per-trade inside a within-day-shuffled Monte Carlo loop and reports the CI lower bound of the resulting pass rate. The official rules page is the authoritative source for the current threshold; this post is an explanation of mechanics, not a substitute for verification.

CFTC Rule 4.41 disclosure: hypothetical and simulated performance results have inherent limitations and do not represent actual trading; no representation is made that any account will or is likely to achieve profits or losses similar to those shown.

Related Articles & Tools

• Pass Lab Methodology: Monte Carlo, bootstrap CIs, and walk-forward windows
• Propfirm Challenge Comparison 2026
• FXOptimize Methodology — §4.2 firm profiles, §4.4 rule simulator, §4.6 limitations
• Frequently Asked Questions
• External: FTMO Trading Objectives (authoritative rules source)