What pass rate can you actually expect?
About 90 out of 100 on a two-phase FTMO challenge, in roughly three months. Here is the full grid behind that sentence: six risk levels, ten rule formats, and the failure rate that belongs next to every pass rate.
On a two-phase FTMO challenge, a well-built Expert Advisor run at sensible risk passes about 90 times out of 100, and takes roughly three months to do it. On a one-phase 8% challenge the same engine passes 93 times out of 100 in about eight weeks. Those are the honest numbers, and they are lower and slower than almost everything advertised.
This page publishes the full grid: six risk levels across ten rule formats, with the time each one takes and how often it blows the account instead. The figures come from 7,000 simulated first-passage paths per configuration, built on 2022-2026 real tick data.
Every number here is a backtest. They describe how one engine behaved on that period, not what any system will do next month. We publish them because the alternative, a single round number with no method attached, tells you even less.
The three things that move the number
1. The format, more than the strategy
A one-phase challenge has one barrier to clear. A two-phase challenge has two, and clearing the second one after a drawdown accumulated in the first is materially harder. The same engine, same risk, same period:
| Format | Passes | Median time | Blows the account |
|---|---|---|---|
| One phase, 8% target, 8% floor | 93.1% | 57 days | 2.9% |
| Two phases, 8% then 5% | 91.8% | 79 days | 2.9% |
| FTMO, two phases, 10% then 5% | 89.7% | 91 days | 2.6% |
| Two phases with a 4% daily limit | 89.4% | 99 days | 2.7% |
Note that the pass rates are close but the times are not. The format costs you weeks more than it costs you percentage points.
2. Risk, in both directions
Raising position size makes the target arrive sooner and the floor arrive sooner too. There is a level where those cancel out, and past it the pass rate falls. On an FTMO two-phase account:
| Risk level | Passes | Median time | Blows the account |
|---|---|---|---|
| Conservative | 79.4% | 147 days | 0.2% |
| Steady | 88.2% | 114 days | 1.2% |
| Standard | 89.7% | 91 days | 2.6% |
| Fast | 87.3% | 74 days | 5.8% |
| Aggressive | 80.7% | 52 days | 11.4% |
| Maximum | 72.5% | 38 days | 16.6% |
The peak is not at the top. Going from Standard to Maximum cuts the median from 91 days to 38, which sounds excellent until you notice the failure rate goes from 2.6% to 16.6%: six times more accounts destroyed. At a typical fee, one failure in six wipes out the time you saved on the other five.
The conservative end is worth reading too. At the lowest risk the account almost never blows up, 0.2%, but the pass rate falls to 79.4% because the account runs out of the window before it reaches target. Too little risk fails in a different way.
3. The drawdown type, which nobody mentions
A static drawdown floor sits at a fixed level below your starting balance. A trailing floor follows your equity peak upward: make 4% and the floor moves up 4% with you. For a strategy that holds positions over multiple days, the difference is brutal.
| Drawdown type | Passes | Blows the account |
|---|---|---|
| Static, 10% | 89.7% | 2.6% |
| Static, 6% | 85.5% | 12.3% |
| Trailing from peak, 6% | 64.5% | 35.5% |
Same target, same risk, same engine. Trailing costs 25 percentage points and more than doubles the failures. If you are comparing two firms on price and one of them uses a trailing floor, you are not comparing like with like.
How long it actually takes
Time is where advertised numbers depart from reality most sharply. A two-phase challenge is two separate races, and the second one starts from zero.
| Risk level | Phase 1 alone | Both phases | Both within 45 days |
|---|---|---|---|
| Steady | 71 days | 117 days | 3.6% |
| Standard | 56 days | 92 days | 9.5% |
| Fast | 45 days | 75 days | 18.5% |
| Aggressive | 32 days | 52 days | 33.9% |
All figures are trading days, not calendar days. At the Standard level, fewer than one account in ten completes both phases inside 45 trading days. Any vendor promising a funded account in six weeks on a two-phase challenge is either running very high risk, quoting phase one only, or quoting the tail of the distribution as if it were the middle.
What happens after you are funded
The challenge is the part everybody discusses, and it is the easier half. On a funded account the arithmetic inverts: during a challenge you risk an entry fee, on a funded account you risk an asset that pays out for as long as it survives.
| Risk level | Survives 12 months | Gross return |
|---|---|---|
| Conservative | 99.8% | +22.8% |
| Steady | 98.5% | +30.5% |
| Standard | 96.9% | +38.3% |
| Fast | 94.2% | +46.6% |
| Aggressive | 88.0% | +62.7% |
The right level on a funded account is lower than the right level on a challenge, and the difference is not small. An 88% survival rate sounds acceptable until you translate it: roughly one funded account in eight is gone within the year, and with it every future payout.
This is also why a monthly return promise of 10% or 15% on a funded account should make you suspicious. The engine described here returns 30-38% a year at sensible risk, which is 2.2 to 2.8% a month. Getting to 15% a month requires roughly five times that, and the survival column shows what happens on the way.
How to read any pass rate you are shown
Four questions, in order. If a vendor cannot answer the first two, the number is decorative.
- Which format? One phase or two, what target, what daily limit, what floor, and is the floor static or trailing. A number without rules attached describes nothing.
- What would a system with no edge have scored on the same rules? On a two-phase FTMO challenge the answer is 25.6%; on a one-phase 8% challenge it is 43.9%. Subtract that from the advertised figure and you have the contribution of the strategy. We ran that control on our own engine and published the gap.
- At what risk level, and what is the failure rate at that level? Pass rate and blow-up rate are two halves of one answer. A high pass rate quoted without its failure rate is half a sentence.
- Over what period, and was any of it out of sample? A strategy tuned on the same data it is measured on will show whatever number was wanted.
What we do not claim
Resampled from one period. The probabilities come from block-resampling 2022-2026. They describe how the system behaved then, and they are not a forecast.
The observed sequence was lucky. Across 8,000 resampled orderings the median maximum drawdown was 32.4R against 27.6R actually observed, and the 95th percentile was 49.7R. Expect a drawdown close to twice the historical one without anything being wrong.
One engine, two markets. Everything here describes a specific system trading the Nasdaq 100 and gold. It says nothing about how a different strategy behaves under the same rules.