Your P&L tells you how each trade ended; MAE and MFE tell you what happened on the way there. With those two numbers you can check whether your stop is wider than the setup needs and how much of the move you give back before you exit. This guide defines both, applies them to ten ES trades and walks through the traps that turn the analysis into noise.
What MAE and MFE are
- MAE (maximum adverse excursion): the distance between your entry and the worst price the market reached while the position was open. On a long, entry minus the low; on a short, the high minus entry.
- MFE (maximum favorable excursion): the distance between entry and the best price reached while the position was open.
- ETD (end trade drawdown): what you gave back from the best point to the exit, that is, MFE minus the final result. NinjaTrader shows it next to the other two.
All three are measured only between entry and exit: whatever price does after you are flat does not count. Two useful properties follow: a trade that gets stopped out has an MAE equal to the stop distance (plus slippage), and a winner whose stop never moved has, by definition, an MAE smaller than that stop.
In ticks, dollars and R
A long ES trade with the stop 6 points below entry (24 ticks, $300 per contract) that went 2 points against you and 8 in your favor:
| Measure | Ticks | $ on 1 ES | $ on 1 MES | R (6-point stop) |
|---|---|---|---|---|
| MAE (2 points) | 8 | $100 | $10 | 0.33R |
| MFE (8 points) | 32 | $400 | $40 | 1.33R |
Dollars depend on how many contracts you trade: they tell you how much the account suffered, not how trades compare. R, the excursion divided by the stop distance, compares everything with everything: an MAE of 0.33R means the trade used a third of the risk you gave it, whether in ES, MNQ or crude. How that initial risk is set is covered in the guide to R multiples.
Where to get the data
In NinjaTrader, the Trade Performance report shows MAE, MFE and ETD for every trade in the Trades tab, and they export with the rest of the report; the guide to exporting NinjaTrader trades has the steps.
If your platform does not provide them, measure them by hand the same day: on a one-minute or tick chart, write down in ticks the extreme against you and the extreme in your favor between your entry time and exit time. It takes thirty seconds per trade at the end of the session, and a screenshot with entry and exit marked lets you measure later without trusting your memory.
A worked example with ten ES trades
Ten hypothetical trades from the same setup, 1 ES each, always with a 6-point stop: 1R is $300. Trade 7 was stopped with one tick of slippage.
| # | Result (points) | MAE (points) | MFE (points) | Result (R) | MAE (R) | MFE (R) |
|---|---|---|---|---|---|---|
| 1 | +9 | 1.5 | 11 | +1.5 | 0.25 | 1.83 |
| 2 | −6 | 6 | 1.25 | −1 | 1 | 0.21 |
| 3 | +6 | 2 | 8 | +1 | 0.33 | 1.33 |
| 4 | +12 | 0.75 | 14 | +2 | 0.13 | 2.33 |
| 5 | −6 | 6 | 3 | −1 | 1 | 0.5 |
| 6 | +4.5 | 2.5 | 7.75 | +0.75 | 0.42 | 1.29 |
| 7 | −6.25 | 6.25 | 0.5 | −1.04 | 1.04 | 0.08 |
| 8 | +9 | 1.25 | 12 | +1.5 | 0.21 | 2 |
| 9 | −6 | 6 | 4.5 | −1 | 1 | 0.75 |
| 10 | +3 | 3.5 | 9 | +0.5 | 0.58 | 1.5 |
Six winners, four losers and +19.25 points: $962.50 gross. As P&L alone, that is all it says. The excursion columns say a lot more.
MAE: the risk no winner used
The classic way to read MAE is a scatter plot: MAE in R on the horizontal axis, result in R on the vertical one, winners and losers in different colors. In the example, the six winners sit on the left, with MAE between 0.13R and 0.58R, and the four losers form a column at 1R. Between 0.58R and 1R there is not a single winner: that is risk you paid for on every loser and that no good trade needed.
With a 4.5-point stop (0.75R, $225), all six winners would have survived and the losers would have lost 4.5 points instead of 6 (4.75 for the one with slippage): the total goes from $962.50 to $1,262.50. And if you keep dollar risk constant, the shorter stop turns into size: $300 buys 10 MES at 6 points and 13 at 4.5.
The reverse reading exists too: if many winners come close to the stop (0.8R or 0.9R), it is already tight. And if you suspect your losers reverse right after taking you out, MAE will not show it, because it stops counting at the exit: look at their charts.
MFE: how much of the move you keep
MFE measures how much the market offered and how much you kept. The most direct indicator is capture, the result divided by the MFE. The six winners made 43.5 points out of 61.75 points of MFE: 70%, from 86% on trade 4 down to 33% on trade 10. Stable capture points to rule-based exits; capture that swings a lot, to impulsive ones.
- Targets. All six winners reached at least 1.25R and only two reached 2R (trade 8 merely touched it, which does not guarantee a limit fill). With a fixed 2R target, four would never have hit it, and their result would depend on what price did next.
- What losers give back. Trades 5 and 9 were up 0.5R and 0.75R before getting stopped out; trade 9 gave back 10.5 points ($525), its ETD. If the pattern holds over a large sample, it deserves a management rule, tested on data first.
- Entry edge. The ratio of average MFE to average MAE, sometimes called the edge ratio, compares what the market offers in your favor with what it demands against you: here, 7.1 points against 3.58, almost 2.
Read that last number with care. The rigorous version measures excursions over a fixed number of bars after entry and normalizes them by volatility (ATR). The journal version is biased, because the stop caps MAE and your exit caps MFE: use it to compare your setups with each other, not as an absolute figure.
Pitfalls: sample size, mixed setups and intrabar data
- Small samples. Same rules as for expectancy: no verdict below 20 trades, broad conclusions from 30 to 50, fine-tuning at 100.
- Mixed setups. An opening-range breakout and a VWAP pullback have different MAE profiles; the combined cloud suggests a stop that fits neither. Analyze each setup on its own.
- Intrabar data. On a five-minute chart you cannot tell whether the bar's high or low happened before your entry or after your exit. Measure on one-minute or tick bars; bar-based backtests have the same problem.
- Order is invisible. On a winner, MAE and MFE do not say which came first; deciding on a break-even rule takes the chart, not just the two numbers.
- Moved stops. With the stop at break-even, your winners' MAE reflects your management, not the market; with a widened stop you get MAE above 1R. Tag those trades.
- Changing volatility. Three points of MAE in a quiet week are not three points in a CPI-and-FOMC week. Work in R and keep news days apart, as the economic calendar guide suggests.
Frequently asked questions
Should MAE and MFE be measured in ticks or in dollars?
For analysis, in ticks or in R. In dollars they depend on the number of contracts: a $100 MAE is huge on one MES and small on three ES. R is the best unit for comparing setups with different stops.
What MAE is normal for a winning trade?
There is no universal number: it depends on the setup and on volatility. What matters is the distribution of your winners within one setup. If nine out of ten never go more than 0.5R against you, a 1R stop has room to come in; if many get close to 0.9R, it is already tight.
Why does NinjaTrader show a different MAE from the one I measure on the chart?
The platform computes it from the prices it recorded while the position was open, against your average entry price; by hand on a bar chart you may count extremes of the entry bar from before your fill, or of the exit bar from after you were out. Always measure from the same source.
Next step
Take the last 30 trades of your main setup and log their MAE and MFE in R. Sort the winners by MAE and find the worst of the best 90%: that number, plus a margin, is the question your stop has to answer. If you decide to tighten it, recompute contracts with the position size calculator and review the guide to position sizing so your dollar risk does not change along with the stop.