MFE, for maximum favorable excursion, is the highest unrealised gain a trade reached before being closed. It is the only statistic describing what happened during the trade, where all the others know only its final result.
What a trade goes through before closing
Contents
A trade statement keeps only two prices: the entry and the exit. Everything that happened in between has vanished. MFE restores part of that path: the most favourable point price reached while the position was open.
The point is being able to judge an exit, which no other statistic allows. A trade closed at +40 pips does not have the same story depending on whether it went to +45 or to +200: in the first case the exit was excellent, in the second the market offered five times more. The final result is identical, the decision is not.
It is also a measure with a technical cost, and that is why it is rare. Computing it requires the price history between each trade's open and close, data no broker statement provides. It has to be fetched elsewhere, trade by trade, which explains why most journals do not display it.
Trade MFE = |most favourable price reached − entry price| — Average MFE = mean across all trades
MFE is always positive or zero: it is a distance, not a result. A losing trade also has an MFE, often close to zero.
It is measured between open and close, never after. What price did once you were out no longer concerns you.
Its precision depends on the candles used. An MFE computed on hourly candles misses intra-hour spikes, so it is systematically understated compared with a minute-level calculation.
| Gain banked against MFE | What it describes |
|---|---|
| over 80% | Exits very close to the best moment. Rare, and often a sign of modest targets. |
| 50 to 80% | The normal zone. You capture most of the move you caught. |
| 30 to 50% | Recurring early exits, or targets systematically too close. |
| under 30% | The market offered three times what you took. The problem is the exit, not the entry. |
| MFE close to the gain on losers | Normal: a trade that never went your way has a near-zero MFE. |
The first row is surprising, and it matters: capturing 90% of MFE is not a good result. It almost always means your targets are so close that price had no time to go further. An excellent ratio can therefore describe a trader who systematically cuts too early, and the only way to settle it is to look at the absolute size of the gains too.
The second point is that MFE does not read the same way on losers. A trade that immediately goes against you has a near-zero MFE, which is mechanical and says nothing. The measure only makes sense on winners, or on trades that were at least positive at some point.
The market offered on average $312 per winning trade and you took $138. The gap is not a loss, since that gain was never secured, but it quantifies room for improvement appearing in no other statistic. The same journal without MFE would simply have shown 'average gain: $138', which says nothing about the exit.
This is the main contribution, and it is decisive because the two problems are not fixed the same way at all. A high average MFE with a low average gain says your entries find the move and your exits give it back: there is nothing to change in trade selection.
Conversely, a low average MFE says the market never moved much in your favour, and there the entry is indeed at fault. Without MFE, these two situations produce the same average gain and are completely indistinguishable.
If your winners' average MFE is $312 and you exit at $138, a target set further out is defensible with a figure. It is not a guarantee, since moving the target also changes how many trades reach it, but it is a measured starting point.
Verification then happens in live conditions: move the target, let thirty trades go by, compare. MFE indicates the direction, it does not replace the test.
The first flaw is hindsight bias, and it is powerful. MFE shows what the market offered, and the mind reads it as what you should have taken. But that point was only identifiable after the fact: at the moment price was there, nothing said it was the top. Confusing the two turns a useful statistic into a regret machine.
The second is data granularity. An MFE computed on hourly candles misses spikes of a few minutes, so it understates; an MFE computed on ticks captures them all, including spikes no order would have been filled on. Two honest tools give two different figures for the same trade, and neither writes down its resolution.
The third is reading it alone. A gap between MFE and banked gain can come from an exit that was too early, but also from a perfectly reasonable target the market overshot one time in ten. Without the other half of the calculation, the number of times the further target would not have been hit, you optimise on one branch only.
The classic mistake. Reading MFE as what you should have taken. A trade's best point was not identifiable at the moment price was there: nothing distinguished it from an ordinary pause. MFE measures an offer from the market, it does not describe a decision that was available.
| Indicator | Formula | What it measures | When to use it |
|---|---|---|---|
| MFE | best point reached − entry | What the market offered at best | Judging the timing of an exit |
| MAE | worst point reached − entry | What the trade went through against you | Judging stop placement |
| Exit efficiency | gain banked ÷ sum of MFE | The share of the move captured, in % | Summing all of that into one number |
| Average win | mean of winning trades | The result, without the path | Saying what a winner returns |
A trade that went against you immediately has a near-zero MFE, which is mechanical and drags the average down without teaching anything. The question 'am I exiting too early' only concerns trades that went your way, and it is on those alone that the comparison makes sense.
An MFE computed on hourly candles systematically understates, and a tick-level MFE captures unrealistic spikes. Knowing what your tool works on changes the interpretation of the same figure, and almost none writes it down.
MFE says what the market offered, not what you would have obtained. Moving the target also changes how many trades reach it, and the net effect can be negative. Thirty trades with the new target settle it, MFE alone does not.
MFE stands for maximum favorable excursion. It is the highest unrealised gain a trade reached between opening and closing, regardless of the final result. A trade closed at +40 pips that had gone to +200 has an MFE of 200.
MFE is the best point reached in your favour, MAE the worst point reached against you. It is the same measure in two directions, computed exactly the same way. What changes is the decision it informs: MFE speaks about your exit, MAE about your stop placement.
By taking the distance between your entry price and the most favourable price reached while the position was open. That assumes having the price history over that period, which no broker statement contains: it has to be downloaded separately, trade by trade.
Not on its own. It has to be compared with the gain actually banked: if the market offered $312 on average and you take $138, the gap is real. But moving your target further also changes how many trades reach it, and the net effect can be negative. MFE indicates a direction, it does not replace a test over thirty trades.
Because it is not in the broker's data. A statement contains the entry price and the exit price, never the path between them. Displaying MFE forces you to fetch price history for every trade, store it and recompute it: that is a whole job, for one extra column.
In Tradoshi, excursions do not come from your broker, and that is worth saying because it explains everything else: no statement contains a trade's path. The app recomputes MFE from downloaded historical candles, trade by trade, with a progress bar and a cache so it does not start over. Some trades have no data available: they are dropped from the calculation rather than counted as zero.
Once computed, average MFE (best moment) becomes a metric in the Reports screen like any other, so it can be crossed with any dimension: by symbol, by weekday, by entry hour, by tag. That crossing is what really answers the question, because people rarely exit too early everywhere in the same way.
To go deeper on this: read the full guide. Every term is gathered in the glossary.