The comparison this page makes, and the one it refuses to make
The obvious way to measure revenge trading is to take every trade you opened shortly after a loss and compare it against everything else. That comparison is wrong, and wrong in a way that flatters the conclusion.
Trades that follow a loss are already unlike your average trade. Something just went against you — the market may be in a state that hurt you a moment ago and is about to do it again. If those trades come out worse, you have not learned that hurrying is expensive. You have learned that losing tends to arrive in clusters, which you already knew.
So the baseline here is narrower: trades that also followed a loss, but where you waited. Both groups start from the same place. The only thing that differs is the pause.
What counts as fast
Your own median pause, not a number this page brought with it.
Every gap between one trade closing and the next one opening is measured, and the middle value becomes the threshold. If you scalp, that is likely minutes. If you swing, it may be a day. Either way, "fast" means fast for you, which is the only version of the word that survives contact with a different trading style.
pause = next entry − previous exit
fast = pause below your median pause
group A = trades after a loss, fast
group B = trades after a loss, waited
gap = average A − average B
carried = gap × number of trades in A
Trades that opened while the previous one was still running get no pause at all, and are left out of the comparison rather than counted as instant. Two positions overlapping is not the same behaviour as jumping back in.
Why order is taken from entries, not exits
Trades are sorted by when they opened, because that is when the decision was made. Sorting by exit would scramble the order whenever a longer position sits across several short ones, and the question here is specifically about what you decided to do next.
The bonus row: fast after a win
The page also shows what your fast entries after a winning trade returned, when there are enough of them.
It is there as a control. If your hurried trades are worse after losses but fine after wins, the pattern really is about the loss. If they are worse in both cases, what you have is not revenge trading — it is that rushing costs you regardless of what came before, which is a different problem with a different fix.
What it does not claim
That hurrying caused the worse result. The statement records the pause and the outcome; it does not record your state of mind, and the direction of the arrow is not in the file.
There is also a real alternative explanation worth stating: the trades you take quickly may simply be the setups that appear quickly, and those may be worse setups regardless of the loss before them. This page cannot separate those two. What it can do is tell you the size of the effect, so you know whether it deserves any of your attention at all.
What happens to your file
It is read inside the browser tab by the same parsers the app uses for MT4, MT5, cTrader, IBKR, Schwab, Trading 212, Tradovate, KuCoin and Coinbase, plus a general CSV reader. No upload, no account, no server in the path.
Frequently asked
Are the figures before or after costs?
After. Commission and swap are subtracted from every trade before any average is taken. A page about the cost of a habit that ignored the most reliable cost in trading would not be worth reading.
Why is the headline sometimes missing?
Because either group had fewer than eight trades. An average over three trades describes those three trades, not your behaviour, and printing it with confidence would be the failure this whole section exists to avoid. The counts stay visible, because counting is honest at any sample size.
I never trade twice in one session — will this work?
Yes, but it may have little to say. The measurement needs trades that followed other trades closely enough for a pause to be meaningful. If your median pause is measured in days, "fast" simply means a shorter wait, and the effect it finds is usually small.