Socius NewsEvery trader has met it. A stop gets hit. Two minutes later there is a new position, bigger than the last one, in the same instrument, in the same direction. There was no setup. There was a feeling.
Most traders call it revenge trading and treat it as a character flaw. It is not a character flaw. It is a measurable pattern, and the research on it is thirty years old.
Why does a loss change the next decision?
In 1979, Daniel Kahneman and Amos Tversky published prospect theory. One of its findings is now common knowledge: a loss hurts roughly twice as much as an equivalent gain feels good. The less-quoted finding matters more for traders. People are risk-averse when they are ahead and risk-seeking when they are behind. Put someone in a loss and they will accept a worse gamble to get back to zero.
A trading account is a machine for putting you behind several times a day.
Has anyone measured it on real traders?
Yes, on professionals. Joshua Coval and Tyler Shumway studied the Chicago Board of Trade T-bond futures pit in 1998, using more than five million transactions and 426 local traders who each had at least 100 trading days. They split each day in two and asked a simple question: does a losing morning change the afternoon?
It does. Traders who lost money in the morning had a 31.2 percent chance of taking above-average risk in the afternoon, against 27 percent for traders who were up. The losers placed more trades, placed larger trades, and accumulated more inventory. And the prices they set in the afternoon reversed faster than the prices set by the winners, which is what you would expect if the rest of the pit had learned to trade against them.
These were full-time floor traders, in a room, with their own capital. Loss chasing did not come from inexperience. It came from being human.
Chance of taking above-average risk in the afternoon
CBOT T-bond futures pit, 1998 — 426 local traders, 5 million+ transactions
Source: Coval & Shumway, The Journal of Finance (2005)
What does it cost?
Brad Barber and Terrance Odean followed 66,465 households at a large discount broker between 1991 and 1996. The 20 percent of households that traded the most earned 11.4 percent a year. The market returned 17.9 percent. The average household turned over 75 percent of its portfolio every year. The authors' conclusion is the title of the paper: trading is hazardous to your wealth.
11.4%
Annual return of the 20% most active households
1991–1996, 66,465 households
17.9%
Market return over the same period
75%
Average annual portfolio turnover
Overtrading is the general case. Revenge trading is the acute one. It is the moment when turnover spikes, size spikes, and the quality of the decision collapses, all inside the same ten minutes.
What does revenge trading look like in a journal?
A feeling leaves no trace. A trade does. Once every execution is time-stamped and sized, revenge trading stops being a mood and becomes a signature with five components. None of them requires you to remember how you felt.
1. Time-to-next-trade collapses. Look at the gap between a losing exit and the next entry. Most traders have a normal rhythm, twenty minutes, an hour, a session. After a loss, the gap shrinks. When your median gap after a loss is a fraction of your median gap after a win, you are not waiting for a setup. You are waiting for the order ticket to load.
2. Size jumps. Compare the size of each trade with the size of the trade before it. A disciplined trader's size is boring: it follows a rule and moves slowly. A revenge trade is typically larger than the loss it follows, because the goal is not to be right, it is to be back to even in one move.
3. Same instrument, same direction. The revenge trade rarely goes looking for a new idea. It re-enters the trade that just failed, because the loss feels like a pricing error by the market rather than a decision error by the trader.
4. The session stretches. Traders who are down keep trading past the hour they normally stop. Coval and Shumway saw it in the afternoon session of the pit. In a retail journal it shows up as a cluster of trades after your usual cut-off, and those trades carry a worse average result than the ones inside your window.
5. Losses arrive in chains. Filter your history for trades taken within a few minutes of a loss. If their expectancy is materially worse than your overall expectancy, the chain is the problem, not the first link. The first loss was a normal cost of doing business. The second and third were paid for by the first.
What your journal would show
Here is what this looks like on illustrative demo data for a trader with 340 trades over three months.
The overall expectancy is positive: 0.18 R per trade. Filter for trades entered within five minutes of a losing exit and the picture changes. There are 41 of them, 12 percent of the total. Their expectancy is minus 0.42 R. Their average size is 1.7 times the trader's median. Thirty-three of the 41 are in the same instrument as the loss they follow.
Remove those 41 trades and the account's expectancy rises from 0.18 R to 0.27 R. Nothing else changed. The trader did not need a new strategy. They needed a five-minute rule.
0.18 R
Expectancy across all 340 trades
illustrative demo data
−0.42 R
Expectancy of the 41 trades entered within 5 minutes of a loss
1.7×
Their average size against the trader's median
0.27 R
Expectancy without those 41 trades
That is the point of a journal that reads your executions rather than your notes. You cannot see the feeling. You can see the gap, the size, the instrument and the hour, and together they name it for you.
What can you actually do about it?
The research is clear that willpower is a poor defence, because the bias is strongest exactly when you are least able to notice it. Rules that live outside your head work better.
A cooling-off rule: no new entry for a fixed period after a loss. Fifteen minutes is enough for most intraday traders to reread their own plan.
A size ceiling tied to the previous trade: the next position may not be larger than the last one after a loss.
A daily stop, in R, that closes the platform rather than the position. Two or three R is a common choice among prop-firm traders, and it is the rule that prop firms enforce for a reason.
And a weekly review of the five signatures above. Not to feel bad about them, but to see whether the rule you set last week actually changed the numbers. If the gap after losses widened and the size stopped jumping, the rule works. If not, the rule is not the problem, the enforcement is.
“You will still lose trades. You will stop paying for them twice.”
Trading involves risk of loss. Socius Trades is an analytics tool, not investment advice.