Revenge trading: a measurable state, not a character flaw
Trading to win back a loss shows up in your log as shorter gaps after losses, larger size and fewer planned setups. Measure those markers and decide your response before the session.
Start in plain English
Revenge trading means taking trades to win back a loss rather than because a setup appeared. Most traders recognise the feeling; fewer can point to it in their log. This lesson treats it as a state you can measure: what changes in your trading in the minutes after a loss. For a small-cap scalper, where a session can hold dozens of trades and a few seconds separate one decision from the next, that state can undo a week of careful work in an hour.
The core idea: revenge trading leaves measurable markers after losses: shorter gaps between trades, larger size than planned, and fewer trades that match a written setup. Once you can see those markers in your own numbers, you can decide your response before the session, when it is easy, instead of during it, when it is not. Calling it a character flaw does not help; counting it does.
Two paths from the same start
All trades, sizes and costs in this section are Illustrative.
The plan, written before the session. Risk $50 per trade. Daily loss limit −3R (−$150): at that point the session ends. After any losing trade, wait five minutes before the next entry. Size stays fixed within a session unless the plan says otherwise. Results below are in R against the plan's $50 risk, so a trade at double size that hits its stop counts as −2R (R multiple).
The start, the same for every path. By 10:12 ET, four trades: −1.0R, −1.1R, +0.4R, −1.3R. The session is at −3.0R (−$150), the daily limit.
Path AThe limit is respected
The trader closes the platform at 10:12 and writes a two-line review: what went wrong on the −1.3R trade (a stop filled three cents late) and whether each trade matched the plan.
Path BThe limit is ignored and the losses continue
The trader keeps going, two minutes after the last loss, at double size "to get it back in one trade". Each of the next four trades, at double size, returns −1.0R, −1.2R, +1.0R and −1.0R on its own risk, which is −2.0R, −2.4R, +2.0R and −2.0R against the plan's $50.
Session: −3.0 − 2.0 − 2.4 + 2.0 − 2.0 = −7.4R (−$370), about two and a half times the planned worst day.
Path CThe limit is ignored and the draw is kind
The same choices as Path B: no pause, double size. This time the next three trades return +2.0R, −1.0R and +0.7R on their own risk, which is +4.0R, −2.0R and +1.4R against the plan's $50.
Session: −3.0 + 4.0 − 2.0 + 1.4 = +0.4R (+$20).
Paths B and C made the same decisions; only the draw differed. Path C is the dangerous one to learn from, because it rewards a choice that Path B shows the cost of. Judged by outcome, Path C looks best; judged by decision, it matches Path B (outcome bias). The rest of this lesson turns those decisions into markers you can count, and into rules you can set in advance.
Measurable markers
Record these with every trade, in practice and in review of real trades:
| Marker | How to record it | What to compare |
|---|---|---|
| Gap after a loss | Minutes from the last losing exit to the next entry (min_since_loss in the practice log) |
Your typical gap between trades |
| Size vs. plan | Actual risk ÷ planned risk; at the planned stop this equals size_vs_plan, shares taken ÷ shares your sizing rule gave |
1.0 is the plan; look at the distribution after losses |
| Plan match | Was a setup tag written before entry? Within plan or outside plan? | Share of tagged trades after losses vs. at other times |
| Same-symbol re-entry | Re-entry in the stock that just stopped you out, within minutes | How often, and its net R |
| Consecutive losses | Count of losses in a row before the trade | Results after 0, 1, 2, 3+ prior losses |
None of these proves a trader was "on tilt". Each is an observation, and the question is whether your trades with these markers do worse after costs than your others.
Odds box
Read the table carefully. The after-loss trades differ in three ways at once: timing, size and setup quality. Any of the three could carry the difference. A useful next split is "after a loss, at planned size, tagged" against "after a loss, oversized or untagged". If the first group looks like your normal trades, the problem is the size and the untagged entries, not the timing alone.
Research points the same way. Kahneman and Tversky (1979) described people as more willing to take risk when facing losses. Thaler and Johnson (1990) described a "break-even effect": after a loss, gambles that offer a chance to get back to even become more attractive. Coval and Shumway (2005) found that professional futures traders with morning losses took on more risk in the afternoon. These studies describe tendencies in groups, not your log.
Pre-committed responses
Rules written before the session are easier to follow than decisions made after a loss. Write them as if-then rules:
- If a trade loses, then wait five minutes and write one line on whether it followed the plan before the next entry.
- If two trades in a row lose, then the next trade is at planned size and carries a setup tag written before entry.
- If the session reaches the daily loss limit, then the session ends, regardless of what is on the screen (Daily loss limits).
- If a stock stops you out, then no re-entry in it for 15 minutes.
The daily loss limit is the circuit breaker: it caps the damage when the other rules fail. Some brokers let you set a hard limit on the account side, which works even when willpower does not.
Observe vs. infer
A learner's end-of-day note: "Lost $370, the market was against me today." Separate the record from the story.
| What the log shows (observed) | What is inferred | Why it is not established |
|---|---|---|
| −$150 by 10:12, −$370 by 10:40 | "The market was against me" | The extra −$220 came after the limit, at double size; the market did not choose those trades |
| Four trades at 2× size after the limit | "I was trying to recover" | The log shows size and timing, not intent; the markers are enough without guessing the motive |
| Entries 2 minutes after losses | "I saw good setups" | Whether they were planned setups is visible only if tags were written before entry |
What shifts the odds
These are hypotheses to record, not findings; no FennEdge study measures revenge trading:
- Size of the triggering loss. A loss beyond −1R, such as a late stop fill, may be followed by stronger markers than an ordinary −1R.
- Consecutive losses. Two or three in a row may matter more than one.
- Distance from the daily limit. Trades taken just below the limit may differ from trades taken early in the day.
- Time of day. Losses late in the session leave less time to "make it back".
- Account-side limits. A hard limit at the broker may change behaviour near the limit.
Cost of being wrong
Using the plan above ($50 risk, −3R daily limit; Illustrative):
| Session | Result |
|---|---|
| Path A: limit respected | −3.0R (−$150.00) |
| Path B: limit ignored, losses continue | −7.4R (−$370.00) |
| Path C: limit ignored, draw is kind | +0.4R (+$20.00) |
| Path B, but a halt turns one 2× trade's −1R stop into −4R | −13.4R (−$670.00) |
Across many sessions, the expected cost of ignoring the limit depends on how often Path B happens compared with Path C. The limit's purpose is to remove the Path B tail, which is the part that can end a trading account (Risk of ruin).
How many trades before this means anything
After-loss trades are a subset of a subset. In the odds box, 70 of 400 trades are after-loss entries; with a per-trade standard deviation of about 1.2R, a rough 95% range for their average is about ±0.28R (Illustrative, normal approximation). Sessions matter too: 70 trades from 12 bad sessions are closer to 12 observations of "after a bad start" than to 70. Keep the markers running for at least 100 after-loss trades before treating the difference as measured; How many trades before your numbers mean anything covers why.
What this lesson does not claim
- It does not claim that every trade after a loss is a revenge trade, or that after-loss trades lose. The odds box is Illustrative.
- The markers describe behaviour, not motive or mood.
- A daily loss limit does not predict the rest of the session; it limits how bad the session can get.
- The research cited describes groups of traders and investors, not any individual.
Go deeper
- FennEdge first.
- If you trade live, Journal Review (Pro or Max, as of 2026-10-05) shows a versioned observation for entries within five minutes after the latest completed loss, for broker-synced trades. It is an observation, not an automatic judgment of emotion or a mistake. You can add personal tags, notes and a "Within plan" or "Outside plan" mark to each trade.
- Accepting an observation as a focus puts it on Desk as a personal reminder (at most two active at a time), with eight weeks of occurrence rates where the weeks are fully covered. Those rates do not measure whether the reminder caused any change.
- On Plus and above, the Trading Journal shows maximum consecutive losses and maximum drawdown in dollars for real trades (as of 2026-10-05). It holds broker-synced trades only, not practice or simulated entries.
- Level Up (Plus and above) is a private challenge scored on your Journal net P&L. Each level can carry a daily max loss, and a day at or below it is recorded as a breach, which gives you a check on your own daily limit. It does not track tags or entry timing, so keep process goals such as "no entries within five minutes of a loss" in your own log.
- Overtrading. Overtrading covers the trade-count side; after-loss clusters often show up in both.
- Daily limits. Daily loss limits on sizing the limit.
- Reviewing. Reviewing scalps on separating decision notes from outcomes; Building a playbook on writing the if-then rules into each setup's entry.
- Tilt indicators in your own log. The same markers, charted by week, show whether a rule changed anything. Changes that coincide with a new rule are not proof the rule caused them.
Exercises
Five exercises follow this lesson: three concept checks (markers vs. motives; what a Journal observation means; why the limit is set before the session), one scenario (reviewing a session that broke the limit and recovered), and one practice task. Exercises, answers and progress are available after signing in.
Practice task: measure your after-loss trades. This is the fifth exercise (ungraded). Before starting, write your if-then rules and daily loss limit. Over at least 40 sim or replay trades, record min_since_loss (blank if there was no earlier loss that session) and size_vs_plan in their own columns before each entry, and add consecutive_losses=… to conditions, in the FennEdge practice log: download the practice log (CSV), with a column guide. Mark session_stop as hit on the trade that reached your daily loss limit and after on any trade taken later that session. Mark plan_followed honestly and note each rule break in deviation_note. Then compare net R for trades with min_since_loss under 5 with all others, and count the after rows. Keep the log separate from the Trading Journal, which holds only real broker-synced trades.
Primary sources
- Kahneman, D. and Tversky, A. (1979). Prospect Theory: An Analysis of Decision under Risk. Econometrica 47(2).
- Thaler, R. H. and Johnson, E. J. (1990). Gambling with the House Money and Trying to Break Even: The Effects of Prior Outcomes on Risky Choice. Management Science 36(6).
- Coval, J. D. and Shumway, T. (2005). Do Behavioral Biases Affect Prices? The Journal of Finance 60(1).
Sources checked 2026-10-05. All sessions, trades, markers and costs in this lesson are fictional and Illustrative. This lesson is education, not a recommendation to buy or sell any security. Past patterns do not indicate future results.