Revenge trading is the behavioural pattern of taking a new position SHORTLY AFTER a loss, often with a weaker setup or a larger size, attempting to recover the loss instead of accepting it. It compounds existing losses with low-quality entries and is the leading cause of single-day account-blowing drawdowns in self-directed trading.
The detector reads two signatures in the trade history without needing the trader to journal anything:
- Time-window signature — how soon after a losing close did the next entry land? Re-entries within minutes of a loser are statistically dominated by emotional decision-making rather than fresh setup recognition.
- Size-up signature — was the re-entry sized clearly above the trader's typical volume? The 'I'll go bigger to make it back faster' move has its own quantitative fingerprint.
A pair counts as a revenge instance when EITHER signal fires. The card emits a card only when the rate across ALL of the user's losing trades crosses the meaningful-only threshold.
30 closed trades over the period; 18 of them losers. Of the 18 losers, 8 were followed by a re-entry within minutes — and several of those re-entries were sized above the trader's median.
Severity bands on the revenge rate: - Mild — occasional revenge; one bad day per fortnight. Not yet structural but worth a journal note when it happens. - Moderate — established pattern. Roughly one in three losing trades triggers a chase; the fix is a written cool-off rule, not willpower. - Severe — dominant trading mode. The trader is trading emotion, not setups. Stop new entries until the rate falls back materially or the next month proves it was a one-off.
The Onyx-Engine assigns the revenge rate to one of these bands; the cutoff thresholds are TradeOnyx-internal calibration.
Why two signals, not just time? The time-window signal alone misses the trader who waits an hour to look 'disciplined' but then doubles size. The size signal alone misses the trader who immediately re-enters at a normal size — still emotional, still revenge. Together they catch both flavours.
The fix is mechanical, not psychological. Willpower fails reliably; rules don't. After every closed losing trade, write the close-time in the journal and commit to a fixed cool-off window before the next entry. The detector's revenge rate falls toward zero within 4-6 weeks of disciplined application.
Tier: Pro. Behavioural-discipline axis. Often co-fires with Stop-Loss-Drift (TRA-218) and Position-Sizing-Consistency (TRA-215) — same root cause (cannot accept the loss), three different symptoms.
How to read the card:
1. Hero (left) — the revenge rate as a percentage with a severity verdict. The single number to commit to memory week-over-week. 2. Two MicroStats — Losing trades (the denominator the rate reads against), Avg size on revenge (the multiplier of your typical volume on the chase trades). 3. Hint line — actionable: the cool-off-window discipline rule.
The Revenge-Trading → Cool-Off-Window pairing. This card's only meaningful intervention is a written cool-off rule between a losing close and the next entry. Most users find 30-60 minutes is enough. The detector measures whether you actually held the rule across the next month — the rate either falls toward zero or it doesn't.
Re-look frequency: weekly. Daily fluctuations are noise; the behavioural pattern stabilises after ~10-15 losing trades.
Tier: Pro.