Exhaustion Gap

BearishReversalIntermediate

The last gap of a long rally, filled within days. The buyers are spent.

Verdict: SkipBitcoin 4H: 0 trades, won —, avg trade — vs random day — (too few trades to judge). Backtest

Long rally
Exhaustion Gap: textbook shape, step by step

A long, steep rally. By now, almost everyone who wanted to buy is already in.

step 1/6

Trade rules

DirectionShort (sell)
EntryIn an advance of at least 10% (7% on Bitcoin) from the lowest low of the last 30 candles, a gap up to a new high, then a candle CLOSES back below the high of the candle before the gap within 5 candles.
Cancelled ifNo close back below the pre-gap high within 5 candles: the gap held, so there is no trade.
Target2x the spike, projected from the breakout
3 common mistakes
  • Shorting the gap itself: until it fills, it could be a runaway gap.
  • Putting the stop at the gap instead of above the new high.
  • Shorting a gap that only half fills: the close must get back below the pre-gap high.

Backtest

same rules every time · equal size · no fees · method
MarketTradesWonAvg tradeRandom day
Bitcoin · 4H · 2017-2026 · too few trades to judge0———
US large caps (50) · 1D · 2016-202625047%−0.15%−0.94%

Verdict: SkipLost money on average on US large caps. Bitcoin had too few signals to judge.

pattern −39%random −235%
US large caps (50) 1D: pattern −39% vs random day −235% over 250 trades

Real examples

Nvidia · 1D

Breakout Feb 24, 2020, exit Mar 16, 2020 (target)+25.20%
Pre-gap highLong rally5.7818 · Jan 6, 2020Exhaustion gap7.9080 · Feb 20, 2020Gap filled
Nvidia 1D, real candles from Jan 6, 2020

Apple · 1D

trapBreakout Jul 10, 2023, exit Jul 19, 2023 (stop)−3.25%
Pre-gap highLong rally170.42 · May 17, 2023Exhaustion gap194.48 · Jun 30, 2023Gap filled
Apple 1D, real candles from May 17, 2023