What wick-to-body ratio makes a pin bar, and is the body the right thing to measure against?
Is the Pin Bar Profitable? 252 Backtests on USD/JPY
Finding
Moving the ratio changed almost nothing. At the default of 2, between 27.57% and 30.38% of all bars qualify. Across the seven steps from 1 to 5 the hourly chart lands between 51.40% and 51.72% in 2024 and falls from 50.83% to 50.00% in 2025, while the candidates drop from 2,898 to 806. Measuring the wick against the bar's range instead, a threshold of 70% beat the base rate in five cells of six (hourly +3.01 and +2.23, 4-hour +6.44 and +1.92). The default wick-to-body ratio of 2 beat it in four, by 0.83 to 4.18 points.
Key results
- Share of bars qualifying at the default ratio of 2
- 27.57 to 30.38%
- Effect of raising the ratio from 1 to 5
- Hourly 2024 stays in 51.40-51.72%; 2025 falls 50.83% to 50.00%. Candidates drop 2,898 to 806
- Wick at 70% of the range or more
- Beat the base rate in 5 of 6 cells (hourly +3.01/+2.23, 4-hour +6.44/+1.92)
- How much the two measures disagree
- Hourly 2025: 1,757 bars by ratio, 1,971 by a half-range wick, 1,506 by both
- Bars excluded for having no body
- 0.13 to 0.42%, with wicks taking 58.65 to 80.26% of their range
- Relation to the hammer
- 306 of the 937 long pin bars (32.66%) on the hourly chart in 2025; 50.69% against 54.58%
- Profitable in both years, exiting on an opposite pin bar
- 3 of 7 hourly, 6 of 7 on the 4-hour, 0 of 7 on the 15-minute
Scope
- Instrument: USD/JPY
- Period: 2024-01-01 to 2024-12-31 and 2025-01-01 to 2025-12-31
- Timeframes: 15-minute, 1-hour, 4-hour
- Wick ratios: 1, 1.5, 2, 2.5, 3, 4, 5
Method
The detection code is a rewrite of the backtester's own test, checked to match the engine's counts and rates in all twelve cells at the default ratio before anything was counted. The control is every bar of the same window, with the up-rate and down-rate blended to each group's own mix of long and short signals. Trades enter at the close and exit either on an opposite pin bar or after ten bars, at a fixed 0.3 pip spread, 0.1 lots and no slippage. 7 ratios x 2 exits plus one spread-free pass x 3 timeframes x 4 windows = 252 runs. Every figure is checked in research/pin-bar-verify.run.ts.
Limitations
- One pair over two years.
- Fills are at the close with a fixed spread and no slippage, so live results will differ.
- The wick-as-a-share-of-range measure is counted here; the backtester has no such condition.
- Bars whose two wicks are exactly equal point in no direction and are excluded.