Not supported · Patterns · 2026-09-26 (found) / 2026-10-10 (checked on other timeframes)
Is 'Gartley in an uptrend' a real edge?
What this means
'Gartley in an uptrend' looked strong on 15m (+0.32R vs −0.10R) but was found by scanning 1,895 pattern-and-condition combinations, and after correcting for that it is not significant (q = 0.77). On the independent 1h, 4h and daily logs it pointed both ways: slightly worse on 1h, clearly worse on 4h, better on daily, all on small samples. This looks like a chance find from a large search.
The explanation is our best reading; the data shows the result but not the cause directly.
By timeframe
| Timeframe | Verdict | n | Result |
|---|---|---|---|
| 15-minute | Not supported | 181 | +0.32R vs -0.10R, q=0.77 after correction |
| 1-hour | Not supported | 105 | diff -0.07 (-0.34, +0.20) |
| 4-hour | Not supported | 31 | diff -0.64 (-1.21, -0.03) |
| Daily | Not supported | 21 | diff +0.42 (-0.30, +1.08) |
- Hypothesis
- Gartley setups with price above the 200 MA and Pivot Point SuperTrend = BUY outperform other Gartleys.
- Method
- Found on 15m among 1,895 pattern x condition cells; then re-checked on the 1h/4h/1d TradingView logs (independent setups).
- Result
- 15m (discovery): +0.32R (n=180-181) vs -0.10R, CI (+0.13, +0.50), q = 0.77 after correction. 1h: +0.08 (n=105) vs +0.15, diff -0.07 (-0.34, +0.20); 4h: +0.02 (n=31) vs +0.65, diff -0.64 (-1.21, -0.03); 1d: +0.61 (n=21) vs +0.19, diff +0.42 (-0.30, +1.08).
- How much to trust it
- Medium: discovery is a multiple-testing artefact candidate; replication samples are small
- Did it change the live rule?
- No.
- Notes
- The formal interaction search reported 0 survivors (shuffled control 0.0). The study's own suggested test was a fresh 15m universe, which has not been run.
Scripts and data for this study will be linked here when the research code is published on GitHub.
Education and research only. Not financial advice. Backtests use today's index members (survivorship bias).