# trending-1005 results, amended 6 October 2026 (B, random control v3)

Generated by `scripts/research/trending-1005-amend.mjs` under `prereg-amendment-2026-10-06b.md` (random control v3). The original `RESULTS.md` and `results.json` are unchanged, and so are the amendment-A (v2) files. Hypothetical historical backtests on cached data to March 2025.

Published family: 722 tests, 0 passed, smallest q 0.1444.

| cell | role | label | CAGR | benchmark | random copies beaten, v1 -> v2 -> v3 | q (all tests) | gate failures |
|---|---|---|---|---|---|---|---|
| tr5-orb-utc-1h-2r-cutoff | best | 1-hour opening range, entries at the 02:00 or 03:00 UTC close only, target 2x the range width, stop on a close past the other edge | -58.3% | 84.7% | 0.5% -> 0.0% -> 0.0% | 1.000 | q, random, halves, net<=0 |
| tr5-orb-utc-1h-eod | variant | 1-hour opening range, stop on a close past the other edge, out at the 00:00 UTC close | -65.1% | 84.7% | 11.5% -> 17.0% -> 23.0% | 1.000 | q, random, halves, net<=0 |
| s29-h-utc-orb | variant | 4-hour opening range, no stop, held to the end of the day (earlier run) | -50.2% | 84.7% | 87.5% -> 93.0% -> 85.5% | 1.000 | q, random, halves, net<=0 |
| tr5-ict-asia-sweep-be | best | Asia range swept in London, wick stop, break-even after 1R, 2R target, out at the 21:00 UTC close | -48.1% | 84.7% | 0.0% -> 0.0% -> 0.0% | 1.000 | q, random, halves, net<=0 |
| tr5-ict-asia-sweep-2r | variant | Same sweep, wick stop, 2R target, no break-even step | -48.5% | 84.7% | 0.0% -> 0.0% -> 0.0% | 1.000 | q, random, halves, net<=0 |
| tr5-ict-asia-sweep-hold | variant | Same sweep, wick stop, no target (archived as 'no stop or target'; it kept the wick stop) | -47.7% | 84.7% | 0.5% -> 0.0% -> 0.5% | 1.000 | q, random, halves, net<=0 |
| s29-etf-turtle-soup | variant | Daily cousin: failed 20-day breakdown on ETFs (turtle soup, earlier run) | -0.4% | 7.0% | 12.0% -> 11.0% -> 9.5% | 1.000 | q, random, halves, net<=0 |
| tr5-vwap-ema9-trend | best | Long above the session VWAP (hourly OHLCV estimate) and the 9 EMA, short below both | -99.1% | 84.7% | 0.0% -> 0.0% -> 1.0% | 1.000 | q, random, halves, net<=0 |
| tr5-vwap-reclaim-ema9-exit | variant | VWAP reclaim (hourly OHLCV estimate), long only, exit on a close below the 9 EMA, out by the 00:00 UTC close | -73.5% | 84.7% | 20.0% -> 62.0% -> 68.0% | 1.000 | q, random, halves, net<=0 |
| social-ta-vwap-bounce | variant | Bounce off a rolling 20-hour VWAP (earlier run) | -49.8% | 84.7% | 1.5% -> 6.0% -> 3.5% | 1.000 | q, random, halves, net<=0 |
| tr5-ema921-1h-both | best | 9/21 EMA cross on hourly crypto, long and short | -51.3% | 84.7% | 19.5% -> 97.5% -> 95.5% | 1.000 | q, halves, net<=0 |
| h3-hourly-ema-9-21-cross | variant | Same cross, long only, above the 200 EMA (earlier run) | 21.8% | 84.7% | 100.0% -> 100.0% -> 100.0% | 1.000 | q, halves |
| tr5-ema921-daily-stocks | variant | 9/21 EMA cross on daily US large caps, long only | 9.1% | 18.0% | 4.0% -> 4.0% -> 29.5% | 1.000 | q, random, halves |
| tr5-crypto-200dma-30d | best | Long after 30 closes above the 200-day average, out on a close below | 14.2% | 62.8% | 24.5% -> 22.5% -> 54.0% | 1.000 | q, random, halves |
| tr5-crypto-golden-cross | variant | Golden cross: long while the 50-day is above the 200-day | 14.8% | 62.8% | 3.5% -> 3.0% -> 38.0% | 1.000 | q, random, halves |
| tr5-crypto-uptober | variant | Uptober: hold through October only | 9.2% | 62.8% | 79.0% -> 78.0% -> 70.0% | 1.000 | q, random, halves |
| tr5a-ict-asia-sweep-nostop | variant | Same sweep entry, no stop and no target, out at the 21:00 UTC close | -43.3% | 84.7% | new -> 5.5% -> 15.0% | 1.000 | q, random, halves, net<=0 |
| tr5a-crypto-200dma-30d-matched | matched | 200-day rule, rule and basket both scored from the first day the 200-day average exists | 20.8% | 32.1% | new -> 43.0% -> 65.5% | 1.000 | q, random, halves |
| tr5a-crypto-golden-cross-matched | matched | Golden cross, rule and basket both scored from the first day the 200-day average exists | 15.2% | 32.1% | new -> 2.5% -> 40.0% | 1.000 | q, random, halves |
