Rule Collider

Backtests · DJI30 · price_vs_50d_high_atr · long

Does price_vs_50d_high_atr below -37.615 work on DJI30? 2023 to 2026 backtest

DJI30 long when Close vs 50-day high (ATR) is below the threshold, tested at five thresholds taken from the feature’s own distribution on DJI30 over this window: its 10th, 25th, 50th, 75th and 90th percentiles. Window 2023-09-30 to 2026-09-30, the last 30 percent held out of sample; stop 2.5 ATR, target 3.5 R, held overnight, max spread 50 pips, one-minute confirmation 0.2 ATR, no hour filter. Fills and spread as measured on a demo fleet (how a result is judged). The direction and the exit are the ones the research catalogue carried for this pair; the thresholds are not.

Five thresholds, one rule shape

PercentileConditionIn sample R / tradesOut of sample R / tradestTrades to confirmLast quarter R / tradesVerdict
p10price_vs_50d_high_atr < -146.4-98.0 R / 543-29.7 R / 199-1.1695-8.5 R / 95No edge shown
p25price_vs_50d_high_atr < -78.142-114.2 R / 995-50.8 R / 479-1.21,355-51.7 R / 212No edge shown
p50price_vs_50d_high_atr < -37.615-196.6 R / 1707-82.9 R / 847-1.51,490-45.7 R / 335No edge shown
p75price_vs_50d_high_atr < -15.421-261.9 R / 2316-134.9 R / 1140-2.2948-64.2 R / 390Active, net negative
p90price_vs_50d_high_atr < -6.179-321.8 R / 2798-165.9 R / 1258-2.6739-67.4 R / 416Active, net negative

R is the risk on one trade. The out-of-sample split carries the verdict: below 100 trades it is “no edge shown” whatever the t; a positive mean with t below 2 is “net positive, not yet distinguishable from chance”; the last column is how many trades that mean would need to reach t = 2.

Each threshold in full

price_vs_50d_high_atr < -146.4 (p10)

Honesty panelNo edge shown

No edge shown out of sample: -29.7 R over 199 trades is -0.149 R per trade, within what chance produces (t = -1.1); about 695 trades would be needed to confirm this mean at t = -2, shown: 199.

In sample
2023-09-30 to 2025-11-05
-98.0 R
543 trades · 25% wins
-0.180 R per trade · t = -2.2, beyond chance
about 465 trades to confirm this mean (-0.180 R per trade) at t = -2 · shown: 543
spread 0.154 R per trade
Out of sample
2025-11-05 to 2026-09-30
-29.7 R
199 trades · 25% wins
-0.149 R per trade · t = -1.1, within chance
about 695 trades to confirm this mean (-0.149 R per trade) at t = -2 · shown: 199
spread 0.140 R per trade
Last quarter
last 91 days of the window
-8.5 R
95 trades · 27% wins
-0.089 R per trade · t = -0.4, within chance
about 2,058 trades to confirm this mean (-0.089 R per trade) at t = -2 · shown: 95
spread 0.177 R per trade
Costs. Spread 0.150 R per trade at 1.25 times the bar’s minimum. Fill offset 0.14 stop-distances on every entry, measured on 931 demo fills. At 1.5 times the bar’s minimum the out-of-sample figure would read -35.2 R (-0.177 R per trade over 199 trades).

Entry: stop order 2 pips past the signal close, cancelled after 4 bars; 751 signals placed, 742 filled, 9 expired; trigger state, an entry on every bar the conditions hold while flat.

Exit: stop 2.5 ATR (mean 343.9 pips, 2.50 ATR); target 3.5 R; exits 75% at the stop, 25% at the target, 0% at the day end.

Filters: signals on every weekday; no entries 22:00 to 01:00 broker time (15,499 bars); max spread 50 pips (1,640 bars over it); M1 confirmation 0.2 atr (287 signals rejected).

Monthly R, price_vs_50d_high_atr < -146.4 on DJI30One bar per month: the R earned that month; the held-out months in the full colour.2023-10: -28.2 R over 72 trades2023-102023-11: -0.4 R over 23 trades2024-02: +3.0 R over 1 trades2024-03: +3.1 R over 1 trades2024-04: -3.1 R over 47 trades2024-05: -5.6 R over 44 trades2024-06: -21.7 R over 52 trades2024-062024-07: -7.2 R over 24 trades2024-08: -1.5 R over 15 trades2024-09: +3.1 R over 1 trades2024-10: +1.2 R over 9 trades2024-11: -6.1 R over 8 trades2024-12: -28.5 R over 35 trades2024-122025-01: +24.1 R over 38 trades2025-02: +1.6 R over 9 trades2025-03: -21.0 R over 64 trades2025-04: +0.9 R over 58 trades2025-05: -10.3 R over 41 trades2025-08: -1.4 R over 1 trades2025-082025-11: -9.6 R over 11 trades2025-12: -1.4 R over 1 trades2026-02: -4.0 R over 3 trades2026-03: -11.6 R over 50 trades2026-04: +5.4 R over 39 trades2026-07: +3.2 R over 1 trades2026-072026-08: -0.8 R over 23 trades2026-09: -10.8 R over 71 trades+24 R-28 R

price_vs_50d_high_atr < -78.142 (p25)

Honesty panelNo edge shown

No edge shown out of sample: -50.8 R over 479 trades is -0.106 R per trade, within what chance produces (t = -1.2); about 1,355 trades would be needed to confirm this mean at t = -2, shown: 479.

In sample
2023-09-30 to 2025-11-05
-114.2 R
995 trades · 26% wins
-0.115 R per trade · t = -1.9, within chance
about 1,150 trades to confirm this mean (-0.115 R per trade) at t = -2 · shown: 995
spread 0.119 R per trade
Out of sample
2025-11-05 to 2026-09-30
-50.8 R
479 trades · 27% wins
-0.106 R per trade · t = -1.2, within chance
about 1,355 trades to confirm this mean (-0.106 R per trade) at t = -2 · shown: 479
spread 0.106 R per trade
Last quarter
last 91 days of the window
-51.7 R
212 trades · 24% wins
-0.244 R per trade · t = -1.9, within chance
about 224 trades to confirm this mean (-0.244 R per trade) at t = -2 · shown: 212
spread 0.115 R per trade
Costs. Spread 0.115 R per trade at 1.25 times the bar’s minimum (more than the trades made before spread). Fill offset 0.14 stop-distances on every entry, measured on 931 demo fills. At 1.5 times the bar’s minimum the out-of-sample figure would read -61.0 R (-0.127 R per trade over 479 trades).

Entry: stop order 2 pips past the signal close, cancelled after 4 bars; 1,485 signals placed, 1,474 filled, 11 expired; trigger state, an entry on every bar the conditions hold while flat.

Exit: stop 2.5 ATR (mean 488.6 pips, 2.50 ATR); target 3.5 R; exits 73% at the stop, 24% at the target, 3% at the day end.

Filters: signals on every weekday; no entries 22:00 to 01:00 broker time (14,223 bars); max spread 50 pips (1,388 bars over it); M1 confirmation 0.2 atr (544 signals rejected).

Monthly R, price_vs_50d_high_atr < -78.142 on DJI30One bar per month: the R earned that month; the held-out months in the full colour.2023-10: +1.4 R over 106 trades2023-102023-11: +18.6 R over 34 trades2023-12: +12.0 R over 4 trades2024-01: -8.2 R over 9 trades2024-02: +0.9 R over 12 trades2024-03: -5.6 R over 20 trades2024-04: -10.7 R over 69 trades2024-042024-05: -11.9 R over 53 trades2024-06: -33.4 R over 72 trades2024-07: -1.6 R over 53 trades2024-08: +2.4 R over 41 trades2024-09: -4.0 R over 17 trades2024-10: -20.7 R over 33 trades2024-102024-11: +5.8 R over 24 trades2024-12: -38.8 R over 71 trades2025-01: +24.3 R over 55 trades2025-02: -1.5 R over 36 trades2025-03: -14.2 R over 68 trades2025-04: -16.5 R over 108 trades2025-042025-05: -19.4 R over 63 trades2025-06: +8.1 R over 8 trades2025-07: +1.7 R over 9 trades2025-08: +0.3 R over 21 trades2025-09: -1.0 R over 4 trades2025-10: -2.0 R over 5 trades2025-102025-11: +0.6 R over 29 trades2025-12: -7.0 R over 32 trades2026-01: -8.8 R over 10 trades2026-02: +17.1 R over 22 trades2026-03: -21.3 R over 97 trades2026-04: +19.7 R over 74 trades2026-042026-05: +1.9 R over 2 trades2026-06: -1.2 R over 1 trades2026-07: -5.2 R over 25 trades2026-08: -8.9 R over 57 trades2026-09: -37.5 R over 130 trades+24 R-39 R

price_vs_50d_high_atr < -37.615 (p50)

Honesty panelNo edge shown

No edge shown out of sample: -82.9 R over 847 trades is -0.098 R per trade, within what chance produces (t = -1.5); about 1,490 trades would be needed to confirm this mean at t = -2, shown: 847.

In sample
2023-09-30 to 2025-11-05
-196.6 R
1707 trades · 27% wins
-0.115 R per trade · t = -2.5, beyond chance
about 1,075 trades to confirm this mean (-0.115 R per trade) at t = -2 · shown: 1707
spread 0.097 R per trade
Out of sample
2025-11-05 to 2026-09-30
-82.9 R
847 trades · 28% wins
-0.098 R per trade · t = -1.5, within chance
about 1,490 trades to confirm this mean (-0.098 R per trade) at t = -2 · shown: 847
spread 0.082 R per trade
Last quarter
last 91 days of the window
-45.7 R
335 trades · 28% wins
-0.137 R per trade · t = -1.4, within chance
about 704 trades to confirm this mean (-0.137 R per trade) at t = -2 · shown: 335
spread 0.087 R per trade
Costs. Spread 0.092 R per trade at 1.25 times the bar’s minimum. Fill offset 0.14 stop-distances on every entry, measured on 931 demo fills. At 1.5 times the bar’s minimum the out-of-sample figure would read -96.8 R (-0.114 R per trade over 847 trades).

Entry: stop order 2 pips past the signal close, cancelled after 4 bars; 2,571 signals placed, 2,554 filled, 17 expired; trigger state, an entry on every bar the conditions hold while flat.

Exit: stop 2.5 ATR (mean 624.0 pips, 2.50 ATR); target 3.5 R; exits 71% at the stop, 22% at the target, 7% at the day end.

Filters: signals on every weekday; no entries 22:00 to 01:00 broker time (10,624 bars); max spread 50 pips (1,246 bars over it); M1 confirmation 0.2 atr (913 signals rejected).

Monthly R, price_vs_50d_high_atr < -37.615 on DJI30One bar per month: the R earned that month; the held-out months in the full colour.2023-10: -14.4 R over 131 trades2023-102023-11: +18.7 R over 43 trades2023-12: +4.3 R over 10 trades2024-01: -24.7 R over 35 trades2024-02: -9.6 R over 51 trades2024-03: -7.0 R over 58 trades2024-04: -35.9 R over 111 trades2024-042024-05: +10.7 R over 85 trades2024-06: -19.7 R over 107 trades2024-07: +13.1 R over 68 trades2024-08: +16.6 R over 73 trades2024-09: +1.8 R over 31 trades2024-10: -45.7 R over 72 trades2024-102024-11: -2.4 R over 59 trades2024-12: -49.3 R over 97 trades2025-01: +12.7 R over 97 trades2025-02: -16.2 R over 77 trades2025-03: -20.0 R over 67 trades2025-04: -26.8 R over 126 trades2025-042025-05: -19.7 R over 86 trades2025-06: +10.8 R over 43 trades2025-07: +16.6 R over 37 trades2025-08: -4.4 R over 52 trades2025-09: +7.6 R over 29 trades2025-10: -7.2 R over 53 trades2025-102025-11: -21.6 R over 71 trades2025-12: -11.1 R over 69 trades2026-01: -5.2 R over 54 trades2026-02: +4.2 R over 55 trades2026-03: -36.5 R over 124 trades2026-04: +31.4 R over 93 trades2026-042026-05: -1.0 R over 19 trades2026-06: -4.0 R over 36 trades2026-07: -1.9 R over 83 trades2026-08: -5.9 R over 104 trades2026-09: -38.0 R over 148 trades+31 R-49 R

price_vs_50d_high_atr < -15.421 (p75)

Honesty panelActive, net negative

Traded 1140 times out of sample, net negative beyond chance (-134.9 R, -0.118 R per trade, t = -2.2); still active in the last quarter (-64.2 R over 390 trades).

In sample
2023-09-30 to 2025-11-05
-261.9 R
2316 trades · 29% wins
-0.113 R per trade · t = -3.0, beyond chance
about 1,049 trades to confirm this mean (-0.113 R per trade) at t = -2 · shown: 2316
spread 0.083 R per trade
Out of sample
2025-11-05 to 2026-09-30
-134.9 R
1140 trades · 28% wins
-0.118 R per trade · t = -2.2, beyond chance
about 948 trades to confirm this mean (-0.118 R per trade) at t = -2 · shown: 1140
spread 0.068 R per trade
Last quarter
last 91 days of the window
-64.2 R
390 trades · 27% wins
-0.165 R per trade · t = -1.8, within chance
about 462 trades to confirm this mean (-0.165 R per trade) at t = -2 · shown: 390
spread 0.075 R per trade
Costs. Spread 0.078 R per trade at 1.25 times the bar’s minimum. Fill offset 0.14 stop-distances on every entry, measured on 931 demo fills. At 1.5 times the bar’s minimum the out-of-sample figure would read -150.5 R (-0.132 R per trade over 1140 trades).

Entry: stop order 2 pips past the signal close, cancelled after 4 bars; 3,485 signals placed, 3,456 filled, 29 expired; trigger state, an entry on every bar the conditions hold while flat.

Exit: stop 2.5 ATR (mean 718.6 pips, 2.50 ATR); target 3.5 R; exits 69% at the stop, 20% at the target, 11% at the day end.

Filters: signals on every weekday; no entries 22:00 to 01:00 broker time (6,081 bars); max spread 50 pips (1,240 bars over it); M1 confirmation 0.2 atr (1,132 signals rejected).

Monthly R, price_vs_50d_high_atr < -15.421 on DJI30One bar per month: the R earned that month; the held-out months in the full colour.2023-10: -13.6 R over 130 trades2023-102023-11: +33.2 R over 61 trades2023-12: +8.8 R over 23 trades2024-01: -12.5 R over 81 trades2024-02: -2.1 R over 64 trades2024-03: -18.5 R over 84 trades2024-04: -42.9 R over 123 trades2024-042024-05: -3.9 R over 96 trades2024-06: -24.3 R over 119 trades2024-07: -7.0 R over 89 trades2024-08: +12.2 R over 98 trades2024-09: +8.1 R over 70 trades2024-10: -52.0 R over 111 trades2024-102024-11: +5.1 R over 80 trades2024-12: -58.1 R over 107 trades2025-01: +24.2 R over 108 trades2025-02: -27.4 R over 112 trades2025-03: -25.3 R over 77 trades2025-04: -22.2 R over 125 trades2025-042025-05: -22.5 R over 113 trades2025-06: +10.0 R over 78 trades2025-07: -11.4 R over 86 trades2025-08: -12.1 R over 94 trades2025-09: -0.2 R over 76 trades2025-10: +1.9 R over 97 trades2025-102025-11: -23.5 R over 98 trades2025-12: -16.3 R over 120 trades2026-01: -22.6 R over 97 trades2026-02: +6.8 R over 87 trades2026-03: -36.5 R over 125 trades2026-04: +17.8 R over 113 trades2026-042026-05: -4.5 R over 46 trades2026-06: -1.2 R over 78 trades2026-07: -8.2 R over 122 trades2026-08: -16.4 R over 118 trades2026-09: -39.6 R over 150 trades+33 R-58 R

price_vs_50d_high_atr < -6.179 (p90)

Honesty panelActive, net negative

Traded 1258 times out of sample, net negative beyond chance (-165.9 R, -0.132 R per trade, t = -2.6); still active in the last quarter (-67.4 R over 416 trades).

In sample
2023-09-30 to 2025-11-05
-321.8 R
2798 trades · 29% wins
-0.115 R per trade · t = -3.4, beyond chance
about 986 trades to confirm this mean (-0.115 R per trade) at t = -2 · shown: 2798
spread 0.078 R per trade
Out of sample
2025-11-05 to 2026-09-30
-165.9 R
1258 trades · 28% wins
-0.132 R per trade · t = -2.6, beyond chance
about 739 trades to confirm this mean (-0.132 R per trade) at t = -2 · shown: 1258
spread 0.065 R per trade
Last quarter
last 91 days of the window
-67.4 R
416 trades · 27% wins
-0.162 R per trade · t = -1.9, within chance
about 478 trades to confirm this mean (-0.162 R per trade) at t = -2 · shown: 416
spread 0.074 R per trade
Costs. Spread 0.074 R per trade at 1.25 times the bar’s minimum. Fill offset 0.14 stop-distances on every entry, measured on 931 demo fills. At 1.5 times the bar’s minimum the out-of-sample figure would read -182.3 R (-0.145 R per trade over 1258 trades).

Entry: stop order 2 pips past the signal close, cancelled after 4 bars; 4,090 signals placed, 4,056 filled, 34 expired; trigger state, an entry on every bar the conditions hold while flat.

Exit: stop 2.5 ATR (mean 743.3 pips, 2.50 ATR); target 3.5 R; exits 69% at the stop, 19% at the target, 12% at the day end.

Filters: signals on every weekday; no entries 22:00 to 01:00 broker time (2,889 bars); max spread 50 pips (1,050 bars over it); M1 confirmation 0.2 atr (1,325 signals rejected).

Monthly R, price_vs_50d_high_atr < -6.179 on DJI30One bar per month: the R earned that month; the held-out months in the full colour.2023-10: -13.6 R over 130 trades2023-102023-11: +42.3 R over 86 trades2023-12: +22.1 R over 85 trades2024-01: -9.4 R over 127 trades2024-02: -3.0 R over 95 trades2024-03: -18.9 R over 91 trades2024-04: -51.8 R over 130 trades2024-042024-05: -2.8 R over 106 trades2024-06: -25.5 R over 120 trades2024-07: -4.3 R over 106 trades2024-08: +2.6 R over 119 trades2024-09: +1.7 R over 107 trades2024-10: -50.5 R over 138 trades2024-102024-11: +3.7 R over 96 trades2024-12: -70.0 R over 113 trades2025-01: +18.8 R over 119 trades2025-02: -29.3 R over 113 trades2025-03: -26.5 R over 78 trades2025-04: -22.2 R over 125 trades2025-042025-05: -26.6 R over 122 trades2025-06: +5.6 R over 113 trades2025-07: -24.4 R over 119 trades2025-08: -18.6 R over 113 trades2025-09: -7.4 R over 109 trades2025-10: -7.0 R over 125 trades2025-102025-11: -19.2 R over 104 trades2025-12: -19.7 R over 125 trades2026-01: -32.2 R over 123 trades2026-02: -8.8 R over 115 trades2026-03: -35.6 R over 105 trades2026-04: +22.2 R over 113 trades2026-042026-05: -11.0 R over 54 trades2026-06: -1.0 R over 116 trades2026-07: -10.1 R over 137 trades2026-08: -17.7 R over 129 trades2026-09: -39.6 R over 150 trades+42 R-70 R

What price_vs_50d_high_atr measures

Close vs 50-day high (ATR) (atr), on M5 bars. Close minus the highest high of the last 14,400 M5 bars (fifty 24-hour days), in ATR multiples.

Test your own variant

Rule Collider runs one rule on one symbol over years of broker M5 data with the fills and the spread a demo fleet measured, and shows the out-of-sample figure first. Early access goes to the waitlist in signup order.

Join the waitlist

Historical simulation on broker M5 data with a measured fill offset, a measured spread factor, swap and a one-minute confirmation model. Past results do not predict future results. This is a research tool, not investment advice. Hours are broker time, the chart time in MT5. Built 2026-11-04 from runs ending 2026-09-30; refreshed monthly.