Rule Collider

Backtests · DJI30 · x_eurusd_vs_ema50 · long

Does x_eurusd_vs_ema50 below 0.03829 work on DJI30? 2023 to 2026 backtest

DJI30 long when EURUSD close vs EMA50 (its 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, no one-minute confirmation, 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
p10x_eurusd_vs_ema50 < -2.853-196.8 R / 1421-130.2 R / 608-3.1257-38.8 R / 184Active, net negative
p25x_eurusd_vs_ema50 < -1.512-269.3 R / 2180-142.0 R / 927-2.6542-73.5 R / 300Active, net negative
p50x_eurusd_vs_ema50 < 0.03829-368.0 R / 2687-141.9 R / 1173-2.3878-53.4 R / 373Active, net negative
p75x_eurusd_vs_ema50 < 1.59-316.1 R / 2854-203.5 R / 1228-3.3450-80.8 R / 404Active, net negative
p90x_eurusd_vs_ema50 < 2.958-332.5 R / 2882-223.7 R / 1297-3.5418-92.4 R / 426Active, 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

x_eurusd_vs_ema50 < -2.853 (p10)

Honesty panelActive, net negative

Traded 608 times out of sample, net negative beyond chance (-130.2 R, -0.214 R per trade, t = -3.1); still active in the last quarter (-38.8 R over 184 trades).

In sample
2023-09-30 to 2025-11-05
-196.8 R
1421 trades · 29% wins
-0.138 R per trade · t = -2.9, beyond chance
about 693 trades to confirm this mean (-0.138 R per trade) at t = -2 · shown: 1421
spread 0.102 R per trade
Out of sample
2025-11-05 to 2026-09-30
-130.2 R
608 trades · 27% wins
-0.214 R per trade · t = -3.1, beyond chance
about 257 trades to confirm this mean (-0.214 R per trade) at t = -2 · shown: 608
spread 0.075 R per trade
Last quarter
last 91 days of the window
-38.8 R
184 trades · 27% wins
-0.211 R per trade · t = -1.7, within chance
about 259 trades to confirm this mean (-0.211 R per trade) at t = -2 · shown: 184
spread 0.082 R per trade
Costs. Spread 0.094 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 -139.3 R (-0.229 R per trade over 608 trades).

Entry: stop order 2 pips past the signal close, cancelled after 4 bars; 2,083 signals placed, 2,029 filled, 54 expired; trigger state, an entry on every bar the conditions hold while flat.

Exit: stop 2.5 ATR (mean 650.9 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 (9,401 bars); max spread 50 pips (1,990 bars over it).

Monthly R, x_eurusd_vs_ema50 < -2.853 on DJI30One bar per month: the R earned that month; the held-out months in the full colour.2023-10: +2.2 R over 53 trades2023-102023-11: +30.0 R over 46 trades2023-12: -8.0 R over 57 trades2024-01: -32.4 R over 72 trades2024-02: -14.9 R over 53 trades2024-03: +0.3 R over 39 trades2024-04: -17.7 R over 75 trades2024-042024-05: -17.5 R over 64 trades2024-06: -18.0 R over 71 trades2024-07: +11.3 R over 54 trades2024-08: +22.0 R over 49 trades2024-09: +5.9 R over 41 trades2024-10: -44.4 R over 78 trades2024-102024-11: +5.8 R over 70 trades2024-12: -31.2 R over 64 trades2025-01: +12.6 R over 54 trades2025-02: -10.5 R over 57 trades2025-03: -5.5 R over 49 trades2025-04: -14.1 R over 52 trades2025-042025-05: -12.3 R over 39 trades2025-06: -9.7 R over 40 trades2025-07: -33.2 R over 75 trades2025-08: -12.7 R over 55 trades2025-09: -4.2 R over 59 trades2025-10: -0.0 R over 49 trades2025-102025-11: -4.7 R over 47 trades2025-12: -5.4 R over 45 trades2026-01: -8.0 R over 55 trades2026-02: -1.3 R over 55 trades2026-03: -28.9 R over 78 trades2026-04: -20.1 R over 56 trades2026-042026-05: -9.4 R over 37 trades2026-06: -13.9 R over 57 trades2026-07: -5.3 R over 64 trades2026-08: +8.3 R over 39 trades2026-09: -41.7 R over 81 trades+30 R-44 R

x_eurusd_vs_ema50 < -1.512 (p25)

Honesty panelActive, net negative

Traded 927 times out of sample, net negative beyond chance (-142.0 R, -0.153 R per trade, t = -2.6); still active in the last quarter (-73.5 R over 300 trades).

In sample
2023-09-30 to 2025-11-05
-269.3 R
2180 trades · 29% wins
-0.124 R per trade · t = -3.2, beyond chance
about 865 trades to confirm this mean (-0.124 R per trade) at t = -2 · shown: 2180
spread 0.095 R per trade
Out of sample
2025-11-05 to 2026-09-30
-142.0 R
927 trades · 28% wins
-0.153 R per trade · t = -2.6, beyond chance
about 542 trades to confirm this mean (-0.153 R per trade) at t = -2 · shown: 927
spread 0.074 R per trade
Last quarter
last 91 days of the window
-73.5 R
300 trades · 26% wins
-0.245 R per trade · t = -2.5, beyond chance
about 196 trades to confirm this mean (-0.245 R per trade) at t = -2 · shown: 300
spread 0.086 R per trade
Costs. Spread 0.089 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 -155.7 R (-0.168 R per trade over 927 trades).

Entry: stop order 2 pips past the signal close, cancelled after 4 bars; 3,190 signals placed, 3,107 filled, 83 expired; trigger state, an entry on every bar the conditions hold while flat.

Exit: stop 2.5 ATR (mean 683.2 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 (6,183 bars); max spread 50 pips (1,623 bars over it).

Monthly R, x_eurusd_vs_ema50 < -1.512 on DJI30One bar per month: the R earned that month; the held-out months in the full colour.2023-10: +1.2 R over 92 trades2023-102023-11: +18.3 R over 65 trades2023-12: -0.9 R over 84 trades2024-01: -26.5 R over 114 trades2024-02: -2.7 R over 83 trades2024-03: -22.0 R over 65 trades2024-04: -14.2 R over 103 trades2024-042024-05: -31.7 R over 98 trades2024-06: -24.5 R over 104 trades2024-07: -5.0 R over 87 trades2024-08: +2.8 R over 92 trades2024-09: -3.4 R over 80 trades2024-10: -17.8 R over 103 trades2024-102024-11: +0.0 R over 102 trades2024-12: -44.4 R over 87 trades2025-01: +9.1 R over 95 trades2025-02: -21.7 R over 92 trades2025-03: -21.6 R over 72 trades2025-04: -18.8 R over 86 trades2025-042025-05: -9.0 R over 63 trades2025-06: +8.9 R over 69 trades2025-07: -24.5 R over 104 trades2025-08: -20.7 R over 79 trades2025-09: -0.0 R over 88 trades2025-10: +1.4 R over 66 trades2025-102025-11: -11.1 R over 67 trades2025-12: -12.7 R over 87 trades2026-01: +0.1 R over 72 trades2026-02: +2.3 R over 86 trades2026-03: -25.4 R over 102 trades2026-04: -10.6 R over 85 trades2026-042026-05: -13.8 R over 53 trades2026-06: +1.1 R over 82 trades2026-07: -19.7 R over 108 trades2026-08: -18.8 R over 75 trades2026-09: -34.9 R over 117 trades+18 R-44 R

x_eurusd_vs_ema50 < 0.03829 (p50)

Honesty panelActive, net negative

Traded 1173 times out of sample, net negative beyond chance (-141.9 R, -0.121 R per trade, t = -2.3); still active in the last quarter (-53.4 R over 373 trades).

In sample
2023-09-30 to 2025-11-05
-368.0 R
2687 trades · 29% wins
-0.137 R per trade · t = -3.9, beyond chance
about 694 trades to confirm this mean (-0.137 R per trade) at t = -2 · shown: 2687
spread 0.087 R per trade
Out of sample
2025-11-05 to 2026-09-30
-141.9 R
1173 trades · 29% wins
-0.121 R per trade · t = -2.3, beyond chance
about 878 trades to confirm this mean (-0.121 R per trade) at t = -2 · shown: 1173
spread 0.069 R per trade
Last quarter
last 91 days of the window
-53.4 R
373 trades · 28% wins
-0.143 R per trade · t = -1.6, within chance
about 613 trades to confirm this mean (-0.143 R per trade) at t = -2 · shown: 373
spread 0.078 R per trade
Costs. Spread 0.081 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 -158.1 R (-0.135 R per trade over 1173 trades).

Entry: stop order 2 pips past the signal close, cancelled after 4 bars; 3,957 signals placed, 3,860 filled, 97 expired; trigger state, an entry on every bar the conditions hold while flat.

Exit: stop 2.5 ATR (mean 716.3 pips, 2.50 ATR); target 3.5 R; exits 68% 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 (3,594 bars); max spread 50 pips (1,187 bars over it).

Monthly R, x_eurusd_vs_ema50 < 0.03829 on DJI30One bar per month: the R earned that month; the held-out months in the full colour.2023-10: -29.2 R over 121 trades2023-102023-11: +32.0 R over 90 trades2023-12: +4.3 R over 101 trades2024-01: -15.4 R over 126 trades2024-02: -25.6 R over 112 trades2024-03: -8.5 R over 63 trades2024-04: -25.3 R over 122 trades2024-042024-05: -8.7 R over 104 trades2024-06: -34.7 R over 118 trades2024-07: -20.3 R over 120 trades2024-08: +3.3 R over 108 trades2024-09: -3.3 R over 97 trades2024-10: -44.3 R over 127 trades2024-102024-11: +0.5 R over 119 trades2024-12: -51.6 R over 104 trades2025-01: +7.3 R over 113 trades2025-02: -24.7 R over 101 trades2025-03: -34.8 R over 84 trades2025-04: -16.8 R over 107 trades2025-042025-05: -27.0 R over 102 trades2025-06: +3.1 R over 107 trades2025-07: -16.1 R over 121 trades2025-08: -12.7 R over 109 trades2025-09: -15.9 R over 109 trades2025-10: +1.0 R over 91 trades2025-102025-11: -13.5 R over 84 trades2025-12: -16.5 R over 102 trades2026-01: +1.4 R over 105 trades2026-02: +3.0 R over 105 trades2026-03: -35.5 R over 122 trades2026-04: -14.5 R over 114 trades2026-042026-05: -16.0 R over 65 trades2026-06: -1.5 R over 114 trades2026-07: -11.9 R over 133 trades2026-08: -1.5 R over 110 trades2026-09: -40.0 R over 130 trades+32 R-52 R

x_eurusd_vs_ema50 < 1.59 (p75)

Honesty panelActive, net negative

Traded 1228 times out of sample, net negative beyond chance (-203.5 R, -0.166 R per trade, t = -3.3); still active in the last quarter (-80.8 R over 404 trades).

In sample
2023-09-30 to 2025-11-05
-316.1 R
2854 trades · 29% wins
-0.111 R per trade · t = -3.3, beyond chance
about 1,059 trades to confirm this mean (-0.111 R per trade) at t = -2 · shown: 2854
spread 0.077 R per trade
Out of sample
2025-11-05 to 2026-09-30
-203.5 R
1228 trades · 28% wins
-0.166 R per trade · t = -3.3, beyond chance
about 450 trades to confirm this mean (-0.166 R per trade) at t = -2 · shown: 1228
spread 0.064 R per trade
Last quarter
last 91 days of the window
-80.8 R
404 trades · 27% wins
-0.200 R per trade · t = -2.3, beyond chance
about 304 trades to confirm this mean (-0.200 R per trade) at t = -2 · shown: 404
spread 0.072 R per trade
Costs. Spread 0.073 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 -219.1 R (-0.178 R per trade over 1228 trades).

Entry: stop order 2 pips past the signal close, cancelled after 4 bars; 4,179 signals placed, 4,082 filled, 97 expired; trigger state, an entry on every bar the conditions hold while flat.

Exit: stop 2.5 ATR (mean 752.0 pips, 2.50 ATR); target 3.5 R; exits 68% at the stop, 19% at the target, 13% at the day end.

Filters: signals on every weekday; no entries 22:00 to 01:00 broker time (2,319 bars); max spread 50 pips (738 bars over it).

Monthly R, x_eurusd_vs_ema50 < 1.59 on DJI30One bar per month: the R earned that month; the held-out months in the full colour.2023-10: -20.8 R over 126 trades2023-102023-11: +35.9 R over 104 trades2023-12: +8.4 R over 97 trades2024-01: -23.7 R over 143 trades2024-02: -0.6 R over 110 trades2024-03: -8.1 R over 67 trades2024-04: -35.4 R over 124 trades2024-042024-05: -13.4 R over 111 trades2024-06: -19.2 R over 110 trades2024-07: -4.0 R over 122 trades2024-08: +16.3 R over 113 trades2024-09: -3.0 R over 112 trades2024-10: -47.8 R over 137 trades2024-102024-11: +1.1 R over 124 trades2024-12: -58.4 R over 109 trades2025-01: +10.4 R over 109 trades2025-02: -32.0 R over 112 trades2025-03: -22.7 R over 74 trades2025-04: -6.6 R over 121 trades2025-042025-05: -22.4 R over 103 trades2025-06: -3.1 R over 117 trades2025-07: -22.4 R over 117 trades2025-08: -20.4 R over 123 trades2025-09: -12.8 R over 125 trades2025-10: -5.8 R over 132 trades2025-102025-11: -21.9 R over 96 trades2025-12: -21.4 R over 116 trades2026-01: -18.2 R over 112 trades2026-02: -13.7 R over 116 trades2026-03: -33.2 R over 107 trades2026-04: +3.1 R over 110 trades2026-042026-05: -20.7 R over 59 trades2026-06: -2.4 R over 120 trades2026-07: -26.1 R over 140 trades2026-08: -20.0 R over 123 trades2026-09: -34.7 R over 141 trades+36 R-58 R

x_eurusd_vs_ema50 < 2.958 (p90)

Honesty panelActive, net negative

Traded 1297 times out of sample, net negative beyond chance (-223.7 R, -0.172 R per trade, t = -3.5); still active in the last quarter (-92.4 R over 426 trades).

In sample
2023-09-30 to 2025-11-05
-332.5 R
2882 trades · 29% wins
-0.115 R per trade · t = -3.5, beyond chance
about 960 trades to confirm this mean (-0.115 R per trade) at t = -2 · shown: 2882
spread 0.073 R per trade
Out of sample
2025-11-05 to 2026-09-30
-223.7 R
1297 trades · 27% wins
-0.172 R per trade · t = -3.5, beyond chance
about 418 trades to confirm this mean (-0.172 R per trade) at t = -2 · shown: 1297
spread 0.062 R per trade
Last quarter
last 91 days of the window
-92.4 R
426 trades · 27% wins
-0.217 R per trade · t = -2.6, beyond chance
about 259 trades to confirm this mean (-0.217 R per trade) at t = -2 · shown: 426
spread 0.072 R per trade
Costs. Spread 0.069 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 -239.7 R (-0.185 R per trade over 1297 trades).

Entry: stop order 2 pips past the signal close, cancelled after 4 bars; 4,272 signals placed, 4,179 filled, 93 expired; trigger state, an entry on every bar the conditions hold while flat.

Exit: stop 2.5 ATR (mean 775.2 pips, 2.50 ATR); target 3.5 R; exits 68% at the stop, 18% at the target, 13% at the day end.

Filters: signals on every weekday; no entries 22:00 to 01:00 broker time (1,790 bars); max spread 50 pips (740 bars over it).

Monthly R, x_eurusd_vs_ema50 < 2.958 on DJI30One bar per month: the R earned that month; the held-out months in the full colour.2023-10: -15.6 R over 126 trades2023-102023-11: +36.2 R over 99 trades2023-12: +3.8 R over 100 trades2024-01: -21.6 R over 144 trades2024-02: +2.4 R over 105 trades2024-03: -7.0 R over 75 trades2024-04: -40.7 R over 134 trades2024-042024-05: -11.9 R over 101 trades2024-06: -27.1 R over 118 trades2024-07: +4.2 R over 115 trades2024-08: +8.3 R over 117 trades2024-09: -1.0 R over 111 trades2024-10: -49.3 R over 135 trades2024-102024-11: +2.3 R over 119 trades2024-12: -63.2 R over 109 trades2025-01: +12.1 R over 104 trades2025-02: -39.6 R over 118 trades2025-03: -21.5 R over 73 trades2025-04: -19.0 R over 129 trades2025-042025-05: -29.4 R over 119 trades2025-06: +6.9 R over 119 trades2025-07: -21.6 R over 124 trades2025-08: -17.5 R over 121 trades2025-09: -11.3 R over 122 trades2025-10: -3.2 R over 131 trades2025-102025-11: -18.8 R over 102 trades2025-12: -19.7 R over 117 trades2026-01: -26.8 R over 127 trades2026-02: -11.1 R over 128 trades2026-03: -38.9 R over 115 trades2026-04: +8.3 R over 114 trades2026-042026-05: -24.6 R over 56 trades2026-06: -7.8 R over 126 trades2026-07: -23.5 R over 149 trades2026-08: -24.2 R over 132 trades2026-09: -44.7 R over 145 trades+36 R-63 R

What x_eurusd_vs_ema50 measures

EURUSD close vs EMA50 (its ATR) (atr), on M5 bars. EURUSD close minus its 50-bar exponential average, in multiples of the EURUSD ATR(14).

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.