London Breakout on GBPUSD (spec lb_spec.md): Asian range 00:00-07:00 London, stop orders 1 pip beyond, entries 07:00-10:00, stop at the other side, exit by 16:00 London. Run 2026-10-21T11:36:02.322Z; 500 control runs per variant. Results in R (fill to stop) after 1.5 pips a round trip. data: 8297247 minutes 2004-01-01 00:05 .. 2026-09-28 15:46 New York; clock check: no week shifted (lb_clock_result.json); 5278 breakout days (2585 buys, 2693 sells; both levels in the first minute: 5) Asian range, median pips by year: 2004 61 | 2005 55 | 2006 45 | 2007 59 | 2008 82 | 2009 88 | 2010 63 | 2011 50 | 2012 31 | 2013 33 | 2014 28 | 2015 28 | 2016 36 | 2017 27 | 2018 27 | 2019 22 | 2020 35 | 2021 28 | 2022 39 | 2023 28 | 2024 20 | 2025 28 | 2026 24 === LB-1R: target 1R, out by 16:00 London === DEV n= 1968 net -1.26 R avg -0.001 R PF 1.00 t -0.03 VAL n= 1683 net -40.44 R avg -0.024 R PF 0.95 t -1.04 HOLDOUT n= 1627 net -54.91 R avg -0.034 R PF 0.93 t -1.47 ALL n=5278 net -96.60 R avg -0.018 R (gross 0.025 R, cost 0.043 R a trade); pips net -2344.50 (gross 5572.50); PF 0.96 t -1.45; at 1.0 pip -0.004 R, at 2.5 pips -0.047 R a trade winners 2686/5278 = 50.9% (95% Wilson 49.5-52.2%); ended by 16:00 1022 (19.4%), target 2191 (41.5%), stop 2065 (39.1%); R median 38.00 pips; without the 10 best -109.80 R longs n=2585 net -101.20 R gross 10.77 R | shorts n=2693 net 4.59 R gross 120.35 R by year (n, net R, net pips): 2004 237 10.16 860.50 | 2005 219 6.29 319.50 | 2006 238 1.72 127.00 | 2007 134 8.57 506.00 | 2008 189 6.35 417.50 | 2009 231 -0.46 266.00 | 2010 240 -16.84 -1402.80 | 2011 237 -19.62 -1061.40 | 2012 243 2.57 194.30 | 2013 236 10.93 323.90 | 2014 235 -25.31 -502.80 | 2015 247 -44.81 -1022.90 | 2016 237 -2.73 171.80 | 2017 236 14.55 250.90 | 2018 251 0.47 -225.40 | 2019 241 6.46 423.50 | 2020 240 8.65 328.70 | 2021 249 -1.89 -320.10 | 2022 234 -23.36 -1017.70 | 2023 247 -22.40 -615.90 | 2024 250 13.47 387.80 | 2025 231 -16.64 -502.50 | 2026 176 -12.74 -250.40 by weekday (n, net R): Mon 1043 -16.79 | Tue 1088 -1.43 | Wed 1063 10.15 | Thu 1051 -46.58 | Fri 1033 -41.95 by Asian range (pips: n, net R a trade, gross R a trade): 0-20: 686 -0.071 0.013 | 20-30: 1327 -0.002 0.054 | 30-40: 1009 -0.050 -0.009 | 40-60: 1183 0.010 0.040 | 60-+: 1073 -0.006 0.011 control (entries moved to other days, same London time, same R in pips, 500 runs, 8 s): mean trades 3733.40, median avg -0.043 R, 5-95% -0.067 .. -0.020 R; real avg -0.018 R at percentile 95.40 [FAIL] 1 PF >= 1.2 in validation: VAL PF 0.95 [PASS] 2 >= 50 trades in each period: 1968 / 1683 / 1627 [FAIL] 3 t >= 2.0 on development: DEV t -0.03 [PASS] 4 above 95th pct of the control: percentile 95.40 [FAIL] 5 positive without top-10 trades: -109.80 R [FAIL] 6 holdout net positive (flagged used): HOLDOUT net -54.91 R PF 0.93 VERDICT Buried (2/6 gates) rule 3: 10 trades are open across a suspect gap; without them avg -0.018 R, gates passed 2/6 -> Buried (verdict unchanged) === LB-2R: target 2R, out by 16:00 London === DEV n= 1968 net 34.22 R avg 0.017 R PF 1.04 t 0.72 VAL n= 1683 net -28.31 R avg -0.017 R PF 0.97 t -0.57 HOLDOUT n= 1627 net 14.37 R avg 0.009 R PF 1.02 t 0.30 ALL n=5278 net 20.28 R avg 0.004 R (gross 0.047 R, cost 0.043 R a trade); pips net 2062.20 (gross 9979.20); PF 1.01 t 0.24; at 1.0 pip 0.018 R, at 2.5 pips -0.025 R a trade winners 2283/5278 = 43.3% (95% Wilson 41.9-44.6%); ended by 16:00 2039 (38.6%), stop 2332 (44.2%), target 907 (17.2%); R median 38.00 pips; without the 10 best -1.85 R longs n=2585 net -43.58 R gross 68.38 R | shorts n=2693 net 63.85 R gross 179.61 R by year (n, net R, net pips): 2004 237 29.16 2051.50 | 2005 219 7.47 275.50 | 2006 238 12.74 517.00 | 2007 134 13.49 776.00 | 2008 189 16.87 830.50 | 2009 231 -0.15 124.90 | 2010 240 -12.64 -1221.00 | 2011 237 -18.88 -984.40 | 2012 243 -13.84 -208.50 | 2013 236 32.73 951.70 | 2014 235 -25.81 -604.00 | 2015 247 -52.49 -1343.40 | 2016 237 16.88 1040.50 | 2017 236 -4.13 -85.80 | 2018 251 1.90 -168.90 | 2019 241 2.62 277.70 | 2020 240 12.90 515.90 | 2021 249 0.44 -250.50 | 2022 234 -6.05 -294.30 | 2023 247 -7.13 -145.10 | 2024 250 25.16 435.50 | 2025 231 -3.11 -220.00 | 2026 176 -7.84 -208.60 by weekday (n, net R): Mon 1043 22.24 | Tue 1088 31.43 | Wed 1063 25.29 | Thu 1051 -18.34 | Fri 1033 -40.35 by Asian range (pips: n, net R a trade, gross R a trade): 0-20: 686 -0.020 0.063 | 20-30: 1327 -0.000 0.056 | 30-40: 1009 -0.027 0.014 | 40-60: 1183 0.041 0.071 | 60-+: 1073 0.013 0.031 control (entries moved to other days, same London time, same R in pips, 500 runs, 9 s): mean trades 3733.40, median avg -0.029 R, 5-95% -0.059 .. 0.001 R; real avg 0.004 R at percentile 96.80 [FAIL] 1 PF >= 1.2 in validation: VAL PF 0.97 [PASS] 2 >= 50 trades in each period: 1968 / 1683 / 1627 [FAIL] 3 t >= 2.0 on development: DEV t 0.72 [PASS] 4 above 95th pct of the control: percentile 96.80 [FAIL] 5 positive without top-10 trades: -1.85 R [PASS] 6 holdout net positive (flagged used): HOLDOUT net 14.37 R PF 1.02 VERDICT Buried (3/6 gates) rule 3: 12 trades are open across a suspect gap; without them avg 0.004 R, gates passed 3/6 -> Buried (verdict unchanged) === LB-CLOSE: no target, out by 16:00 London === DEV n= 1968 net 48.64 R avg 0.025 R PF 1.05 t 0.92 VAL n= 1683 net 14.52 R avg 0.009 R PF 1.01 t 0.24 HOLDOUT n= 1627 net 20.16 R avg 0.012 R PF 1.02 t 0.36 ALL n=5278 net 83.32 R avg 0.016 R (gross 0.059 R, cost 0.043 R a trade); pips net 3935.60 (gross 11852.60); PF 1.03 t 0.86; at 1.0 pip 0.030 R, at 2.5 pips -0.013 R a trade winners 2195/5278 = 41.6% (95% Wilson 40.3-42.9%); ended by 16:00 2884 (54.6%), stop 2394 (45.4%); R median 38.00 pips; without the 10 best 8.16 R longs n=2585 net -21.34 R gross 90.62 R | shorts n=2693 net 104.66 R gross 220.42 R by year (n, net R, net pips): 2004 237 36.54 2396.50 | 2005 219 6.17 323.50 | 2006 238 29.34 1082.00 | 2007 134 19.17 1054.00 | 2008 189 16.36 717.50 | 2009 231 -4.58 -417.10 | 2010 240 -13.70 -1389.90 | 2011 237 -28.89 -1331.00 | 2012 243 -11.78 -191.10 | 2013 236 45.67 1260.20 | 2014 235 -26.24 -554.50 | 2015 247 -39.70 -995.30 | 2016 237 27.94 1373.60 | 2017 236 -8.57 -166.00 | 2018 251 -9.88 -541.60 | 2019 241 25.30 824.60 | 2020 240 12.00 691.80 | 2021 249 -11.14 -472.60 | 2022 234 16.14 493.30 | 2023 247 -1.55 21.10 | 2024 250 13.97 145.90 | 2025 231 1.64 -160.40 | 2026 176 -10.91 -228.90 by weekday (n, net R): Mon 1043 17.85 | Tue 1088 55.13 | Wed 1063 27.49 | Thu 1051 -26.32 | Fri 1033 9.17 by Asian range (pips: n, net R a trade, gross R a trade): 0-20: 686 -0.000 0.083 | 20-30: 1327 0.015 0.071 | 30-40: 1009 -0.012 0.029 | 40-60: 1183 0.055 0.085 | 60-+: 1073 0.011 0.028 control (entries moved to other days, same London time, same R in pips, 500 runs, 9 s): mean trades 3733.40, median avg -0.008 R, 5-95% -0.046 .. 0.035 R; real avg 0.016 R at percentile 80.80 [FAIL] 1 PF >= 1.2 in validation: VAL PF 1.01 [PASS] 2 >= 50 trades in each period: 1968 / 1683 / 1627 [FAIL] 3 t >= 2.0 on development: DEV t 0.92 [FAIL] 4 above 95th pct of the control: percentile 80.80 [PASS] 5 positive without top-10 trades: 8.16 R [PASS] 6 holdout net positive (flagged used): HOLDOUT net 20.16 R PF 1.02 VERDICT Buried (3/6 gates) rule 3: 14 trades are open across a suspect gap; without them avg 0.016 R, gates passed 3/6 -> Buried (verdict unchanged)