CAT researchOverviewEURUSDGBPUSDXAUUSDGBPCADEURCADAUDJPYGBPJPYResearch notes

entry5m v10 (H4 bias) — diagnostic 2: stop widened to 1.5x, with 3R / 1R / 0.5R targets and the MFE distribution (NOT judged)

Declared as diagnostic addendum 2 in the v10 pre-registration (2026-09-07). Stop one tick beyond entry +/- 1.5 x |H4 extreme - entry| (stop_mult = 1.5); the cancel level moves with it; every R figure is measured against the widened stop, so 1R here is 1.5 times the price distance of 1R in addendum 1. Same cell as the judged v10 run (htf 4h, offset 3600, H1-extreme stop, no scale-out, four symbols, five-year Tickstory 5m data); only target_r varies (for the 0.5R cell the reward-to-risk floor min_rr is lowered to 0.5, otherwise the floor of 1.0 rejects every order; nothing else changes). The MFE table reads the judged 3R-target run's trades: the share whose best excursion before the exit reached each R level, i.e. how often the entry "gave some minimum profit" regardless of where the exit rule finally closed it.

Targets at 3R (the judged cell), 1R and 0.5R

Symbol target R n win % net R PF ex-outlier PF expectancy R halves (R)
EURUSD 3.0 118 35.6 2.4 1.03 0.91 0.021 -0.5 / 2.9
EURUSD 1.0 147 55.1 4.5 1.07 1.02 0.031 -5.1 / 9.6
EURUSD 0.5 190 69.5 3.9 1.07 1.04 0.020 -2.7 / 6.6
GBPUSD 3.0 157 29.9 -15.8 0.86 0.78 -0.101 -6.4 / -9.4
GBPUSD 1.0 186 50.0 -13.2 0.86 0.83 -0.071 -8.6 / -4.6
GBPUSD 0.5 239 66.9 -5.0 0.94 0.92 -0.021 -0.8 / -4.2
XAUUSD 3.0 133 30.1 -14.6 0.85 0.76 -0.110 -10.6 / -4.0
XAUUSD 1.0 154 45.5 -24.9 0.71 0.68 -0.162 -9.0 / -16.0
XAUUSD 0.5 201 64.2 -13.1 0.82 0.80 -0.065 0.0 / -13.1
EURJPY 3.0 133 30.8 -0.2 1.00 0.90 -0.001 -1.7 / 1.5
EURJPY 1.0 155 46.5 -16.5 0.79 0.75 -0.106 -2.0 / -14.4
EURJPY 0.5 204 63.7 -7.6 0.89 0.87 -0.037 -0.1 / -7.5

MFE / MAE of the 3R-target trades

Symbol n MFE >= 0.25R >= 0.5R >= 1R >= 2R >= 3R median MFE R median MAE R stopped (n) median MFE of stopped
EURUSD 118 73% 62% 43% 22% 18% 0.72 1.02 70 0.57
GBPUSD 157 80% 64% 42% 27% 14% 0.83 1.05 107 0.56
XAUUSD 133 79% 61% 39% 23% 15% 0.74 1.05 89 0.50
EURJPY 133 78% 62% 40% 27% 17% 0.76 1.02 86 0.46

Reading (observations only; R is measured against the 1.5x stop, so every R here is 1.5 times the price distance of addendum 1's R)

  • Fewer trades, same excursion profile. The wider stop cuts the trade count by about a quarter (118-157 at 3R against 168-207) because more targets fall inside the widened risk and positions stay open longer. The MFE/MAE shape is unchanged: 61-64% of trades reach 0.5R, 39-43% reach 1R, and the median adverse excursion is again about 1.0R — the excursions scale with the stop rather than shrinking relative to it.
  • Stopped-out share falls, net R barely moves. 55-67% of trades still end at the stop (70 of 118 on EURUSD) against 67-72% with the 1.0x stop. EURUSD turns marginally positive at every target (+2.4 / +4.5 / +3.9 R, PF 1.03-1.07, ex-outlier PF 0.91-1.04, expectancy at most +0.031 R); GBPUSD, gold and EURJPY stay negative at every target, with EURJPY at −0.2 R (PF 1.00) at 3R.
  • The 1R and 0.5R rows repeat addendum 1's lesson: the win rate rises to 46-55% and 64-70%, net R does not follow.

Nothing in this read is judged; the v10 verdict stands (FAIL 0/3). A cell built from it (for example EURUSD with the 1.5x stop) would need its own pre-registration and would start from an expectancy an order of magnitude below the +0.05 R floor.