Stop-loss study, 2026-10-22T16:53:49.319Z

Check 1. Inventory
  golden-cross-1-hour: net_usd_per_contract / entry_time, net_usd_per_oz / entry_time
  golden-cross-eurusd-1h: net_pips / entry_time
  coppock-curve-es-1h: net_usd_per_contract / entry_time
  ema-9-20-crossover-gold: net_usd_per_oz / entry_time
  es-gap-fill-strategy: net_usd / entry_ny
  bitcoin-cme-gap-fill: net_pct / entry_utc
  ema-9-21-crossover-1h: net_usd_per_contract / entry_time
  ema-ribbon-5-8-13-nq: net_usd_per_contract / entry_time
  supertrend: net_usd / entry_date
  fx-six-strategies-four-pairs: net_pips / entry_date
  bollinger-band-reversal: net_usd / entry_date
  donchian-channel-breakout: net_usd / entry_date
  rsi-divergence: net_usd / entry_date
  macd-crossover: net_usd / entry_date
  gold-opening-range-breakout: net_points / date
  initial-balance-breakout-es-futures: net_usd / entry_ny
  initial-balance-breakout-strategy-nq: net_usd / entry_ny
  london-breakout-strategy-gbpusd: net_pips / entry_ny
  [ok] 183 pairs, 220 versions, every version found with trades
  [ok] both versions of every pair read in the same log column (unit); a test on two markets may use one column per market

Check 2. Logs against the summaries
  [ok] trade counts equal for all 220 versions; totals equal for the 110 whose summary is in the log's unit

Check 3. Shared entries
  [ok] first entry identical in every A, B and T pair
  kind A: share of the second version's entries that are also entries of the first: min 100.0%, median 100.0%
  kind B: share of the second version's entries that are also entries of the first: min 100.0%, median 100.0%
  kind C: share of the second version's entries that are also entries of the first: min 62.8%, median 91.9%
  kind T: share of the second version's entries that are also entries of the first: min 100.0%, median 100.0%

All checks passed.

Questions 1-5 (the second version has the stop; "better" = higher result after costs)
  A stop only                    pairs  15 sets  8 | better 50% (plain 47%, p 1.000) | worst smaller 94% (p 0.017), med ratio 0.60 | dd smaller 69% (p 0.364), med ratio 0.92 | win lower 100% | trades x1.00 | flips +2/-1 | gates up 4 down 2 same 9
  B stop + target                pairs  70 sets 18 | better 24% (plain 29%, p 0.007) | worst smaller 94% (p 0.000), med ratio 0.55 | dd smaller 65% (p 0.213), med ratio 0.80 | win lower 69% | trades x1.00 | flips +7/-13 | gates up 18 down 30 same 22
  C stop+target+time replace exit pairs  40 sets 20 | better 40% (plain 40%, p 0.454) | worst smaller 93% (p 0.000), med ratio 0.52 | dd smaller 50% (p 1.000), med ratio 0.98 | win lower 78% | trades x1.15 | flips +5/-6 | gates up 8 down 10 same 22
  all stop pairs                 pairs 125 sets 40 | better 34% (plain 34%, p 0.019) | worst smaller 94% (p 0.000), med ratio 0.55 | dd smaller 58% (p 0.275), med ratio 0.88 | win lower 77% | trades x1.00 | flips +14/-20 | gates up 30 down 42 same 53

Question 6 (the second version adds a target to the same stop)
  T target added                 pairs  58 sets 12 | better 9% (plain 16%, p 0.001) | worst smaller 54% (p 0.503), med ratio 1.00 | dd smaller 58% (p 0.646), med ratio 0.85 | win lower 0% | trades x1.00 | flips +3/-13 | gates up 10 down 18 same 30

Question 7 (rule kind of the trend vs mean reversion study)
  stops, trend following         pairs 102 sets 28 | better 36% (plain 36%, p 0.085) | worst smaller 91% (p 0.000), med ratio 0.61 | dd smaller 68% (p 0.032), med ratio 0.80 | win lower 66% | trades x1.00 | flips +12/-16 | gates up 24 down 37 same 41
  stops, mean reversion          pairs  23 sets 12 | better 29% (plain 26%, p 0.228) | worst smaller 100% (p 0.000), med ratio 0.34 | dd smaller 33% (p 0.294), med ratio 1.17 | win lower 100% | trades x1.04 | flips +2/-4 | gates up 6 down 5 same 12
  targets, trend following       pairs  58 sets 12 | better 9% (plain 16%, p 0.001) | worst smaller 54% (p 0.503), med ratio 1.00 | dd smaller 58% (p 0.646), med ratio 0.85 | win lower 0% | trades x1.00 | flips +3/-13 | gates up 10 down 18 same 30
  targets, mean reversion: no pairs

All-six-gate versions in pairs: without stop ["golden-cross-1-hour Gold GH-L-N"]; with stop []; targets: without [], with []
Stop-loss study, independent second computation

IDENTICAL (137 numbers)