The 200-day moving average rule on SPY (spec sma200_spec.md). Run 2026-10-21T09:35:49.275Z; 500 control runs per variant. Trade results in % of the price (total return, dividends included), cost 0.04% a round trip.
data: 8489 days 1993-01-29 .. 2026-10-20; 136 dividends; cash = 3-month T-bill (FRED DTB3)

=== SMA-D200: close vs its 200-day average, trade at the next open ===
  DEV      n=  45 net 123.09%  avg 2.735%  PF 3.74 t 1.42
  VAL      n=  36 net 71.17%  avg 1.977%  PF 2.81 t 1.36
  HOLDOUT  n=  31 net 126.64%  avg 4.085%  PF 4.71 t 1.92
  ALL      n=112 net 320.89% (sum of trade %) avg 2.865% (gross 2.905%); PF 3.71 t 2.67; at 0.10% cost: sum 314.17%
  winners 40/112 = 35.7% (95% Wilson 27.4%-44.9%); days held median 6 max 525; trades of 10 days or less: 69 (net -86.43%); 10 best trades 338.97%; open at the end: yes
  by year (n, sum %): 1993 1 0.20 | 1994 7 -5.71 | 1995 1 40.97 | 1996 3 66.87 | 1998 4 12.23 | 1999 2 -3.15 | 2000 12 -14.94 | 2002 4 -4.10 | 2003 2 24.97 | 2004 4 -2.35 | 2005 5 8.10 | 2006 5 11.02 | 2007 8 -0.89 | 2008 1 -0.36 | 2009 3 12.85 | 2010 4 6.28 | 2011 5 -5.40 | 2012 3 51.10 | 2014 1 8.04 | 2015 6 -11.46 | 2016 2 33.03 | 2018 5 -2.33 | 2019 3 2.79 | 2020 5 42.83 | 2022 6 -8.71 | 2023 7 32.21 | 2025 2 12.33 | 2026 1 14.49
  control (holdings moved to random times, same lengths, 500 runs, 0 s): mean trades 111.28, median avg 2.471%, 5-95% 1.893 .. 3.155%; real avg 2.865% at percentile 81.60
  [PASS] 1 PF >= 1.2 in validation: VAL PF 2.81
  [FAIL] 2 >= 50 trades in each period: 45 / 36 / 31
  [FAIL] 3 t >= 2.0 on development: DEV t 1.42
  [FAIL] 4 above 95th pct of the control: percentile 81.60
  [FAIL] 5 positive without top-10 trades: -18.08%
  [PASS] 6 holdout net positive: HOLDOUT net 126.64% PF 4.71
  VERDICT Buried (2/6 gates)
  whole strategy with cash interest, 1993-11-11 .. 2026-10-20: x16.37, 8.9% a year, worst drawdown -24.6% (1999-07-16 -> 2000-09-29), volatility 11.9%, invested 75.0% of days
  crashes: 2000-02 dot-com -14.9% | 2008-09 financial crisis -4.9% | 2020 Covid -18.9% | 2022 rate hikes -14.3%
  worst year 2000 -0.19048272435331315; years better than buy-and-hold: 7 of 34

=== SMA-M10: month-end close vs the average of the last 10 month-ends, trade at the next open ===
  DEV      n=   9 net 179.83%  avg 19.981%  PF 12.90 t 1.56
  VAL      n=   5 net 75.50%  avg 15.100%  PF 10.49 t 1.35
  HOLDOUT  n=  11 net 108.92%  avg 9.902%  PF 4.90 t 1.69
  ALL      n=25 net 364.25% (sum of trade %) avg 14.570% (gross 14.610%); PF 8.15 t 2.61; at 0.10% cost: sum 362.75%
  winners 16/25 = 64.0% (95% Wilson 44.5%-79.8%); days held median 189 max 905; trades of 10 days or less: 0 (net 0.00%); 10 best trades 384.15%; open at the end: yes
  by year (n, sum %): 1993 1 -6.22 | 1994 1 -3.37 | 1995 1 117.07 | 1998 1 16.76 | 1999 1 6.71 | 2002 1 -5.52 | 2003 1 22.18 | 2004 1 3.37 | 2005 1 28.85 | 2009 1 18.98 | 2010 2 3.12 | 2012 1 56.61 | 2015 1 -3.22 | 2016 1 39.46 | 2018 1 -11.75 | 2019 2 -0.00 | 2020 1 47.03 | 2022 2 -14.71 | 2023 2 29.29 | 2025 1 12.49 | 2026 1 7.12
  control (holdings moved to random times, same lengths, 500 runs, 0 s): mean trades 24.27, median avg 12.014%, 5-95% 8.173 .. 15.588%; real avg 14.570% at percentile 88.40
  [PASS] 1 PF >= 1.2 in validation: VAL PF 10.49
  [FAIL] 2 >= 50 trades in each period: 9 / 5 / 11
  [FAIL] 3 t >= 2.0 on development: DEV t 1.56
  [FAIL] 4 above 95th pct of the control: percentile 88.40
  [FAIL] 5 positive without top-10 trades: -19.90%
  [PASS] 6 holdout net positive: HOLDOUT net 108.92% PF 4.90
  VERDICT Inconclusive (2/6 gates)
  whole strategy with cash interest, 1993-11-11 .. 2026-10-20: x20.38, 9.6% a year, worst drawdown -24.3% (2022-01-03 -> 2023-03-13), volatility 12.6%, invested 76.0% of days
  crashes: 2000-02 dot-com -5.2% | 2008-09 financial crisis -4.5% | 2020 Covid -11.9% | 2022 rate hikes -16.4%
  worst year 2022 -0.20562070150606293; years better than buy-and-hold: 7 of 34

buy-and-hold SPY (total return), 1993-11-11 .. 2026-10-20: x29.77, 10.9% a year, worst drawdown -55.2% (2007-10-09 -> 2009-03-09), volatility 18.7%
  crashes: 2000-02 dot-com -47.5% | 2008-09 financial crisis -55.2% | 2020 Covid -33.7% | 2022 rate hikes -24.5%; worst year 2008 -0.3680716195004692
  by year, buy-and-hold / D200 / M10: 1993 1.1%/0.9%/1.1% | 1994 0.4%/-4.5%/-7.6% | 1995 38.0%/37.0%/34.0% | 1996 22.5%/17.1%/22.5% | 1997 33.5%/33.5%/33.5% | 1998 28.7%/16.9%/11.9% | 1999 20.4%/9.8%/13.3% | 2000 -9.7%/-19.0%/0.4% | 2001 -11.8%/3.5%/3.5% | 2002 -21.6%/-2.6%/-4.1% | 2003 28.2%/25.0%/23.0% | 2004 10.7%/3.5%/7.4% | 2005 4.8%/1.6%/2.0% | 2006 15.8%/9.8%/15.8% | 2007 5.1%/2.2%/5.1% | 2008 -36.8%/1.0%/1.6% | 2009 26.4%/17.3%/22.0% | 2010 15.1%/-0.2%/2.2% | 2011 1.9%/-6.3%/-1.9% | 2012 16.0%/13.9%/10.0% | 2013 32.3%/32.3%/32.3% | 2014 13.5%/11.1%/13.5% | 2015 1.3%/-9.5%/-6.7% | 2016 12.0%/11.1%/9.4% | 2017 21.7%/21.7%/21.7% | 2018 -4.6%/-3.7%/-7.4% | 2019 31.2%/14.9%/6.8% | 2020 18.4%/8.5%/15.6% | 2021 28.7%/28.7%/28.7% | 2022 -18.2%/-15.1%/-20.6% | 2023 26.2%/11.4%/10.9% | 2024 24.9%/24.9%/24.9% | 2025 17.7%/11.6%/12.4% | 2026 13.7%/10.7%/3.3%
rule 3: not applicable (the only gaps in the SPY file are real market closures: Sept 2001, 2 Jan 2007, Oct 2012)