RSI(2) on QQQ, daily closes, Connors' rules replicated from SPY (spec rqq_spec.md). Run 2026-11-07T08:37:06.188Z; 500 control runs per variant. Results in net % per trade, 0.04% cost per round trip, fills at the close.
data: 6947 days, 1999-03-10 .. 2026-10-20; signals from 2000-03-03; 90 dividends

=== RQ2-10: buy at the close when close > SMA200 and RSI(2) < 10; sell at the first close above SMA5 ===
  DEV      n=  36 net 23.69 %  avg 0.658 %  PF 2.47 t 2.32
  VAL      n=  86 net 36.43 %  avg 0.424 %  PF 1.82 t 2.08
  HOLDOUT  n=  91 net 44.62 %  avg 0.490 %  PF 2.00 t 2.79
  ALL      n=213 net 104.74 % avg 0.492 % (gross 0.532 %); PF 1.99 t 4.07; with 0.10% cost avg 0.432 %
  win rate 147/213 = 69.0% (95% Wilson 62.5%-74.8%); trading days held median 3 max 11; open at the end: 1
  the 10 best trades 30.91 %; net without the 10 best 73.83 %, without the 20 best 48.48 %
  worst open loss at a close (open positions included): -12.63 % (trade entered 2000-04-12, exited 2000-04-17)
  next-open fills (descriptive): n=213 avg 0.446 % PF 1.95 t 3.76; by period avg 0.525 / 0.501 / 0.363 %
  whole strategy from 2000-03-03 with T-bill interest while out: growth x4.34 (5.7% a year), max drawdown -15.3% (2000-03-28 to 2000-04-14), volatility 8.5%, invested 11.8% of days; buy and hold x7.68 (8.0% a year), max drawdown -83.0%, volatility 26.4%
  crashes (strategy / buy and hold): 2000-02 dot-com 4.3% / -82.9%; 2008-09 financial crisis -1.8% / -51.6%; 2020 Covid -2.2% / -27.9%; 2022 rate hikes -2.6% / -34.3%
  entry days shared with SPY RS2-10: 93 of 213
  by year (n, net %): 2000 5 -1.48 | 2003 9 14.82 | 2004 9 7.91 | 2005 13 2.44 | 2006 8 1.42 | 2007 8 1.25 | 2008 2 -3.07 | 2009 8 9.43 | 2010 8 4.81 | 2011 9 -5.78 | 2012 11 7.91 | 2013 8 8.07 | 2014 11 5.26 | 2015 13 7.14 | 2016 7 -4.41 | 2017 10 8.85 | 2018 7 -3.48 | 2019 8 0.96 | 2020 6 1.92 | 2021 10 10.04 | 2022 2 -3.69 | 2023 9 4.54 | 2024 13 18.28 | 2025 10 5.13 | 2026 9 6.48
  control (random entries on dips in an uptrend: close > SMA200 and < SMA5; p 1.475e-1; 500 runs, 0 s): mean trades 215.46, median avg 0.346 %, 5-95% 0.164 .. 0.503 %; real avg 0.492 % at percentile 93.20
  [PASS] 1 PF >= 1.2 in validation: VAL PF 1.82
  [FAIL] 2 >= 50 trades in each period: 36 / 86 / 91
  [FAIL] 3 t >= 2.5 on development: DEV t 2.32
  [FAIL] 4 above 95th pct of the control: percentile 93.20
  [PASS] 5 positive without top-10 trades: 73.83 %
  [PASS] 6 holdout net positive (partly used): HOLDOUT net 44.62 % PF 2.00
  VERDICT Buried (3/6 gates)

=== RQ2-5: buy at the close when close > SMA200 and RSI(2) < 5; sell at the first close above SMA5 ===
  DEV      n=  17 net 25.73 %  avg 1.513 %  PF 14.11 t 3.23
  VAL      n=  46 net 20.37 %  avg 0.443 %  PF 1.71 t 1.44
  HOLDOUT  n=  48 net 28.84 %  avg 0.601 %  PF 2.63 t 2.65
  ALL      n=111 net 74.94 % avg 0.675 % (gross 0.715 %); PF 2.55 t 3.80; with 0.10% cost avg 0.615 %
  win rate 84/111 = 75.7% (95% Wilson 66.9%-82.7%); trading days held median 3 max 11; open at the end: 1
  the 10 best trades 34.87 %; net without the 10 best 40.07 %, without the 20 best 18.19 %
  worst open loss at a close (open positions included): -10.09 % (trade entered 2011-08-02, exited 2011-08-11)
  next-open fills (descriptive): n=111 avg 0.646 % PF 2.59 t 3.69; by period avg 1.428 / 0.376 / 0.627 %
  whole strategy from 2000-03-03 with T-bill interest while out: growth x3.37 (4.7% a year), max drawdown -12.2% (2011-08-03 to 2012-05-18), volatility 6.8%, invested 5.8% of days; buy and hold x7.68 (8.0% a year), max drawdown -83.0%, volatility 26.4%
  crashes (strategy / buy and hold): 2000-02 dot-com 10.9% / -82.9%; 2008-09 financial crisis 1.6% / -51.6%; 2020 Covid -2.2% / -27.9%; 2022 rate hikes 2.7% / -34.3%
  entry days shared with SPY RS2-10: 50 of 111
  by year (n, net %): 2000 3 8.86 | 2003 4 6.89 | 2004 3 4.34 | 2005 7 5.64 | 2006 5 -2.01 | 2007 6 6.52 | 2008 2 -3.07 | 2009 5 3.74 | 2010 4 3.88 | 2011 6 -0.68 | 2012 4 -3.24 | 2013 3 4.73 | 2014 3 5.53 | 2015 8 4.98 | 2016 5 -0.47 | 2017 3 3.53 | 2018 6 0.10 | 2019 4 2.66 | 2020 2 -0.39 | 2021 7 9.10 | 2022 1 1.53 | 2023 3 2.71 | 2024 5 4.57 | 2025 6 -0.65 | 2026 6 6.16
  control (random entries on dips in an uptrend: close > SMA200 and < SMA5; p 6.450e-2; 500 runs, 0 s): mean trades 106.66, median avg 0.384 %, 5-95% 0.110 .. 0.644 %; real avg 0.675 % at percentile 97.40
  [PASS] 1 PF >= 1.2 in validation: VAL PF 1.71
  [FAIL] 2 >= 50 trades in each period: 17 / 46 / 48
  [PASS] 3 t >= 2.5 on development: DEV t 3.23
  [PASS] 4 above 95th pct of the control: percentile 97.40
  [PASS] 5 positive without top-10 trades: 40.07 %
  [PASS] 6 holdout net positive (partly used): HOLDOUT net 28.84 % PF 2.63
  VERDICT Buried (5/6 gates)

buy and hold from 2000-03-03: x7.68, 8.0% a year, max drawdown -83.0% (2000-03-27 to 2002-10-09)