Reports

Latest reports

• Growth per trade at the wrong win rate: full Kelly, half Kelly, 2% risk, two ways 8 Nov 2026
• CFR on Kuhn poker: bluff rate, best-reply gain and game value after 1,000 to 100,000 iterations 8 Nov 2026
• The luck bar for 10, 50, 500 and 4,096 tries 8 Nov 2026
• How lucky can the best of N useless trading variants look? Three ways 8 Nov 2026
• Chance a losing decision still shows a net win, by plays: exact binomial, two ways 8 Nov 2026
• How much Sharpe ratio luck alone gives the best of N backtests on five years, two ways 7 Nov 2026
• The best of k no-edge backtests: exact Sharpe quantiles and simulation, two years 7 Nov 2026
• Two dice: 36 rolls, 11 sums; and how fast a dice game tree grows, two ways 7 Nov 2026
• Drawdowns of a 55 percent win-rate strategy: 10,000 simulated paths, checked against the streak arithmetic 7 Nov 2026
• The longest losing streak: exact distribution and simulation, 55% and 65% win rates, and 100 coin flips 7 Nov 2026
• Elo vs Glicko-2, simulated: games to catch a 200-point change in strength 6 Nov 2026
• Hypothesis: an agent finds the right "skill" by meaning. Result: method skills never matched 6 Nov 2026
• Does Bollinger band reversion survive out of sample on AUDUSD=X? 2 Nov 2026
• Does moving-average cross (n over 200) survive out of sample on TLT? 2 Nov 2026
• Does price vs its moving average survive out of sample on EURUSD=X? 2 Nov 2026
• Bell's inequality (CHSH): the 16 local strategies, the quantum value and a finite-sample simulation 1 Nov 2026
• E. coli K-12 MG1655 genome: entropy, conditional entropy and compression 1 Nov 2026
• The hypothesess testing protocol, version 1: register fields, the stage ladder and its kill rules 1 Nov 2026
• Sample size for a backtest 30 Oct 2026
• Strategy Lab: 4,096 strategies, 2006-2026 29 Oct 2026