Unverified50% confidenceBenchmarkExact time
In the benchmark, the random policy obtained an average of approximately 0.56 badges.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
8/29/2026
First Seen
Valid until: 11/27/2026
Sources
Related Entities
Related Claims
UnverifiedWhen a team tests 1,000 combinations of strategy parameters, even if all combinations have zero predictive power, roughly 50 combinations will appear to perform excellently in backtesting purely due to randomness (the multiple comparisons problem).77% similarUnverified一个随机猜测的模型在单次二元预测中有50%的概率押中72% similarUnverified在5%显著性水平下测试100个随机因子，期望会有5个看起来显著（多重假设检验问题）72% similarUnverified对于48支球队上百场比赛的赛制，5万次模拟通常可使冠军概率的标准误差控制在约0.2%以内68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/822090API
curl https://kongchang.com/api/v1/knowledge/claims/822090MCP
get_claim(id=822090)