OMEGAVerified Kalshi YES/NO settlement results, one frozen public quote per resolved contract, estimated fees/slippage, observed win/loss distribution, and calibrated uncertainty. Transparent sample size and correlation limits—no assumed profits.
VaR and expected shortfall here are empirical per-contract downside observations, not certified forecasts of future tail risk. Wilson intervals assume independent Bernoulli outcomes; multiple commodities settling in the same 15-minute window violate that simplification.
Cluster bootstrap resamples entire settlement-time groups, preserving their observed within-group dependence. Fewer than 20 groups make the resulting interval exploratory only. One-day observations never demonstrate long-run strategy confidence.
| Kalshi asset series | Contracts | YES win % | 95% Wilson win range | Mean YES P&L | P&L SD | Worst one-lot | 5% tail loss | Profit factor |
|---|---|---|---|---|---|---|---|---|
| Loading market-by-market risk... | ||||||||
No ranking or automated selection is inferred from this table. Each group has too few independent settlements for strategy promotion. A positive backtest is not evidence of a tradable edge.
Choose a target confidence level and margin of error. The calculator estimates the number of independent binary-outcome trials needed in the conservative worst case (true win rate near 50%). It does not change or artificially smooth measured volatility.
To tighten uncertainty without reducing raw variance, collect additional independent settlement sessions and dates, improve valid-outcome coverage, and analyze fixed models only against truly new observations. Avoid counting thousands of correlated quotes as thousands of independent trades.
The Brier score here measures Kalshi market midpoint calibration, not OMEGA's proprietary alpha. No real-money live execution is enabled or certified. The separate shadow-YES benchmark is not an automated validated trading strategy.