How we test models, software, and claims — and how you can verify them yourself.
Every claim on this site — about a tool, a strategy, or an expected return — must be backed by recorded, reproducible data. We do not publish theoretical “guaranteed” results. If a number cannot be checked, we say so.
Backtesting means replaying a strategy against historical odds to see how it would have performed. It is the only way to sanity-check a model before risking money.
Pitfalls: survivorship bias (only testing markets that still exist), overfitting (tuning to history that will not repeat), and ignoring limits (a strategy that works at $10 rarely works unchanged at $1,000).
The share of bets that win. It is necessary but not sufficient: a 90% hit rate at tiny odds can still lose money if the 10% of losers are large. Always read hit rate together with ROI.
ROI measures profit relative to total stakes — the single number to optimize.
Example: stake $1,000 across 100 bets and end with $1,080 → ROI = 8%. ROI already folds in hit rate, odds, and stake sizing, so it is more meaningful than win rate alone.
A result from 20 bets is noise; a result from 1,000+ bets starts to mean something. Small samples inflate both wins and losses by chance. As a rule of thumb, treat any ROI estimate as preliminary below roughly 500–1,000 recorded bets, and widen your confidence bands as the sample shrinks.
CLV is the difference between the odds you bet and the odds at close (just before the event starts). If you consistently bet at odds better than the closing line, you got “value” the market later confirmed — the strongest known signal that a bettor is beating the price.
Positive average CLV over a large sample is more meaningful than a short-term ROI swing, because ROI is noisy early while CLV reflects pricing skill directly.
The calculator computes exact stake distribution, guaranteed return, and ROI for any set of odds you enter. It does not predict outcomes. Use it to size bets correctly, then record each result (stake, odds, win/lose, closing odds) so you can build your own verifiable track record of ROI and CLV.
Math is exact; markets are not. Execution risk — odds moving before you place the second leg, a bookmaker voiding or limiting your account, line shading — can erase a paper edge. A positive backtest or CLV is evidence of skill, not a promise of profit. Nothing here is financial or betting advice.