Pilot metric · results versus expectation
Conversion+
Who turns the quality of their darts into wins—and who leaves victories behind? Conversion+ compares actual wins with the number expected from match performance, opponent strength and format.
Selected scopeLoading…Qualified players—
Why we built it
Chris Dobey’s 2026 puzzle.
Strong performance numbers, but too many tournaments ending immediately. Conversion+ tests whether that impression survives a field-wide calculation.
Full field
Conversion+ leaderboard
Ranked players require at least 20 connected matches. Smaller samples remain visible but unranked.
| Rank | Player | Conversion+ | Actual record | Expected wins | Win difference | Performance+ | Clutch+ | Modelled openers | Matches |
|---|---|---|---|---|---|---|---|---|---|
| Loading Conversion+… | |||||||||
Method
What enters the expectation.
The model is deliberately kept separate from Power+ and Form+. A player can be excellent at darts and still convert fewer wins than expected.
The field-relative level produced in each match. This is the strongest predictor of whether that match should have become a win.
A chronological pre-match Steddy Rating makes a narrow loss to an elite opponent different from a loss to a weaker one.
Match length and event environment change how reliably a performance advantage becomes a result.
The score is pulled toward 100 when the sample is small. Twenty matches are required for an official rank.
Conversion+ = 100 + 10 × standardized win residual × √(matches ÷ (matches + 20)).
Actual result, checkout percentage, Clutch+ and Pressure+ are not used to create the expected win probability. That prevents the result from being counted twice. World Grand Prix matches are excluded until the double-in format has its own calibration.