WeBall

Confidence score

Also called confidence

How much data sits behind a grade — sample depth, book coverage and consistency, on a 0 to 1 scale.

Confidence measures the strength of the evidence behind a verdict, not the strength of the opinion. A model probability built on eighty games and one built on eight are different objects, and showing them identically would be dishonest.

Three things move it: how many relevant historical samples the engine found, how many sportsbooks are pricing the market (a market only one book offers has no consensus to de-vig against), and how consistent the underlying numbers are.

Below a threshold the grade is suppressed entirely and the bet is published as THIN DATA. Above it, confidence changes what you should do rather than only how the meter looks: a high-edge, low-confidence bet is a reason to look closer, not a reason to bet bigger.

Want the full version? Read the in-depth guide, with worked examples and the math.

See also

  • Verdict — WeBall's grade for a single bet — how the model's probability compares to the best available price.
  • Sample size — How many observations sit behind a number — and whether it means anything yet.

← Back to the full glossary