Metric analysis

Introduction to Success Score Measurement

Armed with a collection of powerful AI-driven tools, it’s only fitting that we fit the pieces together to form the seven measurable components that make up our new passing score (listed in order of weight in the formula):

(I) Expected Points Added Beyond Expectations (EPAOE) accounts for 46% of the pass mark. EPAOE measures production against an expected value (using our new expected yards model) and is calculated as the difference between the actual value of a pass and the predicted value of the pass before the ball is thrown, taking into account the probability of each result pass (e.g. completion, incompletion or interception).

(II) Expected Points Added (EPA) accounts for 18% of the pass mark. Instead of quantifying the success of a game in terms of yards gained, the EPA represents success in terms of points added relative to the game in progress.

(III) Percentage of Completion Above Expectations (CPOE) accounts for 11% of the pass mark. CPOE is a derivative of completion probability, which measures the success of a pass relative to the difficulty of the throw. The CPOE function used in the score adjusts for dropped passes.

(IV) Interception probability (INT probability) accounts for 11% of the pass mark. INT probability measures the probability that a pass will be intercepted if thrown.

(V) Expected air points added (Air EPA) accounts for 7% of the pass mark. Air EPA equals the value of a completion plus the yards a receiver is expected to gain after the catch. Air EPA is an indicator of the optimal reward for a pass under the control of the quarterback.

(VI) Expected Air EPA (xAir EPA) accounts for 7% of the pass mark. xAir EPA equals the value of a completion (plus expected YAC), relative to the probability of a completion (e.g. probability of completion).

(VII) Probability of Victory (WP) is not a feature of the model, but is used as an aggregation set weight. On a given play, the offense’s pre-snap probability of victory is used as a weight in the pass score formula, where a probability of victory closer to 50% is equal to one and closer to 10% or 90 % is equal to 0.6.