ABSTRACT

Detailed measures of player performance arose first in baseball, due to the sport’s structure as a series of discrete events, most of which involve only one or two players on each team. The credit for a home run goes exclusively to the batter, and the blame solely to the pitcher; a ground ball out may involve the batter, pitcher, shortstop, and (perhaps minimally) the first baseman. While basketball is more structurally complex due to continuous free movement of the players, possessions can be considered discrete events and the result of a shot depends largely on the shooter and his primary defender. By comparison, it is more difficult to assign credit or blame in football, due to the interconnectedness of the players’ actions. On a play-action pass, in which the offensive team fakes running the ball before throwing it, most of the 22 players on the field may have some nontrivial role in the outcome. Gerrard classifies football in the category of “invasion sports,” those which “involve a group of players co-operating to move an object (e.g., a ball or puck) to a particular location defended by opponents (e.g., across a line or between goalposts)” (Gerrard, 2007). Among the complexities of such sports noted byGerrard, themost relevant

of Methods and

for football is that of multiple players working directly together for the same action: two players jointly blocking a single pass rusher, multiple tacklers bringing down a running back, or a cornerback and a safety jointly responsible for covering a receiver. Soccer is even more difficult to assess, given the free-flowing spatial movement of competitors, the multiple possible offensive and defensive actions by a given player (sometimes in rapid succession), and the lack of easily discretized events.