Magic, uncertainty, and MVP
Jan 13, 2021
Loved this.
Uncertainty in xG. Part 1: Overview by Domenic Di Francesco
The Expected Goals (xG) metric is now widely recognised as numerical measure of the quality of a goal scoring opportunity in a football (soccer) match. In this article we consider how to deal with uncertainty in predicting xG, and how each players individual abilities can be accounted for.
MVP, POSITIONS, AND THE PROBLEM WITH USAGE IN SOCCER by Kieran Doyle-Davis
The crux of the issue is Pozuelo’s efficiency; at a mere 10%, it becomes difficult to get the volume required to have a g+ value reflecting the positive work he does. When you compare that to Vela, at nearly 18% and an even higher g+ burden, it’s no surprise he put up one of the most statistically dominant seasons ever. This is one of the nice benefits of a possession value framework like g+ (or VAEP, PV+, EPV, etc.), it is really difficult to assess the cost-benefit analysis of high leverage passes like the kinds that these high usage players are required to make. But evaluating it on a usage adjusted basis lets us make accurate determinations beyond what we see. Consider all passes with an absolute g+ value of greater than 0.02, that is, any passes that increase (or decrease) a team’s probability of scoring (or conceding) on this possession by 2% or more. Pozuelo racks up the second most positive passing value from these “home run” type passes, but he also accrues the most negative value by a distance. When you only consider the successful progressive passes and passes into the penalty area, he’s a star, but when you consider the lost value of all of his failures, the “net” value being added is a lot less than might be perceived.