I returned, and saw under the sun, that the race is not to the swift, nor the battle to the strong, neither yet bread to the wise, nor yet riches to men of understanding, nor yet favour to men of skill; but time and chance happeneth to them all.

Ecclesiastes 9:11

I Expect Gifts for Christmas

What happens if a football team massively over-performs and wins the league by a country mile? How do we know if they won by luck, or if they were just _that _much better than everyone else?

One of the metrics in football that the sports analytics industry developed over the last 10 years to assess performance is the much-fabled expected goals (xG). The idea is that a goal being scored is just one of a number of outcomes that could have happened (shot hits the crossbar, keeper saves, shot goes wide, foul, etc.). Now I don’t want to go too much into detail over the xG definition (FbRef has a great explanation and method of computing it here as there are many differing opinions available on what an xG is, or how to calculate it.

For simplicity, let’s think of xG as an indicator of how well a player has taken a shot (or chance) that might result in a goal. A player with a 1-v-1 chance against the goalkeeper? High xG. A midfielder takes a shot on goal from 40 yards past a wall of players? Low xG. Penalty? High xG. A shot taken from near the corner flag? Low xG.

Professional football clubs have been using xG to track performance for a few years now, as opposed to simply drawing conclusions by looking at the final result. It’s slowly made its way to the media to help inform debate- although maybe try avoiding it when speaking to certain pundits.

#machine-learning #data-science #r #python #football