How quickly can a club’s fortunes turn around? Swansea City finished eighth in the last English Premier League season, but quite a few statistics suggested they benefited from luck. So much luck, in fact, that a few months ago I tweeted something rather pejorative about the team: Now, you don’t have to be an analytics expert to see why I tweeted this. Just look at the table: No team above Swansea had a negative goal difference, and Swansea’s goal difference was the same as West Ham’s, down in 12th place. In contrast with the Hammers, Swansea got their goals at…
Read more →

Back in June, after one of Ecuador’s matches in the Copa América, I tweeted about Jefferson Montero being one of just three players from Swansea City with an above average rating in both of my public models. Now in his peak years for performance on the pitch, he’s a player any Premier League club would like to have on its roster. Moreover, coming from a middling team, he might have been available cheaply to a bigger club acting early in the current transfer window. But was he the only hidden gem in the league? The computer says no. Below is…
Read more →

When I talk about ways of evaluating players, I often add that metrics for players can be combined in meaningful ways for teams. Here are two quick examples. First, ratings from different models can be combined to show who’s most important to a team. The graph below shows 2014-15 ratings from two of my models (explained here) along with Shapley values for the club I support, Newcastle United. Shapley values aren’t as persistent from season to season as the other models’ ratings, but they can help to suggest who’s a misfit on a team and who’s pivotal. The diagonal lines…
Read more →