On Winning and Losing Close Matches

Published at SB Nation.

Part 1: On Winning Close Matches (In Which I Predicted Cardiff City’s relegation):

How well a club does in close matches in the first half of the season does not usefully predict how well the club will perform in close matches in the second half.

I’m not saying that winning close matches is “luck.” In general, I think that winning clubs have usually played better than losing clubs. But whatever it is that leads to good performances in close matches, usually it doesn’t carry over as the season progresses. We shouldn’t use points taken from close matches as a good predictor of future points.

Part 2: On Losing Close Matches:

Even these clubs don’t  usually do better than 3-2-1 in close matches, and they average more like 2-2-1. Spurs’ 8-4-2 record so far i close matches would be among the best on this table, which is populated mostly by title contenders. I’d guess here that anything over 1.8 or so points per match in games decided by under two goals is probably unsustainable. Even Mourinho’s 2004-2005 Chelsea only took 2.1 points per match in this subset of the season.

Advertisements

Testing Expected Goals

Published at SB Nation.

Part 1: What stats best predict goals scored and conceded?

The concept is pretty simple. For every shot, you assign an “expected goals” value based on characteristics like the location on the pitch, whether the shot is taken with the foot or the head, whether the shot is assisted by a cross or through-ball, and so on. This is in no way a comprehensive list of the characteristics of each shot, but it provides a reasonable estimate when dealing with larger samples. A club’s expected goals, then, is the sum of all their expected goals values for all their shots.

Part 2: What stats best predict wins and losses?

The relationship between goals and points is humongously complex. As Howard Hamilton showed in his work on the “soccer pythagorean”, having three unequal possible match results creates a weird, non-linear relationship. So instead of dealing with the math, I’m just simulating the games and comparing projected points to real points. There we should expect a simple linear relationship if the projections are good.

Projections and Expected Goals Methodology

Published at SB Nation.

Michael Caley Premier League Projections

My touchstone throughout the process was “does this make football sense?” I am by training something of a skeptic of regression methods. While obviously I had to do lots of regression to create this system, I tried to make sure I was only running regressions when I understand why and how the factors involved related to the creation of better and worse chances in a football match. This has probably let to some infelicities in the math, but I hope it also means that the logic of the system can be communicated reasonably clearly.