Archive | Goalscoring Models RSS feed for this section

# Everybody else is doing it, so why can’t we? Soccermetrics’ foray into expected goals

Expected goals has become a thing — some might say, the thing — in soccer analytics over the last 3-4 years.  It is the one metric that has broken out of analyst circles and discussion groups into the football mainstream, and it seems that every analyst has their own expected goals (xG) model that they […]

# Does the cold really kill goals?

I’m a naturally curious guy, and the other day my curiosity guided me to weather conditions at sporting events.  I was on a site called British Weather Services which is a group of private sector meteorologists who provide forecasts and consulting services for a variety of clients in the UK, sports bettors among them.  My […]

# Maybe goalscoring is Poisson

I see that Chris Anderson has been having fun with nice-looking scatter plots lately, so why let him have all that to himself?I've had my doubts that you could describe goal distributions as Poisson, at least when it comes to deriving derivative expressions from them.  A formulation of the soccer Pythagorean doesn't work with a […]

# Is soccer goalscoring Poisson?

A. Heuer, C. Müller, O. Rubner, "Soccer: Is scoring goals a predictable Poissonian process?", Europhysics Letters, 89 (3): 38007, 2010. [arXiv]Does there exist a formula that can predict the score of a soccer match between two teams?  The answer's not that simple, but three physics researchers from Germany have used German league data from the […]

# Goalscoring variances and league points

While I was writing my previous post on goalscoring variances and Pythagorean estimation, I wanted to know if there were some kind of relation between goalscoring variances and the average number of league points taken per game.  So I extracted variance data, league points (which I then divided by matches played to get an average […]

# Goal parsing tool for soccer results matrix

You might have seen what is typically known in the UK as a "results matrix", which is the collection of all results during a league season.  Here is one example for the Premier League; here is another for La Liga.  It is an elegant and compact way to present results in leagues where all teams […]

# Are there common features of teams with large Pythagorean variances?

My last post has sparked a question in my soccermetric mind: Are there common features in the offensive and defensive goal distributions for teams with large Pythagorean variances?  The Soccer Pythagorean works well at assessing the level of team performance relative to expectations from their goal statistics.  It can even predict point totals within a […]

# A closer look at Ajax’s goal performance

I am so fascinated by Ajax's performance in the league this season, and I'm intrigued that there was such a large gap between the Pythagorean estimate and the actual point total.  To be sure, FC Twente's performance also greatly outperformed their Pythagorean estimate.  But Ajax's record deserves closer scrutiny because their goalscoring record was so […]

# Goal scoring probability over the course of a football match

M. J. Dixon and M. E. Robinson, "A birth process model for association football matches", The Statistician, 47(3): 523-538, 1998.How does the probability of the final score change with the relative strength of the two teams, home advantage, time elapsed, and the current score?  This publication describes what's called a "birth process" model and it […]

# Pythagorean exponents in the 2008-09 English Premier League

A couple of nights ago I presented goal distributions for all twenty teams in the 2008-09 English Premier League season, in an attempt to calculate the exponent that would be used for my expansion of the Pythagorean.  I realized after my calculation that I needed to consider the sum of the squares of both the […]