As expected the first round of group matches has been low scoring, with an average of 1.83 goals per game, this is even lower than the historical average 2.10. Strangely though we haven’t seen a 0-0 draw, with match results clustered around a few outcomes: 1-0, 1-1, 2-0 and 2-1. My model expected a total of 25.6 goals over the 12 matches, compared with 25.3 from the market, so this low level of scoring wasn’t anticipated to this extent. Only 2 out of 12 matches managed to beat 2 goals.
Usually the defensive shackles come off a little in the later group matches, so hopefully this is the case. I’ll allow for increased scoring in my model.
To assess how my model’s performed against the market I’ve used a Brier score method (explained here in a good article by Pinnacle). The lower the score, the better the model is at predicting.
My model has had some good and bad results – beating the market for Italy and Iceland, but underestimating Hungary and Poland. Overall, so far the model has been close to the market – with a Brier score slightly above for match results and lower for over/under 2.5 goals.
Full match data summary