Quick update on the World Cup model following the 2nd round of group fixtures.
My World Cup prediction model is described here and my validation methodology here.
After the 2nd round of fixtures my model has a Brier score of 0.471, but the Bookies are just in front with 0.469 (although I don’t think the differences are statistically significant, yet). The updated probabilities for winning the World Cup now look like this:
I’ve taken off the teams with 0% chance (teams which cannot progress to the next round of the competition). The differences from last time look like this:
A big increase (~4%) for Portugal and large, expected, decreases for Denmark and Argentina. England also have a surprisingly large drop and go down to 8th favourites for the tournament. The impressive performance of Colombia, a potential Round of 16 opponent the likely change here. Interestingly the model actually predicts England have a slightly lower chance (~0.5%) of winning the World Cup than before the tournament began, even though they have maximum points from their opening fixtures.