Round 13 saw a clean sweep of victories by home sides. Repeating our analysis of last week with this updated data, we now get a significant increase in the home sides probability of winning as the competition progresses… its time to publish. Notwithstanding a very lucky exact prediction of the Stormers one-point victory over the Brumbies, our model did worse than the Superbru average. We’re still hovering around the top 25% of the leaderboard.

Here are the outcomes for last weekend followed by predictions for the weekend ahead.

Round 12 saw two unexpected results, with very few picking the Rebels to beat the Chiefs or the Cheetahs to beat the Stormers. Since both the Rebels and Cheetahs were unexpected home-side victories, we looked a bit closer at whether teams playing at home are doing relatively better later on in the competition. The plot below shows, round by round, the proportion of home sides winning. The blue line gives the best fit to the data – it looks like home sides do a little better as the competition progresses. It’s a small effect at best, but worth keeping an eye on. Perhaps playing at home partially offsets the fatigue that sets in later in the competition?

Here are the outcomes for last weekend followed by predictions for the weekend ahead.