Recently, I introduced the concept of how important point differential is in a rugby season. As discussed, there is an incredibly strong relationship between a team’s point differential and their place in the standings. Taking that one step further is to create a model of expected number of points in the standings based on what a team’s average points differential per game looks like at a given moment in time.
Why would we want to do that?
Well, simply to try and make some estimations about how many points a team might finish the season with, based on the information that we have at our disposal. Naturally, the more information we have, the more confident we can be with our estimations. With each additional game we can improve the model.
In my last post, I spoke about the common Bonus Point System. I promised to touch on the French system used in Top 14. The one major difference is with regards to awarding a point for tries scored. Under the common Bonus Point System (used in Super Rugby, English Premiership, Pro 12, etc), a team that scores 4 or more tries receives a bonus point. In the French system, a try bonus is available to the winning team if they score at least 3 more tries than the opponent. This difference means that there are far less try bonuses in Top 14 than there are in other leagues. It also means that a team’s first priority is to win at all costs, and then, maybe attempt to score a bunch of tries. That sounds really hard.
Based on this “unique” system, the Top 14 standings are incredibly tight compared to other leagues. Put another way, it also means that the way that a team performs (e.g., point differential) may not be sufficiently reflected in the standings, relative to other leagues.
Using the Expected Points model, you can see below how the 2013-14 Top 14 season turned out:
I’ll walk you through this. Working from left to right, under the green header, we have: the teams listed where they finished in the final standings before the playoffs; their average point differential per game; the actual points they had in the Top 14 standings; and then what the Expected Points model would have predicted. You can see that some teams slightly underperformed or overperformed relative to their expected level. The model isn’t magic, but it does a pretty good job.
When compared to other leagues, which use a different bonus point system, you can see that Top 14 teams would have higher expected points totals. Even poor Biarritz, relegated last season, would have expected an additional 10 points in the standings if they played in Super Rugby!
The lesson in all of this is to not underestimate the strength of some of these French clubs (except Biarritz).