I have long been a fan of the Grant Fox and Wayne Smith school that your eyes should tell you what is going on during a rugby game, but good data will confirm your views.
But drawing conclusions purely from numbers without context, in any field, is fraught with danger
The world of rugby data is about to take a huge leap forward.
I was put onto this by the BBC Rugby podcast, but the primary source article is by Charlie Morgan, the rugby reporter for the UK Telegraph.
He is worth a read for those who want to keep an eye on the English game, and to his credit is up early watching and commenting on our timezone games on a regular basis.
This new development is called expected points (xP).
Instead of having a database that produces raw numbers counting actions, xP assigns a value to each and every action taken by a player to ascertain contribution to a team score.
The values of each action are predefined (I would think teams and coaches would value different things and therefore actions may have different loadings accordingly) and when totalled give a value of a player’s contribution to a scoring outcome.
Winning the ball back, for example, is a high positive score event.
Here’s the rub: they also provide a value to the cost of errors, which contribute to the scores of the opposition.
Hallelujah, nowhere to hide!
An example given is the wide ranging, offloading Fijian Lock Leone Nakarawa. The article points out that his last season for Glasgow ended with a negative xP score because of how often he gave the ball away.
There are still some clear gaps – how do you score a player who has gone missing, isn’t doing his job or, worse, is leaving it to someone else who doesn’t do it as well and cops the negative points despite trying to do the right thing for the team? From my read, these guys are well on the way to covering this and scoring the game more broadly.
The concept has been in development for some time and the podcast hinted it is in use by the coaches of England and Leicester already. But xP is about to step into the public as the United Rugby Championship (why does Benetton – the product, not the rugby team – leap into my head every time I read that title) use it during their TV coverage. I wonder whether this calm or inflame arguments about players.
The name of the company driving this is Oval Insights (they might be good statistical analysts, but they weren’t up all-night thinking of that name) but their work looks likely to take rugby analysis into the levels we see from some of the bigger world sports – watch a baseball game with an American just for a sense of how sports use numbers to assess outcomes and how quickly the public adopt it in depth.
For mine, you cannot significantly improve anything you can’t correctly measure. Nothing quite like empirical outcomes for assigning personal accountability either.
Rugby union looks like it is about to take a huge leap forward and the URC fans are those who are going to benefit first.