Amazing! I got to meet Chris Robshaw. It was great to hear him talk about leadership, World Cup challenges and even stand up comedy at a #rugby business network event
Well, it seems that Siri does not like Rugby. Only information out baseball, basketball, American football, ice hockey or cricket (!). Apparently golf and tennis to follow…
I have been meaning to post these pictures of the Old Mutual Cup test match between England and Wales on May 29th. It was a great day and England won 27-13 after a very shaky start. In the end England scored 5 tries, and despite the poor kicking of Mr Ford they managed to defeat Wales.
This is a reblog of an article by Bill Gerrard in The Guardian (click on the link to go to the original). I find it very interesting that Rugby is moving forward in this direction. What do you think?
Saracens technical analyst Bill Gerrard is a sports data expert and longtime collaborator with Billy Beane. Here he charts how data is reshaping rugby union
All the top sports team use data these days – and rugby union is no exception
When I started working on data analysis in sport, when I tried to value football players in the 90s before Prozone and Opta, all we had in terms of data was who played, who scored, and basically who was a bad boy. Nowadays all the leading rugby union clubs will use data to monitor fitness, injury prevention and tracking players through GPS. That’s not a surprise: the sport scientists and performance analysts who have come into rugby since professionalism have a background in data because they studied science. However Saracens are probably unique in how much tactical data analysis we do. We have a coaching staff who are strongly wedded to data analysis, and that has a lot to do with the culture Brendan Venter, a practising GP, created at the club. He very much believes in an evidence-based approach to performance and that is what we have implemented.
If I am doing my job well, I need to detect the strengths and weaknesses in my weekly opposition analysis that Saracens’ coaches can use in their game plan. They will look at my report, pick up on things, and then study match footage in detail – and from that analysis they will feed tactical insights to the players in the run up to a game. It is worth stressing that everything I do is fed to the coaches to help support decision-making. Incidentally, I read a lot of analytics work, but it can be quite superficial from a coaching view – sometimes you think so what? For example, knowing that more successful teams typically have played together a lot more is almost a statement of the obvious. And you can’t go out and buy shared experienced.
Apologies to romantics, but the data shows the kicking game works …
When I joined Saracens, one of the first things I did was to show that the coaches’ intuitions about what was the most effective style of rugby was supported by the data: on the whole, teams that use a kicking game more tend to be more successful. It’s one of the basic principles of effective rugby. Now that goes against the purists who want total rugby, and want to play ball-in-hand rugby wherever they are on the pitch. Well, I’m afraid the data doesn’t support that. The less you play in your own half and the more you play in the opposition half, the more likely you are to be successful.
There are really two reasons why that is the case: the more you play in your own half, the more likely you are to be turned over. And the more you are turned over nearer your try line, the more you are likely to concede. Second, the more that you run the ball in rugby, the more collisions and rucks you are going to be involved in and they just sap energy. Incidentally there are certain teams who slightly change the way they play against us – you can see that in the data. Teams kick more when they play against Saracens. The data doesn’t lie.
Deep data analysis increasingly provides insights which can help teams develop successful game plans
The game against Clermont in last year’s Heineken Cup semi-final was one of the best performances in my time at Saracens. And I think it was my best performance in terms of opposition analysis too. We knew how vital Brock James was, so I did an incredibly detailed analysis of his kicking game – the types of kicks, when he made them, from which part of the field and the distance he tended to get. Another important element was deeply analysing the key people in their running game.
But what also came out was that in the seven games prior to that Heineken Cup semi-final, Clermont had only conceded six points in total in the first 20 minutes in those games. They started very strongly and the way they used the kicking game was huge part of that. They would put teams under pressure, get ahead, and in the second half they would adopt a conservative game of keeping what they’d got. And you could see it in the stats in terms of the types of kicks too. They put up bombs far more often in the first half then in the second half. In the second half Brock James would kick for position, he would kick for territory. He was very conservative. That was fed into the coaches and their plan to combat Clermont’s strategy worked perfectly. Indeed, I don’t think Saracens have ever implemented a game plan as well as they did that day.
Data is increasingly custom-made by clubs because it can provide more valuable insights
With a third-party provider such as Opta, I’ll get data on the outcome of every scrum the opposition has had in the league and European Champions Cup. Opta will also tell you if the No8 had a pick and go from the scrum, or passed, or so on. But Saracens’ internal scrum data goes much deeper because the scrum coach, Alex Sanderson, will analyse the technique of every player involved in every scrum and he will pick up on where there has been a failure of technique. Some of it is comment, but that can be transformed into data because it is categorising the outcome.
It is what I call ‘expert data’ because you have an evaluation of technique or decision making. It’s the same with the internal kicking data we use – the kicking coach will evaluate at every given situation how many chases there should have been and go through the video and evaluate how many effective chases were made. Similarly, when it comes to tackles made you’ll get a single figure from Opta, while at Saracens our performance analysts will record several categories of tackles. It all means that we are able to more accurately assess player performance which should help us become a better team.
Data is increasingly important in rugby – and the way technology is going that is not going to change
My academic background is in applied economics and my sporting background is football – I have a Uefa B coaching license. So when I first met Saracens’ coaches their questioned what someone who had never played or coached the sport possibly contribute to a rugby union team. Five years on I am still there, and I have been able to learn from coaches like Brendan Venter, Mark McCall, Paul Gustard and others at first hand. Inevitably I can only see the role of data analysts growing: we increasingly have new technology coming into rugby, which will put more demand on people who can process that data and come up with relevant analysis. But it might be that I am superseded by people who can combine a strong coaching background in rugby union with analytics. After all, I can find patterns in the numbers, tell coaches what the numbers are saying, and pick out where the performance levels are at, but it is the coaches who have the deeper understanding to do something with it.