Sports journalist

A football

Daniel Finkelstein explains how concentration is crucial in calculating the probability of sports results.

Raise Your Game: What is your job?

Daniel Finkelstein: I'm a journalist; my background is actually in political journalism, but I work with statisticians. The best way to describe it is that I can understand someone who speaks 'statistics', even if I can't speak 'statistics' myself. So Dr Henry Stott and Dr Alex Morton who are professional statisticians at Warwick University basically do that statistical work, while I set the questions and interpret their answers and write the column.

RYG: You're a sports fan obviously...

DF: My sporting career is confined to having been substitute in one U15 'B' football match which was cancelled at school. So I wouldn't call myself a sportsman, but I do have a great interest in football and I'm now very interested in the things that I can find out about the game through figures and statistics.

RYG: What first drew you to sport?

DF: Even though I've always been mediocre at football, I've always enjoyed kicking a ball about. Football is a very exciting game, partly because it is low scoring. That means that in a high scoring game the best side will almost always win. But in low scoring games it gives a chance to the team that is not so good, just through a bit of luck, a bit of excitement or just that inspired moment for them to actually win.

One of the pieces of work that we've done shows that football is peculiarly well designed to be an exciting game because it's low scoring and you're never quite sure who is going to win. It's as unpredictable as a team sport between two groups of skilled people can be.

RYG: Do you support anyone in particular?

DF: Absolutely, I've got a share in a season ticket and I enjoy going as much as I can. I shall be going to watch my team, which is Chelsea, play CSKA Moscow in the Champions League this week.

RYG: How did you get involved in Sports Statistics?

DF: During the 2002 World Cup I was listening to Radio 5 and heard a statistician, Henry Stott from Warwick University, who had created a model of the World Cup to see who had the best chance of coming out as winners. This seemed to make a lot of sense, and I persuaded The Times to print a number of his conclusions. One of the first conclusions that we had was that the game taking place on the first day between France and Senegal was actually much closer than people thought. Senegal actually had a 25% chance. Lo and behold Senegal won, thus everyone's attention picked up a bit. I began to realise that using past results was by far the best way of predicting what would happen in the future.

Now of course you can't predict a result, all you can do is say what the most probable outcome is. We thought last season (2003/4), that the chance of Wolves beating Man U was only 6%, and yet win they did, because such outcomes do occasionally happen. But what you can do is get an overall idea of how likely something is. That's good if you want a flutter, but also if you want to understand the game. Using the same statistics we can also look at such things as whether it's true that the form book flies out of the window in derby games and whether the weather affect the football results. We can see whether teams have little flurries of form, and where and when they start and finish. We can see, for example, whether it makes a difference whether Patrick Viera isn't in the team any more. We can do all sorts of little studies like that and get quite interesting results.

RYG: Do you watch matches in a statistical way?

DF: Not completely, but certainly the other day when England were playing Wales, my friend was getting extremely frustrated at the number of times we seemed to be shooting at the target but not scoring. I said to him 'We'll score any second now'... at that point, Beckham had the ball at his foot and a second later it was in the net. Now I can't always get it right like that, but when you know that the more shots you fire at the goal, statistically you're far more likely to score; it's basically a pretty consistent proportion of the shots that go in. Then you do know, you do have a bit of an enhanced understanding of what's going on.

RYG: Are there statistical differences at different levels of football, and other sports?

DF: We haven't looked at other sports, but we have looked at lower levels (of football). We have looked at all the professional divisions, because when we do the FA Cup we need to be able to work out what the result will be if Gillingham plays Manchester United. On our website you are able to simulate the percentage chance of any team in any country, in any division playing a team from any other country in any division.


BBC © 2014 The BBC is not responsible for the content of external sites. Read more.

This page is best viewed in an up-to-date web browser with style sheets (CSS) enabled. While you will be able to view the content of this page in your current browser, you will not be able to get the full visual experience. Please consider upgrading your browser software or enabling style sheets (CSS) if you are able to do so.