Graphic: Team at home in second leg wins

It's an odd way to make a living, combining a football statistics column and political commentary. And not always a convenient mix.

On the day the Iraq war started I was sitting in the office trying to decide whether we should categorise Bolton Wanderers versus Blackburn Rovers as a derby game. And frequently I find myself addressing seminars on education vouchers or some such while a riveting football match is on the television.

It is rare that I encounter someone with a similar professional portfolio. But such a man is Dr Lionel Page, of the University of Westminster. He is one of those behind the riveting politimetrics.com website, which combines the judgment of trading markets and opinion polls to provide political predictions. And he has had a paper published (in the Journal of Sports Sciences) on whether it is better to be home or away in the second leg of a Champions League tie.

Page is also a colleague of the Fink Tank's Dr Henry Stott and Dr Ian Graham, who have been looking at the order of two-leg ties.

Many fans believe that there is an advantage to playing the away leg first and finishing off the tie at home. There are a number of reasons why supporters feel like this. There is psychology - you know what you have to do and the home crowd cheer you on to do it. More concretely, there is the fact that extra time and penalties are played at home. And most concretely, there is the fact that the raw data suggests that such an advantage exists.

Let's go through this. Stott and Graham show, using 1,474 two-leg ties, that the team who play at home second win 55.2 per cent of the time, significantly more than 50 per cent.

Yet is this data quite what it seems? That is where Page's work comes in. For he points out that the seeding of matches distorts the data. The team playing at home second is often the favourite. The Fink Tank calculates that this is the case on 55 per cent of occasions. So you have to adjust for quality to see whether the advantage of playing second is real or simply the result of the better team playing at home second.

This Page did. With the help of Uefa's coefficients, he analysed 6,182 two-leg matches played in Europe since 1955. And he found? That for years there was a real, statistically significant advantage to playing home second, but that from 1986 this advantage ceased being statistically significant.

How could this be? It is hard to be certain, but here is a shot at it. The main reason for it being better to play the second match at home is probably having home advantage in extra time. Over the past century, home advantage has been declining and it may be that this edge is not worth that much any more.

The other may be the way that extra time differs from the rest of the game. The Fink Tank expects goalscoring to rise steadily as the match goes on. But in extra time this process appears to stop and 42 per cent of periods of extra time end goalless. We do not know whether this has changed over time.

Whatever the reason, the data is pretty eloquent. There is no statistically significant advantage to playing at home second.