Regression Theory

The 2020/21 Premier League season has been notable for the sheer unpredictable nature of results. Aston Villa, a team who stayed in the division on the last day of the 2019/20 season beating runaway champions Liverpool 7-2 on the same weekend that Spurs put six past Manchester United at Old Trafford, or Leicester City beating Manchester City 5-2 away only to lose to West Ham 3-0 a week later at the Kingpower Stadium. This has led to a very unusual situation that with a quarter of the season the top 11 teams are separated by just six points. Up in the top five as of the end of November are West Ham United.

At the start of the season, the Hammers were 4/1 to be relegated. Fan unrest with the owners, a less than spectacular performance last season and few new signings didn’t fill the West Ham fans with much optimism for the season. Two back to back defeats to open the season hardly did anything to defy the bookies predictions. COVID struck down manager David Moyes and since then the Hammers have won five out of eight and sit proudly in fifth place in the table. But can it honestly last? I think even the most ardent West Ham fan would suggest not.

There are often seasons where a surprise package will be at the top of the league after the first month or so, only to fall away in the next few months. Leicester City proved that an underdog, one that is at 5000/1, could pull it off, but for every Leicester City there is a Swansea City, Norwich City, AFC Bournemouth and West Ham United who exceed expectations at the start of the season.

Despite the fantastic start they were all realistically never going to stay at the top of the league, were they?  That is because there is a hierarchy in place, whether teams like it or not.  West Ham United at some point in the season will lose three games in a row and essentially return to their natural level.  That effect is the basis of Francis Galton’s work on identifying Regression Theory.

The term “regression” was coined by Galton in the nineteenth century to describe a biological phenomenon. The phenomenon was that the heights of descendants of tall ancestors tend to regress down towards a normal average over a period of years.  I’d like to take his work and overlay that on football. 

Swansea’s case is an interesting case.  For years they had been a third or fourth tier side in English football.  One of their main claims to fame was changing their name from Swansea Town to Swansea City to reflect the changing status of the town/city.  Back in the 1980’s they rose from the 4th tier of English football to top the old Division One in a short space of time, relying on pay big wages to aging stars for one last hurrah (and pay packet).  As quickly as they achieved fame, they quickly slid down the leagues and only avoided the ignominy of relegation from the Football League due to a police dog biting a player (long, urban myth-based story which in reality has nothing directly related to staying up but, hey, the press loved it).  That is a prime example of regression theory.  Short term boost caused by internal (buying good, if old players) and external (traditional good sides becoming crap – reverse regression theory) factors, followed by longer term return to normality.

Fast foward to the start of the 2015/16 season and eight points from their first four games saw them top of the league, including a 2-1 win over Manchester United. All of a sudden the media wondered if Swansea City could be the surprise package of the season. As if the sudden spotlight blinded them they took two points from their next six games. Their form for the rest of the season was unpredictable – a win at the Emirates and at home to Liverpool but defeats to Norwich City and Newcastle United, both of whom ended up being relegated.

Like a number of other clubs, and dare I say it as a West Ham fan, including the Hammers this season, Swansea reset to the regression base line and finished in 12th place. From the 2011/12 season when they were promoted up until the 2017/18 season when they were relegated their average position at the end of the Premier League season was….12th.

There is, however, one factor that “resets” the regression base line in football, and that is money.  Shed loads of it, to be precise. A small team reaching the Premier League today is like a Charlie Bucket finding the last Willy Wonka’s Golden Ticket.  They will never be poor again and could, with careful management and some luck along the way, stay above the relegation zone season after season. And as every season passes, they become more insulated against the threat of relegation and rather than regressing, progress and consequently, Galton’s theory is a load of old bumf!

And that, ladies and gentlemen, is Regression Theory in a nutshell.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s