Predicting a Premier League season’s outcome is more than a numbers game

Jonathan Wilson says he likes Jose Mourinho and Chelsea to take the English Premier League title over Manchester City, but only by the slightest of margins. Andy Rain / EPA

There is a scene in the film Moneyball at which several old scouts, their faces lined by experience, sit around a table looking in horror at the researcher whose statistically driven analysis – they have just been told – will be replacing their gut instinct.

Nobody wants to be a dinosaur, but as stats-based projections proliferate before each football tournament, it is hard not to be gripped by the sense that you do not want them to become too accurate – a large part of the beauty of football is its unpredictability.

A large part of its beauty, too, is that football has both psychological and tactical elements, neither of which – to what I admit is a largely non-mathematical brain – seems difficult to quantify.

Football is not simply won by the “better team”, if you are judging them by adding up their collective ­talents.

One model, for instance, said before the tournament that there was a 49 per cent chance of Brazil winning the World Cup.

I expected them to get to the final, so I cannot claim any greater wisdom there.

But once in Brazil and I had seen them, three things troubled me.

The first two were tactical and inter-related. There was space behind the full-backs, something that had been a problem even in the Confederations Cup, but which I assumed would have been resolved, and there was an over-reliance on Neymar. It was clear that any team that could pack the centre and attack on the flanks would trouble them.

The third was psychological.

I had never seen a team as emotional, playing on such a bubble of hype. It gave cause to consider what would happen if somebody pricked that bubble.

People said that nobody could have foreseen the 7-1 defeat to Germany in the semi-final and, while that may be literally true in that a six-goal margin was shocking, at least four journalists in Britain alone wrote that if Brazil lost it could be by a wide margin.

It is easy to cherry pick examples, and there is probably a confirmation bias that leads us to recall when colleagues have correctly predicted something unusual.

My point is that it is extremely difficult for mathematical projections to take emotional and tactical issues into account. In other words, tournaments evolve, and what seemed certain at the start can soon be cast into doubt by events.

The projections are out again for the English Premier League ahead of the start of the new season on Saturday.

Bloomberg has Manchester City as marginal favourites over Chelsea 38.7 to 38.4 per cent, with Arsenal at 8.5 per cent, Manchester United 7.5, Liverpool 5.8 and Tottenham 0.7 per cent.

They tip Burnley, Queens Park Rangers and Aston Villa to go down, with eight sides having a greater than 10 per cent chance of relegation.

At the top end that tallies with my instinct. City and Chelsea have the strongest squads. While it is not absurd to think Arsenal, United or Liverpool could mount a challenge, they are favourites.

Given that Chelsea's signings seem to have strengthened them exactly where they needed strengthening, and that City may prioritise Uefa Champions League progress this season, I would tip Chelsea, but it is a marginal thing.

It is then a scrap between four sides for the two other Champions League places.

Arsenal should be stronger than they have been for a while but still look short of bite in midfield.

Louis van Gaal has had a positive impact already but United still look short of a couple of players, and nobody knows how Liverpool will react to the loss of Luis Suarez and the pressure of Champions League football.

Then there is Tottenham who, if they were not Tottenham, might be taken more seriously. They have an excellent squad and a promising manager, and if it were not for the years of flattering to deceive at White Hart Lane, there is no reason to think they could not mount a realistic challenge.

At the bottom, a seismic emotional and tactical event has already occurred since the projections were made, namely the departure of Tony Pulis from Crystal Palace.

Before he arrived, they had taken four points from 11 games last season. In 27 games under him, they took 41. His loss is a major blow and immediately puts Palace in danger.

I also fear for Burnley purely because of their lack of resources, and West Bromwich Albion, who have a sense of staleness about them.

The predictions are Chelsea for the title, City to come second, followed by Arsenal and United, with a bottom three of Burnley, Palace and West Brom.

But circumstances change.

By the time the transfer window shuts at the end of August, my thinking may be very different.

Follow us on twitter at @SprtNationalUAE