University of Reading Sports Economists in the media:
See here on the BBC Sport website for the full article: https://www.bbc.co.uk/sport/football/52646128
17 May 2020
Weighted points per game? Unweighted points per game? Going with the table as it stands?
League Two clubs voted to end their season early on Friday, with a “framework” in place to decide how the campaign should be resolved.
But what would have happened if the season had never been halted by the coronavirus pandemic?
The short answer is, of course, that we don’t know, but here at BBC Sport – with the help of Professor James Reade and Dr Carl Singleton from the Economics Department at the University of Reading – we’ve had a go at trying to predict how things might have panned out.
What have the clubs decided on?
Before we delve into the murky number-crunching world of economic modelling, the clubs have already “unanimously indicated” their desire to use an unweighted points-per-game system to decide the final table.
Pending ratification from the English Football League and Football Association, that would mean Swindon Town overtake Crewe Alexandra to claim the title, with Plymouth Argyle staying in the third and final automatic promotion spot.
The four teams currently in the play-offs – Exeter City, Cheltenham Town, Colchester United and Northampton Town – would remain there, but Cheltenham would move above Exeter and into fourth.
At the bottom, Stevenage would stay rooted to the foot of the table – although the clubs have also requested that relegation is scrapped for 2019-20.
Predicting the future
What that method cannot take into account, however, is how the unplayed matches would have turned out.
Crewe still had to play three sides in the play-off race (Northampton, Port Vale and Colchester), while Swindon had a game in hand and only had fifth-placed Cheltenham to play of the sides chasing promotion.
In fact, Cheltenham have a big say in what might have happened – they have a game in hand on the sides around them and still have to face fourth-placed Exeter and Swindon, as well as a derby game with Forest Green.
This is where the University of Reading is able to help.
Professor Reade and his team have employed the same methods they use to forecast things like inflation or gross domestic product (GDP) to predict the results of each of the remaining games left in League Two.
The economists use 10,000 repetitions of their model – which predicts average goals scored for each side in a game – to work out the percentage chance of each team finishing in each position. The percentage drops as you go towards the middle of the division, as there are more chances that a side could finish in a higher or lower position than the one they are predicted.
The Devon dilemma
A huge match was due to take place on 23 March – not long after the league was halted – with third-placed Plymouth hosting their fourth-placed Devon rivals Exeter in one of the most significant derby games between the sides in years.
“You’d see a giant movement in Plymouth and Exeter depending upon the outcome of that match,” Professor Reade tells BBC Sport.
And he’s not wrong. If Plymouth won then Professor Reade predicts they would have an 80% chance of promotion. A loss and it drops to 45%.
For Exeter the odds are even more stark – victory puts them at a 53% chance of going up, whereas a loss drops their chances of automatic promotion to just 18%.
“Given that match was taking place straight away, there are still then a remaining six or so games after that and you can imagine different trajectories from there,” says Professor Reade.
“Plymouth might be discouraged by losing that match and could fade away very badly, but equally their efforts might be redoubled by losing that match and still turn it around, so you have both of those probabilities still happening in some proportion.”
How it would’ve finished – with a twist
Professor Reade’s predicted table is not significantly different to what we are likely to end up with via the points-per-game method, with Swindon winning the title and Cheltenham leapfrogging Exeter. There is one twist, however.
While Port Vale are set to miss out on a place in the play-offs by just a single point – with Carol Shanahan describing the decision to end the season early as “the hardest thing I’ve ever done” – our table has them finishing seventh.
Can Northampton count themselves slightly fortunate to be holding on to that final spot?
Would Stevenage have survived anyway?
Although Stevenage are set to finish bottom, results at that end of the table would prove entirely academic if relegation is scrapped.
Owner Phil Wallace told BBC Sport his preference was to finish the season so they could “play our way out of trouble” – but would they have managed to do so?
While the University of Reading model does still have Stevenage finishing bottom, the chances of them ending up in 24th place have lowered.
“Stevenage remain favourites for the drop, but their probability falls from 85.5% to 62.7%, which is nothing to be sneezed at,” says Professor Reade after factoring in Macclesfield’s latest points deduction.
“Macclesfield go from 6.4% for the drop to 34.2% – the two shifts in probabilities don’t equal out as Morecambe are still down there. Their chances fall with this deduction from 8% to 3.1%.”