Updated probabilities, 1 May 2016

I’ve re-run my E Ratings model after the weekend’s Football League fixtures to see what effect they’ve had on where each club is predicted to finish at the end of the season. Obviously a few fates have already been decided, but there are still clubs with something still to play for in all three divisions.

If you’ve not seen these graphics before, they show the cumulative probability of where each club could finish, in descending order of their average final league position across 10,000 simulations of the rest of the season. The simulations are generated using each club’s current E Ratings and those of their remaining opponents.

You can think of the ordering of the teams down the left hand side as a “best guess” of the final league table, with the coloured bars showing the relative likelihood of each club ending up in a certain section of the table. You can also see how these graphics looked before last night by following the link in the previous paragraph.


CH probabilities 2016-04-30

Middlesbrough‘s draw gives Burnley the advantage in the automatic promotion race, with tough fixtures remaining against Boro and Derby meaning that Brighton have a tough job ahead of them to overtake one of those two.

Sheffield Wednesday‘s win over Cardiff has finally settled the race for sixth place, with the Owls almost certain to meet whichever of the top three fails to secure automatic promotion in their play-off semi-final. However if Derby lose both of their remaining matches while Wednesday defeat Wolves, then the Owls would meet out-of-sorts Hull instead.

Apart from the trio of QPR, Leeds and Wolves, everyone already knows which half of the table they’ll be finishing in.

League 1

L1 probabilities 2016-04-30

The automatic promotion race is all but over thanks to Burton‘s win over Gillingham: Walsall need to win both of their remaining matches and hope that the Brewers lose at Doncaster on the final day.

Doncaster may be lacking a spring in their step as they are all-but-mathematically relegated: they need to beat Burton and secure at least a 12-goal swing over Fleetwood – who have two matches still remaining – to escape the drop, which is incredibly unlikely.

Fleetwood still have a slim chance of being relegated however, although their superior goal difference over Blackpool means that it would require them to lose both of their remaining games and for the Tangerines to win away at Peterborough.

We skipped over the play-off race where the final spot could go to one of three teams on the last day of the season, all of whom have tough away fixtures. Incumbents Barnsley are in a marginally better position than Scunthorpe on goal difference while Gillingham are two points behind them. Therefore the Gills need a win, but should the other two draw than the former’s own superior goal difference comes into play.

League 2

L2 probabilities 2016-04-30

Wimbledon secured the final play-off place as expected, but above them the automatic promotion race rages on. Both Accrington and Oxford are now almost equally likely to complete the final top three after securing away wins, with Bristol Rovers needing one of them to slip up.

Despite Rovers only trailing Oxford by a point, United’s superior goal difference means that a win – likely at home to relegated Dag & Red – is the only way for the Westcountry side to secure back-to-back promotions next weekend. However both their rivals also end the season at home, so they will need a slice of good luck to make it happen.

It’s worth flagging how tight the division is in mid-table: despite there being just one match remaining, only eight-placed Cambridge are assured of a top half finish. The other four berths are still up for grabs, although it would take a pretty unlikely combination of results to bump Leyton Orient down to 13th. There are only three points separating Orient from 14th-placed Luton, but with the Hatters playing Exeter on the final day it’s impossible for both to move above them.


Note: In case you’re wondering why the bars aren’t always in perfect descending order, such as in the case of Fleetwood, there’s a discussion about this on the blog’s Facebook page here. In a nutshell, it’s to do with which fixtures – and how many – each club has left. For example, if a club has to play against weaker teams who can still overtake them, then this will give them a wider spread of possible finishing positions. On one hand there will be more simulations in which they rack up lots of points from those easier games and climb the table, but the downside is that they fall further in the simulations where they lose to a side that moves above them as a result. This is exacerbated by having more matches left to play than others, as it further increases the number of combinations available.