Football boffins use algorithm to predict League One outcome for Portsmouth, Derby County, Bolton Wanderers, Charlton Athletic & Co as season starts
and on Freeview 262 or Freely 565
And the boffins’ outlook for the new term do not make happy viewing for Pompey fans, as they launch their supercomputer.
The model has John Mousinho’s side finishing a disappointing 10th in the table come next May.
Advertisement
Hide AdAdvertisement
Hide AdAnd if the algorithm proves to be accurate, it will be Barnsley and Derby who are celebrating at the end of the campaign.
The Tykes may be facing EFL charges, but they have been predicted to collect 98 points and finish three points clear of the Rams.
It’s Peterborough who finish third (83 points) with Blackpool fourth (81 points), Bolton fifth (80 points) and Wigan back in sixth on 72 points.
Reading, Wycombe and Oxford are all slated to finish above Pompey, who end up behind the U’s on goal difference with 65 points.
Advertisement
Hide AdAdvertisement
Hide AdThe stattos at online bookies BetVictor are behind the formula, which has Cambridge, Burton and Shrewsbury touted to be relegated to League Two.
It adopts the ‘Monte Carlo’ method to arrive at a number of goals each team score in a match.
That is by using ‘poission’ distributions - a mathematical concept which predicts the probability of events happening in a fixed period of time.
To further bamboozle most of us, the key input to the distribution is the ‘lambda’ value, which represents the rate of an event occurring.
Advertisement
Hide AdAdvertisement
Hide AdExpected goals and expected goals against are considered, with with a simulator running the season’s fixtures a mind-boggling 10,000 times and the average standings calculated.
It may be a while before the approach is taken on by the analysts at Fratton Park, who may settle for using football stats platform Wyscout to do their homework for now...
Comment Guidelines
National World encourages reader discussion on our stories. User feedback, insights and back-and-forth exchanges add a rich layer of context to reporting. Please review our Community Guidelines before commenting.