Category Archives: Predictions

Week 31 Predictions

Here are the model’s predictions for Week 31. The marquee game this weekend is probably Arsenal v. Liverpool at the Emirates, and I’ve got Arsenal as a 47% favorite to win, with Liverpool 33% to win, and 20% chance of a draw. This one feels about right to me, but I’ve missed two Liverpool results in a row so the smart contrarian money might be on a draw.

The most surprising prediction is Manchester United only being 33% to beat Aston Villa at Old Trafford, with a 49% likelihood of a draw. My intuition tells me Man U should be a significant favorite here, although I’m not surprised by Aston Villa only having an 18% chance to win.

Finally, in the relegation battle, we’ll start with Leicester City v. West Ham United. Leicester having a 13% chance to win is pretty low, but we’re getting toward must-win territory if they want to have any hope of staying in the league next year. Queen’s Park Rangers have a strong chance (74%) to pick up at least a point, and is surprisingly a small favorite against West Brom away, and Burnley is in way over their heads (4% to win) at home against Tottenham.  Could be a big week for QPR, as none of the rest of the teams in the lower end of the table seem to have much of a chance of taking anything from the matches this weekend.

Should be an interesting week, and some non-intuitive predictions so I’ll be interested to see how the model fares against my instinct.

GameHome Team Win %Draw %Visiting Team Win %Actual OutcomeCorrect Prediction
Arsenal v. Liverpool47%20%33%Arsenal 4 - Liverpool 1Yes
Swansea City v. Hull City42%22%35%Swansea 3 - Hull City 1Yes
West Bromwich Albion v. Queens Park Rangers25%45%29%West Brom 1 - QPR 4No (2nd most likely)
Leicester City vs. West Ham United13%14%73%Leicester City 1 - West Ham 2No (3rd most likely)
Manchester United v. Aston Villa33%49%18%Man United 3 - Aston Villa 1No (2nd most likely)
Everton v. Southampton8%38%54%Everton 1 - Southampton 0No (3rd most likely)
Chelsea v. Stoke City62%25%13%Chelsea 2 - Stoke City 1Yes
Burnley FC vs. Tottenham4%32%65%Burnley 0 - Tottenham 0No (2nd most likely)
Sunderland vs. Newcastle13%50%36%Sunderland 1 - Newcastle 0No (3rd most likely
Crystal Palace vs. Manchester City12%31%57%Crystal Palace 2 - Man City 1No (3rd most likely)

Week 30 Predictions

Week 30 predictions and results are in: my model predicted 8/10  games correctly with an asterisk on Crystal Palace’s road win over Stoke City because the model predicted 35% win probability for both teams.

The two errors were West Ham’s win over Sunderland, where the model predicted a 43% likelihood of a draw, although West Ham winning was the second most likely result at 33%. An acceptable error.

The bigger error was missing Manchester United’s road win over Liverpool, which my model had as the least likely outcome at 19%. Not sure what went wrong there, although it largely agreed with the betting lines on the game. so I’m not the only one who got it wrong.

Overall a 9/11 success rate in a system with three outcomes is strong. There might need to be some tweaks, but overall a good week.

GameHome Team Win %Draw %Visiting Team Win %Actual OutcomeCorrect Prediction
Manchester City vs. West Bromwich Albion60%28%11%Man City 3 - WBA 1Yes
Stoke City vs. Crystal Palace35%28%35%Stoke City 1 - Crystal Palace 2Yes*
Tottenham vs. Leicester City75%12%12%Tottenham 4 - Leicester City 3Yes
Southampton vs. Burnley FC75%6%18%Southampton 2 - Burnley 0Yes
Aston Villa v. Swansea City26%12%61%Aston Villa 0 - Swansea City 1Yes
Newcastle vs. Arsenal11%24%64%Newcastle United 1 - Arsenal 2Yes
West Ham United vs. Sunderland33%43%22%West Ham 1 - Sunderland 0No*
Liverpool vs. Manchester United32%49%19%Liverpool 1 - Manchester United 2No*
Hull City vs. Chelsea70%27%3%Hull City 2 - Chelsea 3Yes
Queens Park Rangers vs. Everton23%34%42%QPR 1 - Everton 2Yes