clock menu more-arrow no yes mobile

Filed under:

Stats and Projections for Week 36

A statistical projection of the remaining Premier League season.

Michael Regan

This one is going to have to be rather quick. It was an annoying week over here for reasons as boring as they were annoying. It didn't help, of course, that Tottenham Hotspur failed to break down Wigan Athletic in a game they absolutely had to win last Saturday, and then on top of that Wigan scored half of their shots on target. I was out of town and watched the game in a Boston bar that I didn't realize was a Liverpool bar until it was rather too late, so it was unpleasant all around.

It should be noted that despite Spurs' draw, and despite Chelsea winning and Arsenal grabbing a point against the league champions, the season is far from over statistically. Spurs are still roughly 50/50 to make the Champions League. Here's an abbreviated version of the projected table.

Projected Table

Club W D L Pts %Title ΔTitle %Top 4 ΔTop 4 %Rel ΔRel
Chelsea 21 9 8 71 0% 0 81% +12 0% 0
Arsenal 20 10 8 70 0% 0 70% +1 0% 0
Tottenham Hotspur 20 9 9 69 0% 0 49% -13 0% 0
Mid-table - - - - - - - - - -
Norwich City 9 15 14 41 0% 0 0% 0 4% +2
Sunderland 10 11 17 40 0% 0 0% 0 5% +3
Newcastle United 11 7 20 40 0% 0 0% 0 8% +4
Aston Villa 10 10 18 40 0% 0 0% 0 15% -18
Wigan Athletic 9 9 20 37 0% 0 0% 0 67% +9

Arsenal have a crazy easy run-in, but nothing is a guarantee in football. Chelsea are in great shape, except for that pesky Wednesday game against Spurs which is probably going to determine things for the top three. (I'll get to that briefly.)

Stoke City are safe now after winning back-to-back must wins. Wigan are in very bad shape even after the draw with Spurs because their rivals Aston Villa dropped six on Newcastle.

Week 36 Projections

As I said, Wednesday's game looms rather obvious big for the top four race. At the same time, the weekend's games will determine how important Wednesday can be. In each case, the desire and drive of opposing clubs, none of whom have anything to play for besides pride and a paycheck, will play a major role. Arsenal have a gimme against Queens Park Rangers, but the spreadsheets give QPR a real chance at stealing points. Spurs have a difficult home game against Southampton, and Chelsea travel to Old Trafford for another important game against United.

Then comes Wednesday.

I'll have time to run the numbers again before Wednesday so we can know just how nervous to be before that game. (Very nervous, is how nervous we should be.) Until then, here are projections for this week's games.

Club xG xGA %W %D #L
Arsenal 2.2 1.3 58% 20% 22%
Chelsea @MUN 1.3 2.2 23% 20% 57%
Tottenham vSOT 2.6 1.1 66% 17% 17%
Tottenham @CHE 1.5 1.9 31% 21% 48%

That 58% number for Arsenal is so low it's normally the sort of thing I'd search for bugs on, but, again, annoying week. And hey, maybe they'll draw. Other clubs have drawn at Loftus Road in the past. The stats obviously explect all of United, Southampton and QPR to perform at or around their season means. If they're all a little more likely to lose, that's probably a wash for expected outcomes, since everyone's got one of those games.

Outcome W D L
Arsenal 83% 60% 44%
Chelsea (@MUN) 94% 83% 75%
Tottenham (vSOT) 59% 34% 24%
and then Wednesday... --- --- ---
Chelsea (vTOT) 97% 82% 57%
Tottenham (vSOT) 81% 47% 29%

Chelsea can put things away with six points. Tottenham can tank the season with fewer than four points, or get into pretty ok position with six. Arsenal can screw it up, but they probably won't. See you all in the morning for the games.