clock menu more-arrow no yes mobile

Filed under:

Premier League Projections and Power Rankings, Week 14: Bad Wins

Tottenham Hotspur won yesterday at Craven Cottage, but the underlying stats from the game reflect a stronger performance by Fulham overall, trumped by a couple great shots from distance by Spurs and perhaps Hugo Lloris' best shot-stopping of his Tottenham career. Overall, Spurs do not move up.

Paul Gilham

The basic stats tell the story. Tottenham Hotspur had 18 shots, Fulham 15. Spurs put five shots on target, Fulham six. From the danger zone, Spurs took nine shots and put two on target, Fulham had eight with five on target. Previous to this match, Fulham had been massively outgunned by opponents in the most productive areas of the pitch, taking 31 shots from the danger zone against 95 for their opponents. Spurs registering merely even numbers against a club which had previously been steamrolled by West Ham is not a good performance overall.

Now, yesterday's Fulham were not Saturday's Fulham. New manager René Meulensteen overhauled his lineup, making seven changes and instituting new tactics better designed for a squad battling relegation. Out went the suicidally high line and the elderly centerbacks and midfielders who could not play it, out went the attacking midfielders who refused to track back and cover, in came a simpler, deep-sitting counter-attacking set of tactics which left primary creative duties to Dimitar Berbatov as the point man in the attack. So we shouldn't have expected, I don't think, a Fulham performance to match their earlier awfulness. But still, this is not a strong side, and Spurs allowed them way, way too many good chances.

But Spurs were not the only big club to register a bad win by the numbers. Everton's big 0-1 away victory at Old Trafford came significantly against the run of the underlying stats. Manchester United took eight shots from the danger zone, three on target. Everton took only one shot from the danger zone, a blocked attempt by Kevin Mirallas in the 41st minute. Everton did take a bunch of shots from wide areas in the box, six total with three on target, but overall the balance of chances was heavily in United's favor.

You'll see in the table that Everton have moved up with their bad win, while Spurs drop. This is because of the nature of the projections. Spurs' bad win was a win that they were already basically projected to get, I had Spurs taking an average of 2.2 points from the match. Everton by contrast, playing away to Manchester United, were projected to drop points to begin with. So their stealing three points adds more in expectation than the drop in the underlying stats costs them. Spurs salvaged a win they should have taken regardless, which is not worth as much.

Spurs Shot Conversion

As Andre Villas-Boas mentioned after the match, Spurs were probably a bit lucky to win last night. Since I talked about Spurs' struggles converting shots on target, the club has put fourteen shots on target and scored four, at a league averageish 28% rate. Of course the kind of shots they've converted has been weird. Villas-Boas likes his clubs to shoot from distance more than any other manager in the Premier League, and the clubs average shot quality has been consistently low under his watch. Spurs have now attempted 137 shots from outside the box and put 48 of these on target. You would expect about five goals based on those shooting numbers, and so Spurs with five goals from outside the box are exactly at league average. Of course, we've gotten there quite oddly, scoring only one of our first 42 SoBoT and now four of our last six.

Expecting that a club which has underperformed expected goals will then massively overperform to return to league average is a form of the "Gambler's Fallacy." I'm not arguing that I or anyone else foresaw Spurs SoB conversion returning to league average in the course of four days. But it does suggest that it would have been wrong to project Spurs' goal conversion rates to remain in the garbage.

Hopefully the same will be true of our shots inside the box conversion rate. With 39 shots on target from inside the box and 107 total SiB, we'd expect Spurs to have scored 13-14 goals from inside the box, and instead we have only five. SoB/SoBoT overperformance has carried Spurs for two matches while we have still failed to put away good chances from more promising positions.

Projections and Power Rankings

Do remember that because of rounding, not all the numbers necessarily add up quite right.

Club W D L Pts GD Team+ Top4% ΔT4 Rel% ΔRel Title% ΔTitle 5th
Manchester City 24 7 7 79 +54 160 94% +1 0% 0 46% +2 3%
Arsenal 23 8 7 77 +36 140 89% +1 0% 0 33% -1 6%
Chelsea 21 10 7 73 +29 135 72% +0 0% 0 12% +0 12%
Liverpool 20 9 9 69 +26 133 45% +6 0% 0 4% +1 18%
Everton 18 13 7 68 +21 127 40% +5 0% 0 3% +1 19%
Tottenham Hotspur 19 10 9 67 +13 133 32% -4 0% 0 2% -0 18%
Manchester United 18 11 9 65 +19 129 23% -8 0% 0 1% -1 17%
Southampton 15 11 12 56 +7 107 2% -2 0% 0 0% 0 3%
Swansea City 15 9 14 54 +4 108 1% +0.5 0% 0 0% 0 1%
Newcastle 14 10 14 53 -7 90 0.5% -0.5 0% 0 0% 0 1%
Aston Villa 12 12 14 49 -4 88 0% 0 1% -1 0% 0 0.5%
West Bromwich Albion 10 13 15 44 -8 83 0% 0 5% +1 0% 0 0%
Stoke City 9 13 16 40 -14 79 0% 0 14% +2 0% 0 0%
Norwich City 10 9 19 40 -26 76 0% 0 17% +4 0% 0 0%
West Ham United 9 12 17 40 -12 82 0% 0 15% +7 0% 0 0%
Hull City 10 9 19 39 -20 65 0% 0 19% +2 0% 0 0%
Cardiff City 8 12 18 37 -23 68 0% 0 27% +2 0% 0 0%
Fulham 8 7 23 32 -32 64 0% 0 65% -5 0% 0 0%
Sunderland 7 10 21 31 -30 69 0% 0 67% +2 0% 0 0%
Crystal Palace 8 7 23 31 -31 63 0% 0 70% -12 0% 0 0%

I don't have a ton of commentary on the movement in the table. The top three continue to separate themselves, leaving a four-team race for the final Champions League place. Crystal Palace got a good win but remain in the relegation zone. Southampton's top four chances have collapsed after stompings by Arsenal and Chelsea and then a terrible shot conversion performance home to Villa.