149 down, 231 to go. The Premiership so far

This post comprises of a big data dump of six tables followed by a summary for each team. For an explanation of any of the metrics, click the (link) following their initial mention.

First, how the Premiership teams have performed in terms of TSR (link). The higher the number, the better the team controls the ball in the attacking areas of the pitch:

Screen Shot 2012-12-04 at 12.59.43 PM

How about the teams they’ve faced? Here’s qualcomp (link) and, in this case, the higher the number, the stronger a teams opponents have been:

Screen Shot 2012-12-04 at 12.59.51 PM

Next is the qualcomp of the opponents each team has to face before the mid-way point of the season (19 games). Once more, the higher the number, the tougher the opposition:

Screen Shot 2012-12-04 at 1.00.01 PM

The shooting percentage of each team, where shooting percentage = goals / shots on target for:

Screen Shot 2012-12-04 at 1.00.08 PM

The save percentage of each team, where shooting percentage = (shots on target against – goals against) / shots on target against:

Screen Shot 2012-12-04 at 1.00.19 PM

Finally the teams ranked by PDO (link), a measure of the amount of luck each team is getting. The higher the number, the more luck a team has been benefiting from in its matches:

Screen Shot 2012-12-04 at 1.00.24 PM

So, what does it all mean? On to the summaries:

Arsenal have controlled the ball well against tough teams. An easier run around Christmas should restore some normality to the situation, although a top four finish is probably a toss up at best right now.

Aston Villa hover just above the relegation zone despite having one of the easiest schedules in the league to this point. They’ve got a tough run around christmas but over the long run their sh% should improve. Whether that’s enough to keep them up is less certain.

Chelsea, as highlighted at the end of last season (link) struggled to control the ball under RdM. Three games isn’t enough to judge Benitez but the initial results aren’t exactly promising. They have some relatively easy fixtures around christmas but a trip to the World Club Cup could complicate that.

Everton have been excellent at controlling the ball and faced average opposition in the process. Given the way that the top four is split (a top two and then the chasing teams) they only require one of the teams ahead of them to slip to have a chance of sneaking into the top four. The most promising part is a 13th placed PDO, they haven’t needed luck to get themselves into this position.

Fulham are an average team in each and every one of the categories above. There’s not a lot to say here except to expect few changes over the rest of the season.

Liverpool have controlled the vast majority of the ball against good opposition and received not a lot for their troubles. Being 7 points behind 4th place is a lot to overhaul in 23 games but they should at least have a decent shot at the top six. Their sh% is horrendous (14.5%) but that’ll regress towards the league average and help their cause somewhat. Their four christmas fixtures come against teams in the bottom eight in terms of TSR so things may well look brighter by the half way stage.

Man City are the class of the league, once again streets ahead of anyone else. They have relatively easy fixtures around the christmas break and a sh% that may improve slightly. I still think they’re really solid title favourites.

Man United continue to defy TSR and pick up a ton of points despite not controlling the ball. Their opposition in the next four games is, on the whole, pretty average, and they’re going to keep this title race interesting just as long as they keep scoring with 30% of their shots on target, although that isn’t likely. (Over the past ten seasons their sh% has averaged 22.8%, with a high of 25%. The league average has been 21.3%. In essence they’re five times further above the league average than their historical performance. They’re also on course to post the fourth best sh% season in the last ten years (link). I wouldn’t bank on that continuing.)

Newcastle are a team I can afford to be a little bit smug about, as their current 43 points pace (a drop of 22 from last seasons 65) was quite accurately predicted by my shots models prior to the season (link). In reality they’ve continued performing as a mid table team, against below average opposition, but the luck has run out, and their sh% and sv% are middle of the pack. Their christmas opponents aren’t easy (Fulham, QPR, and the Manchester’s) and I’m feeling quite confident about nailing this one.

Norwich have performed like a mid table team whilst facing mid table teams. Their sh% will probably pick up a bit but where they are right now is roughly where I’d expect them to be come the end of the season, a la Fulham.

QPR have by no stretch of the imagination been the worst team in the league in terms of controlling the ball, and they have some winnable games coming up (Wigan, Fulham, Newcastle and West Brom). Their sh% will regress upwards and help out but, realistically, 7-8 points is a lot of ground to make up over 23 games. They’re not a terrible side, and it could be done, but it’s definitely odds against right now.

Reading aren’t good and have some tough games coming up. The game in hand against Sunderland is the difference between being one point from safety and five. Their sh% is likely to cool off, although an improvement in their sv% should compensate somewhat. In all I’d say their chances of staying up are about as good as QPR’s.

Southampton are probably going to make a fist of this. They’ve done ok against the toughest opposition that any team has faced this season. Their next four fixtures (Reading, Sunderland, Fulham, Stoke) provide the opportunity to bank some points, and things could be looking much rosier come the New Year. Their sh% may fall slightly but that should be more than compensated by an increase in sv%. There’s reason for quiet optimism amongst Saints fans.

Stoke defy shots models year after year. Once more they’ve been poor at controlling the ball, against relatively easy competition, and, once more, they sit a very safe distance from the bottom three. They’re being powered by an outrageous sv% of 88%, compared to values the last four seasons of 81, 84, 81, 82. This will drop off but they, like West Ham, probably have the requisite points in the bag to safe already, although they may well suffer around christmas, three of their opponents across their next four matches sit amongst the top five teams in terms of TSR this season.

Sunderland do not look in good shape. They’ve been bad at controlling the ball against mediocre opposition, and granted slight reprieve thanks to an above average sv%. It’s not unsustainably high but it’s likely to regress somewhat. However, with QPR and Reading looking likely to go, they only really need to do better than one of the four teams surrounding them. Their odds of avoiding the drop probably aren’t much better than one in two though.

Swansea have, much like Fulham and Norwich, been average against average opposition. Expect their sv% to drop off a bit and I doubt they reach the 58 points they’re currently on pace for but what we’ve seen so far has been a pretty fair reflection of their ability.

Tottenham are probably my favourites for one of the two available top-four spots. They already have points in the bag, have reasonable opponents over the next four games, and all this despite having an unsustainably poor sv%. That’ll improve as the season wears on and I’d be surprised if they weren’t in the mix for third come the end of the season.

West Brom have had a terrific start to the season. Unfortunately it’s largely smoke and mirrors. They’re in the bottom half in terms of ability to control the ball, and are riding unsustainably high sh% and sv%’s, giving them a league high PDO. They’ll falter over the remainder of the season, but have an outside chance of a top six spot by virtue of points already in the bag. I think they’re more likely to finish between 7th and 11th, still a pretty solid season.

West Ham fans should be happy about points already in the bag. In reality the team has fared poorly against the easy combination of opposition faced of anyone in the league. Their next four games are particularly difficult (Liverpool, Everton, West Brom and Arsenal) and it could be a sorry christmas season. Their sv% is going to fall at some point and the results could be ugly. Fortunately they’re only 16 points away from 38, a record of 4-4-15 (W-D-L) would still get them to 38 points.

Wigan have had a declining TSR for a few years now (0.541 -> 0.514 -> 0.504 -> 0.475) and this trend has continued this season. Alarmingly they’ve faced relatively easy opposition thus far. This poor performance has been compounded by a lower than expected sh%, which should rise a little by the end of the season, but this year is the first time that the underlying numbers have given me cause for concern about their ability to get out of trouble.

In short, I think City will win the league, Spurs and one of Chelsea, Arsenal, or Everton (with decreasing likelihood) will round out the top four, and finally the most likely combination of the bottom three are QPR, Reading and (just) Villa. Thoughts?

P.S. You may well have read already but myself and Richard Whittall have begun the Soccer Analytics Forum. We’re hoping to gather together as much of the community in a place that has room for more coherent discussion than we see on twitter but is less formal than blogging. I’d encourage anyone interested in the stuff I write to sign up and share their idea’s. And as always, feedback and suggestions as to what we can do to improve the forum are more than welcome. James


4 thoughts on “149 down, 231 to go. The Premiership so far

  1. Loved this article, fascinating analysis of strengths and weaknesses enabling predictions to be made.

    One stat I would love to see, which I am sure is impossible to find, is which team is most likely to concede a goal to an ex-player as fans of every club complain that this happens to them the most!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s