Chelsea's unbeaten run finally ended last weekend. Not at the hands of a Manchester City or an Arsenal, or even a revived Manchester United. No, it was away at St. James' Park on Tyneside that Jose Mourinho's men met their first defeat of the season. The same weekend, Stoke City beat Arsenal and West Ham United climbed to fourth place. Southampton follow close behind the Hammers in fifth. These are far from the traditional heavyweights of the modern English Premier League, but these clubs just as much as traditional powers Liverpool, Everton and Tottenham look like legitimate competitors for those last European places.
The basic shape of the English Premier League is being remade before our eyes.
If that sounds dramatic, well, it is. We're only 15 matches into the season. Sometimes weird things happen in a couple months of football. But the indicators are striking and they suggest this is something quite new. In the past five years, the top of the table has been nearly fixed. The "Superior 7" of Chelsea, Arsenal, Manchester City, Manchester United, Liverpool and Tottenham have been beaten out for a top seven finishing position only twice. In 2009-2010 Aston Villa came fifth and Everton fell to eighth, while in 2011-2012 Newcastle snagged a fifth place finish with Liverpool dropping to eighth.
Can we know this early that the shape of the table has changed? There have been week-15 interlopers in the past, from Newcastle in seventh last year to West Bromwich's hot start that saw them rise to fifth a couple seasons past. But in the past, these teams have almost always been riding hot conversion rates and could be identified easily as likely half-season wonders. (West Brom had a plus-3 goal difference but a minus-5 expected goals difference, and they slid out of contention.) The difference is that this season the underdogs have the underlying statistics to stay in contention.
To evaluate team quality, I use a measure called "expected goals ratio" over the last several seasons. Expected goals ratio is a measure of team quality according to chances created and conceded. It is calculated as xG created divided by the sum of xG created and xGA conceded. 0.500 is average and anything over 0.600 is very good.
In the last five seasons, just as there have been only two interlopers in the top seven at the end of the season, there have also been only two interlopers in the top seven of expected goals ratio after 15 matches. Last year Southampton were sixth in the league, bumping United to eighth. And in 2010-2011 a hot start by Newcastle had them 7th over Liverpool in 8th. Expected goals ratio, then, has a much better record at predicting who will finish in the top seven.
Now, after half a decade during which just two unexpected sides crashed the top of the expected goals table, there are four this season alone.
This year, the expected goals table shows these non-traditional powers in strong positions. So far, the underlying statistics are more impressed with Southampton, West Ham, Newcastle and Stoke than with Spurs, Liverpool, Everton or even United. West Ham have attempted 96 shots from the danger zone this season, the fourth best in the league, while Southampton are second with 99. These numbers lap those of the traditional clubs they've displaced. Only Manchester United is within 20 danger zone shot attempts of West Ham (they have 77) while the other sit below 70. In multiple key statistical measures, these unlikely teams have been outplaying their moneyed competitors.
This change affects the competitive ecology of the Premier League. It has been common for the league to have a glut of below average sides making up a desultory mid-table. These teams have no meaningful chance of contending for European qualification, but they should stay in the top division by taking enough lucky points off the top seven and beating up the bottom five. These sides would typically put up expected goals ratios between 0.450 and 0.500. They were below average but not so bad that they became interesting.
In a typical season from 2009 through 2014, there would be five or six teams in this middling brigade. This season, only two clubs have xGR between 0.450 and 0.500. Instead there is a much larger number of good teams and several more bad teams.
The Premier League appears to be shifting from a league with only seven truly competitive sides to one with ten or eleven. It is likely, as Mike Goodman recently argued, that the effects of the league's new television deal are showing up in the league table. None of the newly-competitve sides looks like a world-beater, but neither do they display the sorts of easily-expoitable weaknesses that kept poorer teams in the bottom in past years. Just by using the new money to shore up holes in their squads, these clubs have found a path to competitiveness.
At the same time, the easy mid-table may be disappearing. With more quality teams that should beat up on lower-table opposition, the bottom seven to ten clubs in the Premier League may face a more precarious existence. Sunderland and Crystal Palace may be a step better than relegation-bound Leicester City, but now they are likely to struggle to get points off the newly-elevated midtable. There may be less room for error at the bottom if ten or eleven EPL clubs play quality, competitive football.
It is not yet clear what this will mean for the future of the Premier League table. Manchester United is surely too big and too powerful to be kept down for long. But for teams like Spurs, Liverpool and especially Everton, the wearing away of their economic edge could have lasting effects. One cannot be sure how events will settle when something new starts breaking into the world, but this is a trend worth watching.