Into The Stats: Around The Premier League...
Home and Away Again
There was some chatter around earlier in the season noting the high level of away wins in the league this season. It seems as much as anything the lack of dominant home teams contributed to this perception. So far no team in the league has drawn or lost fewer than six home league games combined meaning no team has won more than eleven home games and no team will exceed thirteen by the end of this year. We are far more used to at least a couple of teams posting formidable home records. Last season three teams won 14 or 15 home games (Chelsea, Manchester City and Manchester United), the season before had City (17 wins), Liverpool (16) and Chelsea (15), 2012-13 had Manchester United (16) and so on.
In fact you have to go back to 2001-02 to find a Premier League season with such an egalitarian spread of home wins. That year West Ham placed 7th and matched title winner Arsenal's 12-4-3 home record (the positional difference was explained by Arsenal's 14-5-0 away record compared to West Ham's 3-4-12). Overall it is an uncommon situation for one or more of the larger teams to fail to dominate the opposition on their own patch
Overall home wins have exceeded away wins, as you might expect, but only after a temporary early season skew moved towards the latter. This fooled a few into thinking that a home advantage had disappeared (the numbers bottomed out around Christmas) and although a super long term trend is moving the gap together, the evidence for home field advantage is strong and consistent and remains. Currently, and long after this storyline has faded from view, the score is 132 home wins (40%) to 108 away wins (~32%) with 9o draws (~28%). This pegs home to away at 55 : 45.
Looking at this season, this advantage endures as we bounce merrily through the numbers.
- Goals H:A ~54:46
- Shots H:A ~55:45
- Shots on target H:A ~54:46
Also, home teams complete more passes: ~78% to 76%, and have more of the ball:~52% to 48%. What is relevant here is that the shots advantage persisted throughout, and it was more a case of a temporary skew in conversion feeding into the slew of away wins to create a perception that an advantage had softened.
These biases continue to be reflected at team level. At home compared to away, this season every single team in the league has better shot rates (however you slice them) and every team bar Leicester has better shot on target rates. With regard possession and passing percentages, only Tottenham exceed their home totals away from White Hart Lane.
There is comfort at home and it's reflected in the numbers.
Weekly wonderings about Leicester
Leicester have happily thrown the traditional "How to Succeed" playbook out of the window and lead the league despite moderate shot totals (-21 shots, +31 shots on target), a league worst passing percentage (68.4%), bottom three possession numbers (43.1%) and we saw a reconfirmation this week of the fact that their likely 5000/1 league success would be the largest individual odds defied in the history of sport. That's pretty cool and means the chances of aliens arriving to join in the celebrations now seems a whole lot more likely while Elvis might show up and ask if they named the King Power Stadium after him.
Parts of the wider media are also starting to notice the statistical oddities that encompass Leicester too, from bookmakers to newspapers, we've seen a good deal of comment on these issues. Indeed it's pleasing that there are larger organisations reaching out and looking for these stories, they're valuable and informative. Back out here where the most of the good content lives, in a wide ranging analysis, Constantin Eckner pulled it all together with a pretty definitive tactical shift over at Spielverlagerung and Michael Bertin investigated their insanely low opposition conversion rates. He calculated a value for the chance that Leicester would concede so few goals since the halfway point; it's around a 1 in 300 chance. This reflects some wider truth, for when we find unlikely success, it's not entirely surprising that we also find other large odds occurrences taking place. What odds the season long total lack of injuries? What odds finding Vardy and Mahrez and them elevating to this level peaking alongside each other? What odds the huge plus column full of penalties? What odds it all comes together at once? We know that: it's 5000/1.
So next season if your club decides to give up the ball and play directly, on balance, it's more likely that they will end up like the many clubs that have tried and failed with the style before, scrapping for their Premier League lives.
Some have made a comparison with Atletico Madrid and the formulation and execution of different to usual systems. We have far more evidence that Diego Simeone has built something sustainable, his teams have put up solid shooting numbers during the nearly four full seasons he has been in charge of after taking control midway through the 2011-12 campaign. Their defensive solidity is borne of shot suppression and has shown to continue despite personnel changes and is altogether different from the good but not great Leicester numbers. Indeed while reliant on the attacking talents and goals of Antoine Griezmann or Diego Costa and Radamel Falcao before him, that Atletico have been able to compete over multiple seasons in multiple competitions reflects that they are not purely reliant on talent alone; the system is strong, as their heroic resistance of Barcelona testified. One suspects that the removal of any of Leicester's talismanic trio of Jamie Vardy, Riyad Mahrez or N'Golo Kante would hugely impact the effectiveness of their attack; something that has already faded to a degree.
Of course, the West Ham game finally found a heady mix of misfortune for Leicester. Vardy deserved his second booking, but would most referees call that? Maybe not. Most referees probably wouldn't give a penalty for pushing in the box, but they possibly would if they'd just doled out a warning. These things can go both ways, as the last ditch penalty showed, and finally a few went against Leicester but they finally conceded the goals to go alongside the volume of opposition's shots and Tottenham, for a day at least, spy a gap in Leicester's armour. This result and even the format of its arrival were overdue but the warning remains, celebrate Leicester, but unless they blow their summer budget on Cristiano Ronaldo or Zlatan Ibrahimovic, somewhere along the line, the talent gap will be exposed and they will refind their level.
This isn't a football revolution.
Second Half Stories
Most analysts would agree that a league season is an insufficient volume of games to apply sufficient rigour to many analyses but in a short term world, trends can be noted over far smaller samples. Indeed the back end of last season offered small clues to the trajectory of this season's mystery teams, Chelsea and Leicester. Too small to predict their eventual destination, but enough to be able to reflect that there were signs that everyone entirely disregarded in favour of broader takes. This being so, there must be some value in at least noting movement in trends between the two halves of the season. Not everything is shifting narrative powered by the natural oscillation of lucky or "unmeasured" factors.
Here's a handful of the likely contending types:
- Chelsea haven't improved overall. Now seemingly on holiday and operating like a try out team for some of their younger players, their shots numbers haven't moved since the half way point. Guus Hiddink may have brought a safe hand and an early run of decent results, but taken as a whole this entire season leaves them no higher than they deserve. Predictions of a big year next season are likely to be thin on the ground.
- Liverpool have a positive skew on their front end for the first time since the back end of 2013-14, 41% of shots on target since game twenty have turned into goals. How high is that? Well, that's Leicester early season hot, pretty smokin'. The other thing that resembles that season is the opposition conversion: far too generous at 39%. High event stuff, shots, goals and fun. That they needed a seven goal thriller to edge out Dortmund is entirely in character with their whole second half of this season. I'm pretty positive that Klopp can at least bring a little order to proceedings next year, but for now it's harum-scarum stuff with little sign of let-up. The Merseyside Derby against this Everton incarnation has the makings of madness.
- Tottenham's shot numbers are significantly better in the second half of the season. They have been a 20 shot to 10 team all through this stretch and the on target numbers are loopy (7.5 to 2.7). They've improved throughout the year and project extremely favourably going forward, all the while without skewing off their conversions, likely a function of their propensity for long range shooting but nonetheless, married to defensive solidity, hard to criticise.
- Arsenal are a 53% shots team in 2016, which goes some way to understanding that while the autumnal incarnation of the team was good but unlucky, the winter to spring edition is underwhelming. Their squad has been fit enough through most of February and March to make a run at things and you would normally expect a top four contender to be about 5-7% ahead of this rate.While we know their shot profile has evolved into a pro-inside box method, it feels like they've sacrificed volume somewhere along the way. Whether the underperformance in expected goals is a blip that will right itself over time remains to be seen, but for now, perched typically at the arse-end of the top four, it hasn't paid off.
- Manchester United have been a 45% shots team in 2016 and still average under 11 shots per game. It is a fantastic achievement to find a team in 5th place with borderline relegation shooting numbers. Credit to van Gaal. The victory over Aston Villa once more enthralled the faithful.
That'll do for now. We'll find some more next week.
Thanks for reading!