So we’re near enough a third of the way through this glorious Serie A season and we’re all seeing the storylines unfold. But are these storylines for real? Udinese’s fall? Fiorentina’s Scudetto hopes? On eye value, yes, but advanced statistics offer a deeper view.
Statistics have been met with hostility in football, in recent months, just like they did with baseball, NFL and the NBA, statistics such as xPG (expected goals) are trying to help quantify this chaotic albeit beautiful game we all love. But can statistics quantify such a fluid game? Football is evidently not baseball, as baseball is a hell of a lot more static and easier to quantify, but that doesn’t mean statistics are totally redundant in fooball.
Michael Caley (@MC_of_A) is doing some absolutely stupendous work in the statistics community, providing free advanced statistics to all of his followers and regular updates. I highly recommend you follow him.
The full link to these statistics can be found here. I know, I know. It all looks very confusing right now, but I’ll help guide you through it.
Expected goals
Expected goals can seem a really daunting statistic at first, but it essentially tries to measure the quality of a shot using several different variables. As Caley identifies:
“It’s a method for estimating the quality of chances that a football team creates or concedes in a match. Such factors as where was the shot attempted from? What sort of pass assisted the shot? With what body part was the shot taken? Did the attacker dribble past his defender before trying the shot? How fast was the attacking move that led to the shot? Was the shot off a rebound or from a set play?”
Overperformers
And the XPG winners after twelve games is Roma, steaming ahead of everyone else it seems. However, Roma’s actual goal total is 27, whereas their xG is 21. What does this mean? Well, we can infer from the data that Roma’s current scoring rate is not sustainable and they will eventually regress to their mean as the season progresses. But they lead the charts in goals and xG, meaning they’re by far the best attacking team in Serie A.
The other team that stick out like a sore thumb in this column are Fiorentina. There is a crazy 8.9 goal variation between their goals and their xG. A large portion of this comes from their over performing striker, Nikola Kalinic, who is performing light years ahead of his expected performance. Kalinic found solace during his time at Dnipro, in the seasons of 2011-12 and 2014-15, he scored 10 goals in 19 games and 12 in 23 respectively. Could Kalinic finally be ‘getting it’? Unlikely. A hot start is the most likely explanation for Kalinic’s start, and statistics show he’ll even out towards the end of the season.
It’s actually their goals that explains Fiorentina’s impressive start and as nice a storyline it would be, their attacking production can simply not be maintained near this level. Kalinic will regress, Khouma Babacar showed promise last season but he’s also benefiting from a team performing well over expectations. As Fiorentina’s attacking game regresses to the mean, their league position will suffer. It’s a shame but that’s statistics.
Another over performing team, somehow, some way, is AC Milan. Yep. The Rossoneri are currently striking above their xG tally, which is quite remarkable to anyone who has seen Milan play this season. The individual brilliance of Carlos Bacca, there should be a worry amongst fans of the red and black as to where their goals are going to come from if Bacca goes on a dry streak (which inevitably he will). The joint second goalscorers at Milan include Bonaventura and Luiz Adriano (who is looking more and more like a bad buy), so there’s even more reason to worry than originally thought at Milan. They’re overperforming offensively!
Underperformers
Inter Milan have played some woeful attacking football this season. With Mauro Icardi not taking that next step that he showcased last year, scoring three goals in ten games, a poor continuation from his capocannoniere form last year, whereas Stevan Jovetic looks like he’s still getting used to first team football again. The Nerazzurri’s defensive displays have bailed them out this season and have usually only required one goal to win a game – as illustrated by Inter Milan’s SEVEN 1-0 victories so far this season, meaning they’ve won more games 1-0 than they haven’t won games one nil.
There’s reason to look up for the Champions League hopefuls, as their attacking displays will eventually regress to the mean and Inter fans can expect more fruitful attacking displays as the season goes on, as I suspect this team glues and finds more of an ‘identity’, which Mancini is struggling so desperately to find.
The Bianconeri are not only under performing on the pitch but also in the statistical department. Juventus have a similar xG to Napoli, yet there’s a staggering seven point difference between the two sides. It’s been a bad combination for Juventus that they’re under performing, and their Scudetto rivals are over performing, Juventus are performing quite close to their xG total, so the goal scoring will pick up in Turin, but more importantly their rivals will slow down.
Expected goals against
This statistic is pretty self explanatory in that it’s the quality of chances the team gives up, compared to what it creates.
Overperformers
Holy cow, Inter Milan. Mancini’s men are on the two opposite ends of the spectrum in opposite departments. While Mancini’s defensive transformation of this Inter Milan side, but it actually appears they’re simply over performing in that area. In fact, we can eventually expect to see the Nerazzurri’s defence completely capitulate, considering their xGA is almost double the total of their actual goals conceded.
There’s not too many over performers at this stage of the season in the defensive areas in terms of the big boys, but the under performers? Oh boy.
Under performers
Despite boasting the fifth best defence in Serie A even with their horrid (by their standards) start, Juventus’ defensive unit is massively underperforming. An eventual regression to their mean xGA of 6.4 should help Juventus regain their crown of the defensive Kings of Serie A. So chin up, Bianconeri. Your defence isn’t as leaky/unlucky as the footage suggests.
Bad news, Milan fans, your attack is over performing, good news Milan fans, your defence is underperforming! Anyone who’s seen Milan’s defence this season can vouch for the fact it’s an absolute mess. When presented with the fact Milan’s defence is underperforming, it begins to make sense as we look for reasons. Twenty year old Alessio Romagnoli is playing on his second full season as a starter, but is making impressive strides with every game, but it’s more so Milan’s failure to establish a centre back pairing, and a stable back four. Stability increases solidarity within a defensive unit, and shifts by Mihajlovic and unfortunate bites from the injury bug have prevented Milan from creating anything defensively stable.
Lazio are the most notable standout on this on the xGA list. You know how they say stats are nothing without context? Well this beautifully applies to the Biancocelesti. While Pioli’s men have conceded a staggering amount more than they should’ve (20 actual GA, 12.6 xGA), several results easily explain this. A five-nil defeat at the hands of Maurizio Sarri’s Napoli indicated this season might just be a bumpy one for Lazio, of course the predecessor to this defeat two weeks earlier was a four-nil defeat to Chievo. That’s nine goals in two games. Lazio’s defence isn’t nearly as bad as it’s made out to be and will -guess what?- regress to the mean as the season goes on. Although the absence of Stefan de Vrij is a tough one
Strength of Schedule
This represents how much more difficult or easier a team’s schedule has been. An SoS number of 1.10 means a schedule about 10 percent harder than average.
I bring good news for Milan fans! Yes! Two pieces of good news and only one piece of bad news. Milan have actually had one of the toughest schedules in Serie A (and have Juventus in Turin coming up this Saturday), so we can infer Milan actually aren’t as incompetent as they’re showing right now.
The statistics continue to betray Fiorentina, as their schedule has been one of the kindest in the league, combined with an overperforming attack, it’s safe to assume La Viola will come crashing down soon enough.
Inter Milan have more cause for concern along with their soon to regress defence, as their schedule has been the third easiest in the league. And this team doesn’t even seem like it’s properly gelled yet and still has holes that needed to be addressed. Yikes.
Oh Sampdoria. Why did you think it was a good idea to hire Walter Zenga? Not only have you been riding the momentum from last season, but the Blucerchiati currently reside in tenth place having played the joint easiest schedule in the league. Combine this with a massively overperforming attack (19 G, 11.5 xG), it seems as though tenth place is kind (we’ll get to this in more depth). Vincenzo Montella was hired and was subsequently spotted awkwardly cuddling Massimo Ferrero, so there’s hope for Sampdoria yet.
Adjusted expected goal difference
Fairly simple. This is the ratio of expected goals scored to expected goals allowed, with the strength of the schedule played taken into account.
First we have to establish as what relevance goal difference has in terms of league position.
Unfortunately, I am not talented enough for such a group, this was made over here. It’s also a different league and a past season, but it still helps illustrate my point.
There’s a correlation that cannot be ignored, as goal difference goes up, so does league points, and thus league position and the chance of winning a Scudetto.
In Serie A, 22 men kick a ball and in the end, Juventus win. Of course, this does not mean Juventus are going to win the Scudetto simply because they top the xGD table, there is not a total correlation between GD and league points, but there’s a strong one. But the numbers throughout the board clearly indicate Juventus will get going soon enough and will be around the Champions League places come May. More good news for Juventini.
It seems there’s three elite teams in Italy this season, Juventus, Roma and Napoli. In fact, their xGD is so much higher than the rest of the league, the Champions League places may be decided by April.
Fiorentina’s absurd xG is illustrated in the difference between their GD and xGD and although La Viola are enjoying a fairytale start to this season, they’ll struggle to maintain to be in the Champions League places come May considering the gap between theirs and Napoli’s xGD.
Sampdoria, shockingly, look to be close enough to the relegation zone, boasting the third worst xGD in Italy, with distance between them and Genoa. In fact, Sampdoria are completely over performing in every advanced statistic, and a crash ‘n’ burn second half of the season could be on the way.
A word on Udinese
Ho. Ly. Crap. I honestly had to rub my eyes at these numbers for Colantuono’s team. Adjusted xGD has them finishing near enough in the European places, yet the current Serie A standings show Udinese just a point above the relegation zone. There’s a staggering 8.7 swing between their GD and xGD but the main problem seems to come from their xG. With a 5 goal variation between G and xG, which, if they were performing up to expectations, Udinese would boast far better than the joint second worst attack in the league (quite the contrary, they should have the fifth best attack in Serie A). But it’s also a combination of their defence underperforming, too (albeit not as badly). It’s a combination of two factors that make for an awful storm for Udinese, but we can only expect them to shoot up the table in the second half of the season.
The post What can advanced statistics tell us about this Serie A season? appeared first on IFD.
from IFD http://ift.tt/1NFIQNM
via IFTTT berita selengkapnya Follow @jadwalliga
