Contact us for a free demo Contact us
for a free demo
StatsBombIQ StatsBomb Data

Serie A 2016/2017: Half Season Review

By Flavio Fusi | January 13, 2017 | Main

Atalanta's midfielder of Slovenia Jasmin Kurtic vies with AC Milan midfielder from Italy Giacomo Bonaventura (R) during the Serie A football match AC Milan vs Atalanta at San Siro Stadium in Milan on December 17, 2016. / AFP / TIZIANA FABI (Photo credit should read TIZIANA FABI/AFP/Getty Images)

The first match-day of 2017 coincided with the conclusion of the first half of the 2016/2017 Serie A championship. Or, rather, it should have coincided, since the Italian Super Cup and the snow of this cold start of the year, mixed up the schedule with Juventus, AC Milan, Crotone, Bologna, Fiorentina and Pescara played a game less than the others did.
The symbolic turning point of the season remains the best moment to make a partial appraisal of the championship and having fun making some predictions about the league outcomes.


To compensate the different amount of games played, it was necessary to use averages and this explains why expected goals difference and non-penalty goals difference are presented in a per game format.

Expected goals data are from l’Ultimo Uomo xG model created by Alfredo Giacobbe. For an explanation of the model click here (Italian).
How Juventus is superior to everyone else

Even with a game less than their rivals, Juventus are solidly leading the Serie A with a 4-point advantage over Roma. Thus, they have the possibility to go +7 once they will have played their game against Crotone. Allegri’s team is reinforcing its position as favorite for the Scudetto, which would be their sixth in a row. The distance between the Bianconeri and the rest of the league is well depicted by their average non-penalty goals difference: +1.33. A figure significantly bigger than any other team: Napoli stops at 1.00, Roma doesn’t exceed +0.79, while Lazio (+0.53) is the only other team with a difference bigger than +0.5.

Looking at the average expected goals difference, Juventus are still in front of everyone else, even if in this case the difference gets thinner: Roma and Napoli answers with +1.03 and +0.97 respectively to the Bianconeri  +1.05. Juve confirm their leadership also in regards of expected performance, but they have created the gap in the table thanks to their technical superiority over the rest of the league, which allows them to go (once again) above expectations.

As it was the case during past seasons, defence it’s Juve’s secret weapon, since they concede just 0.6 xG per game.

It seems self-evident but Juventus have an exceptional defence because they concede the least shots (8.11 per game) and the least shots on target (2.39 per game) overall, all with a low average xG (0.074 xG per shots allowed)
It seems self-evident but Juventus have an exceptional defence because they concede the least shots (8.11 per game) and the least shots on target (2.39 per game) overall, all with a low average xG (0.074 xG per shots allowed)


The integration of Benatia and Rugani allowed Allegri to compensate for the various aches and pains of the “BBC” (Barzagli-Bonucci-Chiellini, in case you didn’t know) maintaining pretty much the same defensive solidity of last season. From an offensive point of view, Juventus are third in expected performance: with 1.65 xG generated on average, they are distanced by Roma and Napoli and almost tied with Inter. It’s once again thanks to their great defensive phase if they are leaders also in TSR (66.0%) and SoTR (71,5%).


Roma and Napoli: power to the offence

While Juventus rule the league tank to their defence, the second and third team in the table rely on an offensive production well over the norm.

If we do not consider penalties, Roma are just fourth for goals scored (Juventus, Napoli and Roma exceed the Giallorossi 1.63 non-penalty goals per game), but this stat doesn’t do justice to a team who should have scraped together a bigger scoring haul. Indeed Roma are first for expected goals generated, with 1.94 xG per game and they would have probably deserved to score more.

Dzeko and Salah deserve a lot of the credit, since together they have generated 23.22 xG, or around the 63% of Roma total offensive output. The two have developed a nice understanding of each other if we consider that Salah, who leads the league with 8.78 xA, has already refurnished his Bosnian team-mate with 3 assists, who instead is first in xG overall, with 15.19 xG but just 12 non-penalty goals. It is not by chance that the absence of the Egyptian has determined a drop in Roma offensive production and his further unavailability due to the upcoming African Cup of Nations will be a relevant problem (this is why Roma are eyeing up Chelsea’s Charles Musonda and West Ham’s Sofiane Feghouli).


Dzeko’s fallibility in front of the goal is by now a fact of life, other than one of the main causes of Roma’s non-penalty underperformance. Anyway, the Bosnian is able to compensate with a shooting volume of 5.3 attempts per 90. Making a simple projection, if Dzeko maintained the same averages, the same efficiency and the same usage also in the second half of the season, he would reach 25 goals in total, a target which would make him immune to critics.

Napoli, which is second per xG for and third per xG against per game – with a non-penalty goals difference of +19 and an expected goals difference of +18.36 – has pretty much offered performances in line with those expected. Despite Milik’s injury and Gabbiadini’s evanescence, that regardless of a similar usage has generated more xG per 90 than the Polish (0.48 vs. 0.33), Sarri has already found 11 different goal-scorers and above all a Mertens capable of scoring 10 non-penalty goals and 8.80 xG (4th overall in the league).

Mertens’ new role has inevitably and positively influenced the distribution and dangerousness of his shots
Mertens’ new role has inevitably and positively influenced the distribution and dangerousness of his shots

The Azzurri are a football-machine that nearly attempts 18 shots per game, more than a third of which on target, and that still manage to have the best possession percentage (59.1%) and pass completion (87.3%) of the league. It will be interesting to see how the introduction of a pure striker, the newcomer Pavoletti or the close to be available Miik, will affect these already electrifying numbers.

Analyzing the performances of Roma and Napoli at 360 degrees, we could place these two teams pretty much on the same level, even if as of now Spalletti’s squad has a  3-point advantage. Napoli keep a reasonable advantage in terms of non-penalty goal difference (19 to 15), but a team like Roma, which scored 9 times from the penalty spot, is surely penalized by such a comparison.

We need to consider the fact that, as Fiorentina last season, Roma won’t likely score just as many penalties, but at the same time their open play scoring haul seems destined to increase. Hence, there are all the premises to witness one hell of a fight for the second place between the Giallorossi and the Azzurri, considering that, Juventus (which anyway is not unreachable) excluded, these two teams have been at least a step beyond everyone else in the league.

Atalanta and Inter: two European outsider?

If we wanted to build a potential Serie A power-ranking by using expected goals difference, Atalanta and Inter, respectively sixth and seventh on table, would be tied at fourth, immediately after Juventus, Roma and Napoli, with a ∆xG of +13.12 (Atalanta) and +13.27 (Inter).

It’s also surprising how the xG/shots of the two teams is pretty much the same, with 0.98 for the Nerazzurri of Bergamo and 0.96 for the Nerazzurri of Milan. The bigger overall shots volume rewards Inter (17.05 shots per game to. 14.84)  with a greater amount of xG generated, but Atalanta square up thanks to a really solid defence which concedes just 0.76 xG per game and it’s second only to Juventus’.

Ladies and Gentlemen, Juventus excluded, the best defensive phase of the league
Ladies and Gentlemen, Juventus excluded, the best defensive phase of the league

 Atalanta could consider themselves “unlucky” to concede 20 non-penalty goals and with an offensive efficiency (Npg/xG) of 1.05, Gasperini’s team could continue to surprise. It feels weird to say but actually, they could even aim higher.

Inter has yet some work to do (like stopping to attempt 30+ crosses per game) to get back in contention for an European spot, but sure enough Pioli’s arrival has given a concrete steering to their form, since the Nerazzurri have not only recorded 4 wins in a row but also begun to register better xG numbers.

Analizying Inter xG numbers game-per-game, the change of pace after Pioli’s appointent is clearly noticeable
Analizying Inter xG numbers game-per-game, the change of pace after Pioli’s appointent is clearly noticeable

Lazio’s ambitions

With the menacing Atalanta and Inter and very short distances in the table, Lazio are currently fourth, with a point less than Napoli and one more than Milan.

So far Simone Inzaghi’s team has been protagonist of a positive path in the league and they have also harvested an amount of points in line with expectations. After 19 games, the Biancocelesti have a +8.86 expected goals difference and a +10 non penalty-goals difference. Anyway, what they showed so far doesn’t seem to be enough to keep the pace of the top three teams of the table, so it’s much more probable that we will see Lazio contend for the fourth place rather than for Champions League qualification.

It is interesting to highlight how Lazio has been the team with the biggest number of different goalscorers (14): so far Ciro Immobile has emerged above everyone else, thanks to 8 non-penalty goals and 9.08 xG.

The overperformance of AC Milan

When it comes to AC Milan, a certain degree of overperformance persists, which is even bigger if compared to when Alfredo Giacobbe examined the Rossoneri’s situation after one third of the league on the pages of l’Ultimo Uomo. If after 12 games the difference between non-penalty goals difference and expected goals difference was of around +3.5, with half of the games behind, that same difference as increased to +4.11 (their ∆xG equals to +0.89 and their ∆NpG to +5).

To say it in other words: if the performances of Montella’s men will not get better, the only hope they have is to maintain the same level of overperformance registered so far, since among the first seven teams on the table – which are basically the only realistic contenders to European qualification – they are the weakest candidate. Consider that they are the only team among the first seven which does not reach a TSR equal to 50%: they stop at 49.1%, just 12th best in the league.

Does a relegation battle even exist?

As never before, this year relegated teams seem already decided after just one half of the season. The last three teams on the tabele Pescara, Crotone and Palermo, have collected just 28 points overall (even if with 2 games less) and maybe it is not even correct to talk about a relegation battle. And think that Empoli and Cagliari, the worst Serie A team for xG difference per game (-0.93 and -0.79) have respectively a margin of 7 and 13 points over the 18th position.

Crotone (with an average non-penalty goals difference of -1.11=, Palermo (-0.84) and Pescara (-0.84) are all three clearly underperforming and of a considerable amount: evidently they do not have the required quality to survive in the Serie A and, miracles or tecnico-tactical twisting apart, they would be likely relegated to the Serie B.

The rest of the league


Looking at the rest of the league, there are a handful of other things that need to be mentioned. First and foremost the situation of Genoa, 12th in the table is quite strange, since as per xG, they could be even five places above. We can do a similar, yet inverse, reasoning for Torino, the team with the highest level of overperformance of the entire league: they sit at 8th, at least five places above their expected performance.

Apparently, there aren’t many chances for Fiorentina to get back in contention for Europe, all the more so if they will lose Nikola Kalinic (7 NpG, 6.71 xG) to China. If this would have been a “normal” season and if variance would not have played a role – since they are both going above expectations – Bologna and Udinese would be fighting to avoid relegation. Giampaolo Samp’s does not seem ready to take off, while Chievo’s performances will assure them a 9th consecutive championship in the top tier of Italian football.

Finally yet importantly, the most unlucky team of the league needs to be mentioned: after a never-ending series of injuries and an exhausting Europa League campaign, Sassuolo moved from a European qualification to the depths of the table. Di Francesco is struggling to choose his starting XI week-in week-out and even if they scored with 13 different players, they are performing under expectations of a considerable amount.

Article by Flavio Fusi