Offensive performances in the 2010/11 Eredivisie

Now that the Eredivisie is over, it is time to sit back and take a look at some of the performances delivered by the different teams throughout the season. Ajax came out victorious in the end in the unique event of a true Eredivisie final against Twente. Excelsior and Vitesse battled to the very end to see the tiniest of margins decide who stays up and who faces the dreaded promotion/relegation play-offs.

In the end it all came down to the very essence of football: scoring more goals than you allow your opponent to do. This articles kicks off a planned series review articles regarding the past Eredivisie season with a simple, though interesting look the offensive side of the game. It evaluates goals scored, shots taken and shot accuracy to come up with some interesting stats.

 

The data

But before can dive right in it pays off to look at the source of these data. The data dare collected from the NOS website, which offers the parameters mentioned above that are provided by a company called Infostrada Sports. The three main input parameters for this review seem quite well defined, as goals, shots on goal and shots wide are generally clearly distinguished. I’ve assumed that blocked shots are grouped under shots off target.

 

Shots per match

Every goal grew from a shot on goal (let me include headers etc. among shots here), so let’s first look at the teams’ amount of shots per match. The average amount of shots was 12.91, which is slightly lower than the 14.7 presented in a recent article on the excellent Soccer by the Numbers blog, including data from the first half of the Premier League season.

Over the past Eredivisie season, PSV and Ajax were positive outliers with 17.7 and 17.5 shots per match repsectively. Compared to those two, Twente’s 15.8 seems on the low side. Best-of-the-rest positions are for Feyenoord, Heracles and AZ at around 13.6 shots per match. At the other end of the spectrum, Willem II created only 9.8 shots per match, which was only marginally less than the 9.9 shots per match created by NAC, who in spite of this figure manage to spend most of the season in a safe mid-table position.

 

On target percentage

But winning matches may not all be about the number of the shots per game, but also about the quality of the shots created. Let’s therefore look at two other graphs. The first one depicts the fraction of shots that were on target and the second one plots both the amount of goals scored per shot and the amount of goals scored per shot on target.

On average 57% of shots were on target, with top and bottom teams quite randomly spread among the field. Of note is Twente’s low percentage of 54.5% of their shots being on target. NAC, who’d also scored low on absolute number of shots created, were bottom of the table with only 52.7% shots on target. Overall spread seemed much lower in this department compared to the absolute number of shots created, but the difference of 6% more shots on target between for example Heracles and Groningen is definitely of relevance.

 

Goals per shot

The second graph shows the average amount of goals scored per shot (yellow) and per shot on target (blue). It turns out that ADO outperformed the other teams in the amount of goals per shot (0.17) and per shot on target (0.29). Groningen and Roda and Heracles, ADO’s rivals in the Europa League play-offs,  also had  good seasons in terms of the amount of goals scored per shot and per shot on target. NAC made up for their low amount of shots and shots on target, particularly with their impressive 0.25 goals per shot on target.

 

In conclusion

These graphs split out the Eredivisie teams according to several aspects of their offensive game. Ideally, of course, each team would rank among the top teams in the amount of shots, the fraction of shots on target and the amount of shots on target that are converted to a goal. But it turns out that none of the teams consistently scores high in all categories.

Breaking the process of scoring goals down into these different aspects of scoring seems like a valuable tool in identifying a team’s general strategy. Take the examples of NAC and ADO, two teams that seem to aim at creating fewer, though qualitatively superior shots. In contrast, for example, AZ ranks sixth in terms of creating shots, but only fifth from bottom in terms of goals scored per shot, which indicates a more opportunistic offensive approach.

 

 

Finally, a word of thanks to Infostrada Sports, who’ve been very helpful and quick in answering my questions regarding their data.

3 thoughts on “Offensive performances in the 2010/11 Eredivisie

  1. bart

    Funny but I recently looked up stats for strikers in La Liga when it came to shot % on goal and goal % in relation to the shots.
    I looked at the first and second strikers of all the teams and came up with the following which correlates interestingly with your numbers …

    For a good striker they need the following stats:
    % between shots and shots on goal (on target percentage) – .450
    % of goals per shot – .130 to .150
    % of goals per shot on target – .330

    I’ll put up some individual results later tonight and a quick comparison … for now this.
    cheers,
    b.

    Reply
  2. bart

    So … back in March I dived into some stats of strikers/attackers in La Liga, here are some of the results, they include penalties (missed too) and free kicks … I’ve included some of the more interesting strikers that were under discussion at the time …

    Player SH/SG(%) G(%/%) Assists GamesPlayed(start/from bench)

    Adebayor 27/16(.593) 2(.074/.125) 0 5/5
    Benzema 64/36(.563) 10(.156/.278) 3 14/12
    C.Ronaldo 208/80(.385) 27(.130/.338) 8 27/0

    D.Villa 118/50(.424) 17(.144/.340) 5 28/1
    Messi 130/72(.534) 27(.208/.375) 16 26/1

    Caicedo 33/19(.576) 11(.333/.579) 2 17/4
    Rondon 56/26(.464) 12(.214/.461) 1 22/2
    Rosenberg 82/37(.451) 6(.073/.162) 0 23/3
    Drenthe 28/10(.357) 4(.143/.400) 1 12/2
    Llorente 103/41(.398) 15(.147/.366) 1 30/0

    A short overview of some of the more interesting names … unfortunately I can’t find what I had on Baptista … anyway, anything like this available for the Dutch league … would live to see figures for Berg (as a PSV supporter, one who would like to see him play for Racing Santander next season instead of his countryman Rosenberg – ex-Ajax and who was horrible then too!), the brothers De Jong, El Hamdahui (spelling) etc and how they all compare. I’d also like to se the difference between Berg this past season and his stellar season at Groningen.

    See the post above on what are in my opinion some good percentages. Interesting too see the C.Ronaldo vs Messi stats and also at that time the comparison between the new and old strikers at R.Madrid.
    It should be noted that team system doesn’t really influence the percentages, the only thing that could influence the stats a great deal is the taking of penalties where you would expect the shot to be on target and then score the majority of the time. Thus penalty takers will have higher percentages, a bit “unreal” and thus maybe penalties should be taken out and put in their own category with stats and get merit that way (C.Ronaldo would be one, or old Danny Landzaat, although not a striker a great penalty taker.)

    I used the ESPN Soccernet site for the data … I know that their data isn’t perfect but hard to find other sites so complete. I’ll have a look to see if I can find/compute some stats …

    Reply
  3. Tom

    Interesting. How about some sort of ANOVA à la Why England Lose to see how these parameters influence the number of points a team earns?

    Reply

Leave a Reply