A famous politician once said that there are three kinds of lies: lies, damned lies, and statistics. Despite this, I will try and use some basic statistics to explore one of the most fiercely debated questions on this website: “how good is Denilson?” Or perhaps more accurately: “is Denilson good enough to play regularly for Arsenal?”

About 7 months ago, another blog featured an article in which they used statistics to argue that Denilson was Arsenal’s best player last season, and the player with the most interceptions in the entire league. I will try and use slightly more in-depth data to compare his performances this season with those of 26 other midfielders in the Premier League.

What is Denilson’s role in the team?

Denilson generally plays as a holding midfielder. His job is to break up opposition attacks by tackling, intercepting loose passes and ‘picking up the scraps’. He must also cover the defenders if they join in the attack. As well as this, holding midfielders play an important role in keeping possession by playing short passes to the defenders, fullbacks and other central midfielders. Attacking contributions- shots, goals, assists- are a bonus for someone like Denilson, but not essential to the job.

How do we compare Denilson to other players?

There are two main ways to do this: compare Denilson to other midfield players at Arsenal, or compare him to similar players at other clubs. I prefer to compare with players at other clubs, because for statistics to be useful we have to compare ‘apples with apples’. However, I have also included statistics for Song, whose role is similar to Denilson’s. I think it is easier to compare the effectiveness of players in the same position, than between teammates with different roles.

There are some problems with this. Not all teams play with a ‘holding’ midfielder- Chelsea play a ‘diamond’ and use players like Essien, Mikel and Ballack who are more all-round midfielders. I have nonetheless included such all-round players in the data.

What statistics should we use?

Given the job description for players in the ‘holding role’, we should look at statistics that measure passing (especially short passing), tackling, positional sense, anticipation, work rate and stamina. We can also briefly look at attacking statistics (shots, goals, assists)- but these are not as important.

However, not all these statistics are available on the net. Using www.soccernet.com, the Guardian’s online chalkboard and the Telegraph’s online database, I compiled some data on height, weight, passing, tackles, interceptions, shots, assists and goals. These were all adjusted for the amount of minutes each player spent on the pitch, to avoid bias towards those who had played more.

I have sorted these into three sections:

1. Possession contribution (percentage of passes completed and number of interceptions per 90 minutes).

2. Defensive contribution (height, weight, percentage of tackles won)

3. Attacking contribution (shots, goals and assists per 90 minutes)

I then manipulated the data to get a ‘percentage’ score for each player in each of these sections.

Results

I have included only the final table showing how each player performed in the three categories, and their ‘overall contribution’ across all three. This overall score was weighted with 10% to attacking, 50% to possession and 40% to defensive and physical contributions.

If anyone is interested in seeing the full results, I would be happy to send you the original Excel spreadsheet.

Discussion

These results indicate that Denilson has some areas of great strength, and others where he is very weak. In particular, he is one of the best in the league in the ‘possession’ category, due to his accurate passing and ability to intercept the ball. It would not necessarily be accurate to compare him to players like Fletcher, Ballack, Essien and Mikel. While they might be better players in an all-round sense, they make higher-risk passes, and do not intercept enough passes to be compared.

To be fair to him, we should only compare him to players who are also strong in this area: Song, Mascherano, de Jong, Jenas and Karl Henry of Wolves. With the exception of Jenas, who is an all-round midfielder, these players are all holding midfielders. They do roughly the same things Denilson does best, to about the same level as he does.

So how does Denilson compare? His defensive statistics (tackles, height and weight) are significantly worse than any of the other players mentioned, and in fact are the worst of any midfielder on the list. His attacking contributions are higher than Song and de Jong, about the same as Henry, and much lower than those of Jenas or Mascherano.

Conclusion

Denilson has it tough. His midfield role is perhaps the least glamorous position on a football field, and it is often difficult to appreciate the short passes and interceptions that are his strengths. However, he is not the only one doing the dirty work- most clubs have at least one, and often two, holding players. The statistics above suggest that Denilson’s interceptions and passing are indeed among the best in the league. In fact, they were even better last season.

However, his defensive contribution is the lowest out of any of the 30 listed players. He wins the lowest percentage of tackles, and at 68kg is the lightest player on the list. Similarly, his attacking contribution is in the bottom third.

The question then becomes: is this good enough? The obvious answer is “no”. Arsenal is an established top-four club with title aspirations, and it seems that the side must have “top-class” players to compete. While Denilson’s short passing and interceptions are very good, he has serious defects as a defensive midfielder. However, Denilson is young and will improve his tackling and strength with experience and conditioning. He has a role to play in Arsenal, but maybe not a permanent role in the starting line-up.

So what can we say for sure? Statistics do show that at least some of the players mentioned above can do Denilson’s ‘fetching’ job as well as he does, while offering more of a defensive presence and attacking threat. Moreover, Denilson is not significantly ‘better’ overall than similar players in teams stuck in mid-table or relegation trouble. For someone who last year played the most minutes for the club, and is an automatic starter when fit this year, fans can expect more.