I wanted to back up yesterday’s prose about Arsenal’s Premier League comeback with some data. So I took a look at my database, split the season into two halves, and drew up the following table:
First, you might want to click on the table to get it big enough to see the data. Second, to explain the shorthand GFA = Goals for Arsenal, GAA = Goals against Arsenal, SPG = Shots per game, SOG = Shots on goal, GPSOG = Goals per shot on goal, and CONV = Conversion rate (Goals per shot as a percentage, e.g. Arsenal convert 12% of their shots into goals). Anything with an “O” in front of it indicates our opponent’s averages when playing against us.
To give you some context to the numbers above, Arsenal’s defensive averages for the whole 2011-2012 season are at a six year high. For example, the previous high for Arsenal’s defensive conversion was just below 11% last year. Meaning that Arsenal only allowed 11% of their opponent’s total shots to be converted into goals. Similarly with stats like goals allowed per game, 1.13 was the high water mark during this rebuilding project and that also happened last season.
Meanwhile, Arsenal’s “keep-ball” defense is actually doing well and limiting the opposition’s shots on goal and shots per game to below 4 and 11 respectively. Those two numbers are in line with the last three years’ averages.
When you look at the fact that Arsenal do very well at limiting their opposition’s looks on goal and yet the conversion rate is at an all time high this gives some evidence to the common sense notion that Arsenal are a team who allow goals on the opponent’s first or second shot.
With the rates of conversion being so close (12% for Arsenal and a whopping 13% for the opponents) this would indicate close margins of victory (or loss) but you have to consider that Arsenal create on average 6 more chances than they allow the opposition to create per game.
Thus, we have a numerical model of the Arsenal system: keep possession, create chances, limit the opponent’s chances, and (hopefully) win the game. I’m not sure how uniquely “Arsenal” that is, but there you have it.
What is truly amazing, however, is how big the Arsenal turnaround has been in the second half of the season. I intentionally chose to split the season directly in half rather than use a biased split after, say, the Blackburn debacle. Because I felt that would be too easy to pick on in terms of trying to debunk the numbers. So I split the season arbitrarily in half and figured that what we would see is a slight uptick in Arsenal’s defense.
What we see, however, is that the Arsenal are half a goal per game better (defensively) in the second half of the season than they were in the first. Arsenal’s offense has dipped slightly but the team’s defense has gotten dramatically better: shaving nearly an entire shot per game and nearly a half shot on goal per game off the opposition’s offense. This drop in opposition shots is a big deal when you play a possession based scheme.
But what’s really amazing is the change in opposition conversion rate. In the first half of the season Arsenal were allowing 15% of all shots in as goals. That means for every 10 shots allowed (Arsenal’s season average) the opposition were scoring 1.5 goals.* So, to bring that number down four points is a big deal and could mean that Arsenal’s opponents are not getting anywhere near the same quality shots they got early in the season.
With just ten games left in the Premier League season and with Arsenal getting great news that Andre Santos is back playing in the reserves and Jack Wilshere is kicking a ball again, the Gunners could see an uptick in the one number that we didn’t want to see go down over the second half: goals scored per game.
It wasn’t a huge drop but when you see that Arsenal’s attack is mostly generated from the right side this season and you see how Newcastle was able to change Arsenal’s approach by bringing on Perch (who led the Geordies with 5/5 tackles in just 45 minutes and forced Arsenal to spread the ball around the park in the second half) you have to wonder if teams haven’t figured out that Arsenal’s Theo-van Persie combination is a real threat and are playing defense accordingly. Thus, bringing back a real left-sided attacking threat in Andre Santos (who already has as many goals for Arsenal as Gael Clichy had in 230 games) will help open things up for Arsenal and hopefully generate more chances, which Arsenal will (hopefully!) convert.
Regardless how the last ten matches stack up offensively, I for one, am keen to see Arsenal’s defensive prowess grow and hopefully kick that opposition conversion rate below 10%.
Ten clean sheets ought to do the trick.
*Obviously, this means that sometimes the opponents wouldn’t score, would score one, score three, or score 8 goals on 20 shots. It’s an average. I’m not suggesting that there were scorelines of Arsenal 1.63 v. Newcastle 2.12