One criticism of the Arsenal over the last five seasons is that they tend to up their game when the season is already effectively over. Specifically that they play better in the last 10 games of the season compared to the rest of the season. This critique is usually paired with a psychoanalysis of the squad with fans and pundits saying that Arsenal play their best only after the pressure is off.
I looked at Arsenal’s final ten games of every season since the year 2000 and compared them to the season average (which includes those final 10 games). The stats do show that Arsenal have improved performances over the last five years but when I compare those improved performances to the other top teams in the League, Arsenal are actually just playing to their relative average. Let me show you what I mean.
First, the “season” here means “year ending in…” So, “2017” is the 2016/2017 Premier League season. Second, I used the tables at Statbunker to create custom tables based on date. I would have preferred to use Statto but I believe that the site is completely down now. There is one small problem with the tables created in Statbunker: I was only able to use date instead of number of matches played. So, when selecting Arsenal’s final 10 games there were always teams in the League who played more or less matches. It was usually within one or two games, however, and I didn’t feel like putting in the time to make a League table for each of the last 17 years. Besides, if the idea is that Arsenal play better after X event (that they are out of the title race) then that should show in the data regardless of one match difference.
What the data shows is that there has been a points per game improvement and even a goal difference per game improvement by Arsenal over the last 10 games in four of their last five seasons. But the same data shows a drop in performance over the final 10 games in five of the six seasons between 2007-2012.
Between 2007 and 2012 we also see a drop in table position in the final 10 games relative to the other teams in the League. In other words, Arsenal finished 4th in the League in 2010/2011 but if we just look at the last 10 games table they were 12th in form. In 2010 they finished the season 7th in form, in 2006/2007 they were 8th in form, and so on. These data points show that Arsenal “lost the League” and/or stopped being competitive in the final 10 games of each of those seasons.
Conversely, despite the uptick in points per game and goal scoring in the last five years Arsenal’s table position has remained relatively stable or improved. Between the 2012/13 season and 2016/17 Arsenal had just two seasons where they finished the season strong: 2013 and 2017. The remainder were actually the same form as the rest of the season when compared to the League.
This is important because the League isn’t won with a set number of points. So, relative to the opposition, Arsenal have only overperformed in the final 10 games in two seasons (2013 and 2017) and have actually underperformed in 6 of their last 18 seasons. In 10 of their last 18 seasons Arsenal basically performed the same in the last 10 games, relative to the League table, as they had, relative to the League table, in the season. In other words, in the last 10 games of the season Arsenal may score more and win more, but so do all of the top clubs and as a result Arsenal don’t make up any ground.
Arsenal supporters are correct that Arsenal have finished the last five seasons strongly. And this perception is enhanced by the fact that Arsenal finished the previous six (2007-2012) seasons weakly. But the overall record for Arsene Wenger’s Arsenal is one of amazing consistency and his trend shows that his teams are more likely to finish a season in the exact same form as they would throughout.