Arsenal’s had an abysmal end to 2020 saw them drop to eighth, sneaking into the Europa League via a stunning FA Cup win over Chelsea. After, I couldn’t help but think: Thank God that’s over; it can’t get any worse than that.

Well. Here we are.

There were positives to take from another struggle-filled 2020-2021 season, including improved form after Emile Smith Rowe’s introduction as a starter, Bukayo Saka’s continued rise into the stratosphere, and even some semblance of defensive consistency for the first time in what had felt like forever for AFC.

But that’s not what we’ll remember. What will stick out about the season is finishing 8th, missing European football for the first time in nearly three decades, a rash of injuries to the likes of Gabriel Martinelli, Thomas Partey, Calum Chambers, Pablo Mari, Kieran Tierney, and even Pierre-Emerick Aubameyang’s absence due to malaria. MALARIA! What an event that was.

After the win over Brighton, we all agreed to throw 2020-2021 in the bin and move on. But dissecting the season can still offer some value. So with that, let’s take a look at Arsenal’s team data for the year and figure out how everything went so wrong.

Disclaimer: What follows is a lot of data. Team style absolutely affects some of these ranks, and not all teams are created equal. But I’ll attempt to use each point to establish a larger theory of the season. Hope you’re able to follow along!

Arsenal’s Defence: strength or not?

Let’s start with the area of the game where data has the hardest time truly quantifying quality. Arsenal’s defence allowed the third-fewest goals last season, which is the most important statistic, right? After all, the league champions allowed the fewest goals, and the second top in the category made the Champions League.

But what else can data tell us about Arsenal’s defence? Well, for one, it wasn’t very aggressive.

Arsenal were dead last in the Premier League with 315 tackles won, a statistic that on its own may not say a lot since Leeds and Southampton topped the table. But they were 19th in number of successful ball pressures, and just 11th in successful pressure rate. Manchester City were also near the bottom of the league in number of ball pressures, sitting 16th, but they were highly effective when they did press. They recovered the ball 32 percent of the time, best in the league ahead of Leicester, Liverpool and Brighton.

Arsenal were 17th in the league in number of balls blocked, and they were 20th in the combined “tackles+interceptions” statistic.

Some of this is likely a function of Arsenal possessing the ball for more than half the game, right? Well, maybe.

Among clubs with at least 50 percent ball possession last season, no club approached Arsenal’s lows in tackles and successful ball pressures. Liverpool tackled about once more per game and City pressures about three times more per game, the next lowest in each category. But both of those clubs were over 60 percent possession, and they also happened to rank in the top four in terms of successfully pressuring the ball. Basically, they completed more defensive actions in less time out of possession. Whether it’s by design or failure to execute, Arsenal just didn’t put enough pressure on the ball to win it back.

One more defence-related stat: Arsenal created 10 shots and one goal last season with defensive plays. The shots created number is better than only West Brom and West Ham (and tied with Everton). Perspective: Leicester created three times as many shots and five goals with defensive actions. City were just below LCFC in shots created and also created five goals through defensive actions. Again, that’s showing a lack of successful pressure on the ball that could be creating opportunities.

Ball possession: This club needs a ball carrier

Moving right along into ball control, Arsenal were largely successful at holding on to the ball. Their 319 miscontrols were about 50 better than second-place Tottenham’s record, and their 361 times dispossessed were just behind league-leader Burnley. That’s good, right?

The difference between Arsenal and Spurs, two of the best in the league at simply not losing the ball, is how they acted once they had it. Spurs successfully dribbled over 100 times more than Arsenal – nearly thrice a game, and good for sixth in the league behind Fulham (wow), Wolves, City, United and Liverpool. Arsenal’s dribbling record placed them down in 17th, just behind Leeds and ahead of only Burnley and relegated Sheffield United and West Brom. The Gunners simply lacked that player who could consistently beat his defender with the ball at his feet. Their most frequent dribbler, Bukayo Saka, was only in the 38th percentile among attacking midfielders in dribbles per game, and in the 34th percentile in successful dribble rate. Their next-best, Nicolas Pepe, was in the 75th percentile in number of dribbles completed, but just 11th percentile in success rate. (Note: Pepe played about 10 games’ fewer minutes than Saka this season).

Some good news: Arsenal fared well in ball carries, ranking behind only City, Chelsea, Liverpool and United in successful carrying distance. They were also fifth in carries into the penalty area and seventh in carries into the final third.

The data just seem to indicate that a player who can make something happen with the ball at his feet would be a major help. Yves Bissouma, for the record, ranked 24th among all Premier League players in dribbles completed last season, and he converted them at a 65.8 percent rate, slightly above average for a midfielder. Andre-Frank Zambo Aguissa from Fulham ranked third and was successful about 73.1 percent of the time, a 72nd-percentile rank for a midfielder. For the record, James Maddison was in the 43rd percentile in quantity of dribbles among attacking mids.

Passing and creation: A mixed bag

It’s difficult to know just what to make of Arsenal’s passing numbers. As I’ve documented well here, the club passed somewhat progressively this year, ranking sixth behind some of the usual suspects (Liverpool, City, Chelsea, ManU) and one surprise (Leeds). But the data also show that perhaps there’s some difference in the quality of their progressive passing versus their rivals.

Arsenal’s progressive passing distance, the total length of all their progressive passes, ranked just 13th in the Premier League. For reference, Leeds ranked fourth in that category. The top four in progressive passes completed made up the other clubs in the top five in progressive distance.

The explanation? Most of Arsenal’s passes were obviously short or medium length, where they ranked fifth and fourth in the league in number of passes attempted, respectively. The most progressive passing team by distance in the league, Liverpool, launched about 25 percent more long passes than Arsenal, on the other hand.

Arsenal fired a decent number of passes into the penalty area, ranking eighth, and into the final third, ranking fifth, but those are two numbers that fans will want to see brought up, particularly if no one can move the ball that close to the goal by dribbling.

Another shortcoming? Arsenal’s passes were less incisive. The club registered just 323 key passes this season, which are defined as passes that lead directly to a shot. That number ranked 12th in the league, with only Tottenham doing worse among clubs that finished ahead of Arsenal on the table. Top four in key passes were Liverpool, Chelsea, City and United.

The club ranked 11th shot-creating actions (SCA) and seventh in goal-creating actions. Again, the only high-ranking club to even come close to the Gunners in that area was Spurs, who actually were behind Arsenal in SCA but well ahead in GCA.

But it’s interesting to study the difference between Tottenham and Arsenal in that area, too. Which leads me to…

Shooting: What just happened?

Add up everything below and you get a slightly-above average club in terms of goal-scoring. Arsenal’s 53 goals scored were ninth in the league, somehow just below Chelsea but also below the likes of West Ham and Leeds.

That all jibes with their xG rank (ninth), and their non-penalty xG rank (eighth). But when you start to dig into some of the post-shot expected goal data, it gets pretty rough.

Let’s start with the most basic: Here’s the ball, now shoot it. Did you hit the target? For Arsenal, the answer was no far too often. Their 31 percent of shots on target was 17th in the Premier League. Manchester United led this category, followed by Spurs. These are the clubs who made the most of their chances and shot the ball accurately.

But of course, there’s that issue of the keeper standing there. Sometimes outstanding keeper play can cost you a goal, or two, in a game! To solve for that, take a look at non-penalty goals minus expected goals. How many goals did you score, and how many would average finishing have gotten you?

For Arsenal, this is a pretty bleak picture. The club was 12th in npG-xG. Terrible luck, right? Well, maybe not. For all their successes, Chelsea came 15th in this category, while Liverpool were a woeful 17th. So how can all three clubs score far less than they should have and still have such differing results? For one, the other factors such as midfield and defensive play. For another, consider the cause of the poor scoring numbers.

Both Liverpool and Chelsea put their shots on target more often. Liverpool were 13th in this category, while Chelsea were tenth. So Liverpool and Chelsea, data would suggest, should have scored more goals because their shots were that much more on target. Some of that will be bad luck, some will be goalkeeper play, some of it is just a higher level of finishing needed.

For Arsenal? They just missed the target too often, and you can blame a lot of that on Auba. He took the second-most shots of anyone in the squad (56, to Saka’s 61). Neither of those two did well at putting the ball on the mark – Auba’s shot-on-target % was 33.9 percent, while Saka’s was just 34.4%. Both of those are below average for wingers, and 33.9% is in just the 25th percentile among forwards, below Olivier Giroud, Aaron Connolly and Bertrand Traore.

Auba’s poor finishing is especially notable because it followed a season where 44.4 percent of his shots were on target, a 76th percentile rank among forwards. FBRef has six domestic seasons’ worth of shooting data for Aubameyang. He has put 44 percent of his shots on target in that time.

How does that affect the team? Well, 31 percent shots on target this year, following 35.9 percent the year before. That would have ranked ninth in this year’s Premier League, still not great, but much better. 34.1 percent of Arsenal’s shots were on target in 2018-2019, and 37.8 percent were on target in 2017-2018. Simply put, Arsenal missed Auba this year.

To quote the old adage, there’s nothing wrong with Arsenal that changing everything wouldn’t fix. But as detailed above, some good places to start are aggressiveness, both in tackling and pressing, and in passing the ball progressively. The club would do well to develop or acquire more players who can progress the ball past defenders themselves. And they desperately need Pierre-Emerick Aubameyang to bounce back as a shooter. Accomplish all that, and who knows? Maybe Europe is back in the conversation again.

You can find Adam on Twitter @AdamVoge.

P.S. Before you get going on your merry way. I am stepping into the light on Twitch at 6pm GMT later on. Johnny and I will be recording a live podcast. Jump on to make me feel better about the 6 people who will watch, we’ll be answering questions if you have any. Also, tell all your friends about the innovation. The twitch destination is BELOW

TWITCH: 6PM BRITISH UK TIME FAM