A lot has been made in the last few days about what might be wrong with Arsenal and what might need changing. The Gunners have ceded the top spot in the Premier League table to Liverpool on the back of a middling run of results. Arsenal are 1-1-2 over their last four league matches with only three goals scored, despite lighting up the shot counter.
You might not like what I’ve got to say, but I’m here to be the voice of reason, to counterbalance all the Chicken Littles out there. Arsenal are fine.
Looking at the big picture, Arsenal aren’t all that far from where they were last season when you look at corresponding fixtures. They’re four points worse off. The defense is markedly better, although less so in terms of actual goals allowed. The attack is slightly worse, and significantly worse in terms of actual goals scored.
Scoring is the big difference, so let’s drill down.
Arsenal Compared to Corresponding Fixtures Last Season— Scott Willis (@scottjwillis) December 29, 2023
Points: -4 (40 vs 44)
xG: -3.3 (34.7 vs 38)
G: -12 (36 vs 48)
xGA: -4.6 (15.2 vs 19.8)
GA: -2 (18 vs 20)
xG Dif: +1.3 (+19.5 vs +18.2)
G Dif: -10 (+18 vs +28) pic.twitter.com/4EBs3TSx25
Arsenal went into the match against West Ham slightly ahead of their non-penalty expected goals for the season. They came out of the match with their actual versus expected goals roughly even. The average Big 6 side runs about 10-15% over their expected goals for a season — individual seasons shown in the table below.
For a good team like Arsenal, scoring even with your non-penalty xG is running cold. In fact, if Arsenal continued to score at roughly the same pace, it would be close to the worst season for a Big 6 club in that seven year dataset.
Were Arsenal to improve towards the average for a good team in the second half of the season, they’ll be right in the thick of things, in contention for the title. If they trend towards last season’s level of finishing, you’d have to think about making them favorites.
If we want to look specifically at the West Ham game, unfortunately Arsenal were living in the 1% world, as I like to call it. Technically they lose that match more than 1% of the time. What I mean is that there are things that are likely to happen and things that are unlikely to happen. That match falls squarely in the highly unlikely category.
Per The Athletic, Arsenal-West Ham was just the second time since 2016-17 a Premier League team took 30 shots in a match and lost. Ironically, the other instance was that David De Gea masterclass against Arsenal at the Emirates in December 2017. A few more tidbits:
- Arsenal had 77 touches in the opposition box against West Ham without scoring, the most of any team since Opta started collecting that kind of data in 2008-09.
- From the xG generated in the match, Arsenal had a higher probability of scoring 5 goals than scoring 0.
- Arsenal failed to convert all five of their big chances in the match.
You could go on and on with unfortunate superlatives achieved by Arsenal against West Ham. The Gunners dominated the match and fell victim to flukes and bad luck. Sometimes it be like that.
We don’t need advanced stats to tell us that Gabriel Martinelli is in a scoring slump. The eye-test works just fine there. The stats tell us that Gabriel Jesus, Bukayo Saka, and Martin Ødegaard are all converting chances below their career rate. Even as a known xG under-performer, Gabriel Jesus is lagging well behind his finishing rate.
In spite of that and while playing the 1st, 3rd, 6th, and 8th teams in the table in December, Arsenal’s attack is trending in the right direction (in every category that isn’t actual goals scored). In the past month, the Gunners have managed 4 of their 6 best non-penalty expected goals performances of the season. They are up to 7th in the Premier League in overall npxG and 2nd in the PL in open play chances created from passes, both of which are high-water marks for the campaign thus far.
Arsenal have had the best defense in the Premier League all season. They’ve had between the 4th and 6th best attack in the PL all season, too. Over the last 10 matches, the attack has gotten slightly better overall, improving to the 3rd best in the Premier League in terms of Scott’s weighted xG and G model.
Premier League Weighted xG (70%) and Goals (30%) through round 19 and over the last 10 matches played. pic.twitter.com/Kh8Dgo56so— Scott Willis (@scottjwillis) December 29, 2023
Arsenal have also been one of the best teams in the Premier League this season in Expected Threat, a measure of possession value, specifically what is the chance of scoring or conceding a goal based on a team possessing the ball in a particular area of the pitch. For those of you with a subscription to The Athletic, they wrote an xT Primer two years ago.
Data at https://t.co/dUh5MC3R20 updated for all leagues! Graph shows Expected Threat created and conceded for Premier League teams. I.e. how often each team enters and prevent entering into the most threatening areas pic.twitter.com/KT68jG6P9m— markstats (@markrstats) December 29, 2023
The Gunners rarely allow their opponents to have the ball in dangerous areas and they do a pretty darn good job of getting the ball into dangerous areas themselves. Since November 1st, Arsenal have improved their creative threat, moving above 1.6 to post the third-best mark in the Premier League, behind only Manchester City and Liverpool (and a small margin behind at that). That improvement from Mikel Arteta’s side has not come at a cost defensively, either. Arsenal’s allowed threat has held steady.
That’s a (very) long way of saying that this season Arsenal are playing the kind of football that puts you in a position to win Premier League title if things go your way. And they’ve been improving on that football over the last 4-8 weeks of play!
Just to underscore that improvement and that Arsenal are in the midst of a patch of bad luck and depressed finishing:
There isn't one, perfectly linear story to Arsenal's attacking issues this year. There are two: before the update (fewer chances/shots, more pens, better luck) and after (way better everything — 12.32 more open-play SCA + 6 more shots per game — but worse luck and misses). pic.twitter.com/4J61wlgPBc— Billy Carpenter (@billycarpy) December 30, 2023
If you think back over the campaign so far, it’s hard to come up with matches where Arsenal were convincingly outplayed. Slightly outplayed? Maybe. Played evenly, absolutely. But for the most part, the Gunners have been the better of the two teams on the field.
The answer to the $64,000 question “what should Arsenal do” seems to be an emphatic “nothing” or at least nothing significant. They should stay the course, continue to play (more or less) how they’re playing, and bank on the finishing improving towards (or beyond) where you’d expect it to be based on past history, player skill, and the volume of chances created.
That’s more or less what Mikel Arteta said — keep doing what we’ve been doing, find a way to improve the finishing.
It’s not panic. It’s about trying to do more and better. If the team plays like this it’s going to win a lot of games.
Without spark, you don’t generate what the team generated yesterday. It’s impossible. But it’s the final thing, the final touch, the final action that puts the ball in the net or not. That’s what we need.
When specifically asked whether he thinks Arsenal can win the title playing as they are now, he did acknowledge that Arsenal won’t manage it “if we don’t improve in the boxes” in the second half of the season.
He’s not wrong. Odds are that if Arsenal were to keep converting chances at their current rate, they won’t win the league. But the odds are also that Arsenal aren’t going to continue converting chances at their current rate. It’s likely to improve.
So we’re back to stay the course and ride out the negative variance. That’s not a particularly satisfying answer. When we perceive a problem, we want to fix it / for somebody to do something to fix it. Take action!
That drive to act is especially potent as the January transfer window opens. There is a compelling urge to do something. New personnel must be the way to improve.
For a second, let’s put aside the pesky details like which players might be available, at what price, can Arsenal afford them (both under FFP regulations and from a cash flow perspective), and how does it fit into the bigger picture of building a squad capable of competing over multiple seasons. Although I’ll point out that nearly all of those factors cut against the Gunners making an earth-shattering move in the next few weeks.
Change for the sake of change is not the way to go. Especially big change. The phrases “don’t just do something, stand there” and “if it ain’t broke, don’t fix it” come to mind. As I’ve laid out, Arsenal are playing solid, title-winning football (if they get even average finishing). Making dramatic changes to good process based on a small sample size of poor results is folly.
Arsenal are dangerously close to “more harm than good” territory right now. They don’t need drastic changes. Fortunately, I think Mikel Arteta recognizes that. It’s not an accident that he talked about not panicking. He’s cautioning against overreacting to bad luck and negative variance. He knows that playing like they are playing right now is the way to win a bunch of games, even with sub-par finishing. He knows that if the finishing improves, and it should improve, they’ll be in really good shape.
It’s always been about maintaining a consistent level of performance this season, and Arsenal have done well in that regard. Last season, they started the campaign on a 90-point pace and ended it on a 65-point pace. Two years ago, they collapsed down the stretch and fell out of the Champions League places despite sitting in pole position.
This season, with the Premier League looking as if it will be won by 84-86 (ish) points, Arsenal are on an 80-point pace at the halfway post. If they keep playing as they have been with just the slightest bit of improvement (i.e. the finishing rising to expected levels) they’ll be right there. Arsenal are good enough to win the league. The question is whether they can play at or close enough to that level over a 38-game season rather than just for pieces of it. And we really won’t know the answer until we get into March and April.
That’s not to say Arsenal shouldn’t tweak things in an attempt spark that small, needed improvement. A team should always be doing that, even if they’re racking up wins. A new tactical wrinkle, like we’ve seen with Martin Ødegaard dropping deeper and having more freedom of late (that has him playing fantastically well, I might add), seems the most likely thing to jumpstart the Gunners. Gabriel Martinelli seems most likely to benefit from some kind of change, perhaps that’s what he needs to snap out of his poor form.
But really what Arsenal needs is patience. The attack has gotten noticeably better since the struggles to start the season without sacrificing the defensive side of the ball. The finishing should improve, hopefully sooner rather than later. And even if that doesn’t happen overnight, the side is playing well enough to win in spite of the finishing struggles.
A healthy dose of shooting drills in training might be in order, too.