Arsenal's finishing slump has hit at the wrong time

There is never a right time to go on a cold streak but having it happen now, with two key cup matches and chasing Liverpool could not have been worse.

Martin Odegaard has his penalty saved against Manchester United

It has been a tough week for Arsenal.

They lost 2-0 to Newcastle in the first leg of the League Cup semi-final, putting themselves in a terrible position for trying to advance.

They lost on penalties after a 1-1 draw to a 10-men Manchester United in the FA Cup.

On top of that they have not exactly pulled up trees in the Premier League winning 1-0 against Ipswitch, 3-1 against Brentford (actually pretty good), and drawing 1-1 with Brighton.

The crazy part is outside of the Brighton match Arsenal have been pretty convincingly the better team in each of those matches

Arsenal (1.8) 1-0 (0.1) Ipswich

Brentford (0.3) 1-3 (1.8) Arsenal

Brighton (1.5) 1-1 (0.8) Arsenal

Arsenal (3.0) 0-2 (1.2) Newcastle

Arsenal (3.4) 1-1 (0.6) Manchester United

That is a combined scoreline of (10.9) 6-5 (3.7).

This doesn’t change the fact that this is just 2 wins, 2 draws, and a loss but it does highlight that sometimes a team needs some bounces to get things going their way, and even then a team might just be having one of those seasons.

I think that this was a bit of what happened Sunday against Manchester United. Arsenal got a few bounces that went their way, Manchester United had a red card for a second yellow (with the number of fouls it did seem like it was coming but you never know with referees) and a penalty that was probably generous.

Those decisions were worth nearly a 2 full goal swings towards Arsenal and it just still didn’t matter in the game and it was one where again the team saw more misfortune still losing Gabriel Jesus for 9-12 months with what looks like an ACL injury.

Going Cold at the Wrong Time

It feels like a bit of deja vu for Arsenal. This time last season, almost to the day I wrote an article that has a similar theme to this one. This was coming off the back of Arsenal going through a tough finishing run as well:

There are no satisfying answers to the Arsenal slump

Not long after this, Arsenal went on a pretty historical heater for finishing and propelled themselves to within a few bounces of winning the title. That is not a guarantee to happen again, and that isn’t how variance works but it is a good signal that this isn’t a situation that lasts forever.

Taking the shots Arsenal have had over this stretch and putting them through the simulator you get the following spread for Arsenal:

This is 2.2 standard deviations away from the average, and it would be expected that the team would have scored 9-13 goals from these chances but have instead scored just 6 in this span.

Looking at the subset of these shots that are just on target (or would have been on target if not blocked) it looks like this:

Both of these match expectations a bit better but still have Arsenal on the left-hand side of the spread of potential outcomes.

This definitely suggests one of the big explanatory factors is that the team has done a poor job of converting the chances that they do create into tough chances for the opposition goalkeeper to keep out.

Is this a Systemic Problem?

Looking at the full season Arsenal have 60 goals scored from 63.8 xG. That is a bit below the total expected and it covers basically the last run of matches as taking the team from above xG for goals scored to slightly below expected.

In statistical terms this is basically right in line with the expected values (less than a standard deviation below) where it would be on average 64 goals scored with a +/- 5 goals spread as the main expectation for the range of outcomes.

The problem recently has been that the team has done poorly with the shots on target being significantly below the xG but that is not the case with the season as a whole.

This looks like more like an outlier statistically, the average here is 66 goals with a +/- 4 goals spread and this ends up over 1.4 standard deviations below expected.

So while the current run does show that Arsenal have not added value with the shot placement, as a whole that has not been a particularly large problem for the team this season. I would take that as a sign that it would be improbable that this is something inherent in the team.

Maybe it is an Arteta Problem?

Every time something like this Manchester United or Newcastle match happens, and you talk about how it is a bit of a variance or heaven forbid say it is “unlucky” people like to bring up that this happens to Arsenal more often than other teams and that would suggest that this is not just unlucky and a problem with Arsenal or the way Mikel Arteta has the team set up.

I have looked at this at this previously and it didn’t seem to hold water but I thought this would be a good chance to look at things even larger.

For this I have pulled data from FBref Stathead for all matches in the top 5 European Leagues since Artata has taken over where a team had an xG difference of +1.0 or greater.

If it is true that there is something specific to Arsenal I would expect that they have more draws and losses in these situations than average or maybe even the other big teams that they are striving to compete with.

Since the start of the 2019-20 season in League matches, Arsenal have had 77 matches where they had an xG difference of +1.0 or greater. This is the 9th most of all teams in this span.

For these matches, Arsenal have won 67, drawn 8, and lost 2. That is 2.71 points per match and an 87% win rate.

The overall average is 2.42 points per match and a 74.8% win rate. This suggests that Arsenal do not have these sort of fluke results happen more often than expected. If we set the cut off at 50 matches that meet the criteria to eliminate “weaker” teams the average is 2.57 points per match and a 81.4% win rate so still much higher than average.

Out of the 29 teams that have had 50 or more matches at +1.0 or greater xG Difference Arsenal rank 3rd for points per match (behind Lazio and Tottenham) and 4th in Win Rate (behind Lazio, Dortmund, and Tottenham).

Arsenal have been better than Real Madrid, Manchester City, Liverpool, Bayern Munich, Barcelona in these situations under Arteta.

Looking at the different subsets of xG difference and the results hold true with Arsenal coming out much better than average and as good or better than their peers.

At +1.5, Arsenal are 2.88 ppm vs 2.64 ppm and 93.9% vs 83.9%

At +2.0, Arsenal are 2.92 ppm vs 2.76 ppm and 96.2% vs 89.2%

Final Thoughts

I am not going to lie to you and say things are great right now. They aren’t but that for me is tied much more to the injury situation at the club and the ramifications of the results and where it has put the team for the things they want.

Goals are more important than expected goals.

That is not in doubt. A team plays for scoring (or not allowing) goals and while a good way of doing that is by creating good scoring chances (what xG measures) it is not 100 percent aligned on everything.

I don’t think the attack has been perfect, but the level of what is wrong is probably out of proportion to the actual problem for the team.

Just like when a team runs hot, you don’t have to give back points or wins that the team may not have earned but you should worry about how sustainable it is. Running cold is basically the flip side, you get some comfort that this probably won’t go on forever but it doesn’t make the points dropped get added back.

If the team can address the pure depth issues that they have from the long term injuries I would feel a lot better but overall it is probably a spot where the mood has sunk below the actual performances. Hopefully the performances can continue and the results will come back towards matching them.