Thread by @adamvoge

OK everyone, we need to talk about Pierre-Emerick Aubameyang. I call this one "Why Auba can't be as far off as he was last season..." (1/x)

Stop me if you've heard this one before: Pierre-Emerick Aubameyang had a poor 2020-2021 Premier League season. He scored 10 goals in the league, two from penalties. Unlucky? Ted Lasso might call it "New York Jets."

Perspective: Auba scored 10 for Arsenal in 2018...in 12 gms.

What happened?! For one, his shooting dropped off. Auba hit the target with 33.9 percent of his shots in PL play and was 36.9 pct in all comps. Auba is a career 44.2 percent shooter.

2020-2021 was also the first domestic campaign Auba failed to hit at least one SoT/90, (0.73).

(Perspective: Last season Auba's shooting percentage was in the bottom quarter of Prem strikers. Neal Maupay, Willian Jose and Aleksandar Mitrovic were among his closest comps.)

So Auba missed more often. But he also shot less often.

With 56 shots, Auba averaged 2.19 shots per 90 minutes. He shot 90 times in 19-20 and 89 in 18-19, both 2.5 per 90 or higher.

Look at that cluster of red and yellow right in front of goal. There's 4+ xG right there.

I find that cluster disturbing because they were high odds opportunities. Here's his 19-20 shot map, when he was admittedly finishing at something like a career-best rate. Fewer missed sitters, more goals on lower-odds shots. Plenty of chances missed, too.

Here's his first year at Arsenal. Again, you see misses from in front of the goal, some pretty big red or yellow circles here. Auba isn't automatic. But the sheer volume of chances and shots is noticeably bigger, and some of those tiny green circles went in, credit to him.

Here's another way to look at his chances missed. Forwards miss chances, that's not a secret. But Auba didn't actually miss that many last season. In fact, 7 big chances missed is arguably his Arsenal-career-best (2017-2018 was a half-year at the club; this is the PL's stat).

Let's look at progressive passes received and touches in the penalty area/90 from all years available. Hmm. Last year alone, Auba was receiving 1.5 progressive passes less per 90. Among PL forwards, that's the difference between 50th and 85th percentile. That feels significant.

Another note, though I'm not sure how significant it really is: In 2020-2021, only one player assisted more than one Auba goal: Willian (2). Two PKs. The other seven were assisted by seven different players.

2019-2020:
Pepe (3)
Saka (2)
Niles (2)
five unassisted

Another thing that seems problematic: Auba didn't beat defenders off the dribble as much. Dribbling was never really his golden skill, but his dribbles completed/90 was in the eighth percentile among PL forwards last year, down from 22nd and 40th the prior two years.

Some of these things, particularly the shooting accuracy and reduced number of chances, can be fixed. Arsenal want a creator to help him get more chances, and Auba is statistically likely to shot better. But here's something Arsenal should be thinking about...

Auba's output is on the way down. That decrease from Bundesliga to EPL would be expected, and last year's dropoff is more than it should be. I do suspect this is age-related, and it would be wise to start preparations soon for a future without the Gabonese striker.

Auba feels likely to top that 0.31 non-penalty goals per 90 this season, but can he get back above half a goal per match? Mathematically speaking, even posting a 0.50 npg/90 last season would have netted him five goals, or 15 overall instead of 10. ~9% of the club total.

The shooting is likely to bounce back. But how much of the rest will? That's a huge thing to watch this season. Arsenal certainly need Auba's output, particularly in an attacking four potentially devoid of other skilled scorers. For now? It's good to see him smiling.