Analytics are all the rage these days and while I do think they are useful, I tend to use them after I’ve developed a theory. I look to see if there are numbers that back up my thinking or if there are numbers that tear my thinking apart.
In short, I use the eye test first then I check the stats for validation.
I have this theory I’ve been working on about what kind of player Jesus Ferreira is and where FCD should play him in their shape. So I dug into some of his numbers from 2019 (2020 is too small a sample, you see) to see what they would tell me about him.
Then I wrote up this whole section for my theory post. I found, though, that the analytics bogged the thing down too much so I took them out and put them in this separate story.
Goals
At the very top surface level, Ferreira lead FCD in scoring with 8 goals and 6 assists in 2019.
The first 2/3s of the season (20 games) – playing mostly as a 9 – Ferreira had 5 of his 8 goals and 4 of his 6 assists. In the final third’ish part of the season (13 games) – playing mostly as an 8/10 – Ferreira had 3 goals and 2 assists.
So the surface level stuff makes logical sense. He was pretty much just as effective as a striker or mid. It’s simply a matter of more games meaning more goals and assists.
When I broke the game down by literal position (based on the MLS post-game reports and FCD lineups), rather than splitting the season into chunks it looked like this…
Pos. | Games | Goals | Assists |
9 | 17 (1 sub) | 5 | 4 |
Wing | 4 (3 sub) | 0 | 0 |
10 | 4 | 2 | 1 |
Free-8 | 7 | 1 | 1 |
For clarity, the difference between 10 and free-8 is 10 plays in front of a double pivot and the free-8 is paired with a linking-8 in front of a single pivot.
Seems like he’s about a goal every 2.4 games no matter how you slice it.
So let’s dig deeper.
Expected Goals
For the season, the ASN xG numbers in total say he scored half a goal less than he should have in 2019.
Over the first 20 games – again, mostly as a 9 – Ferreira’s expected goals (xG) was 6.26 (thanks American Soccer Analysis) and he scored 6. Over the final 13 games – again, mostly as an 8/10 – Ferreira’s xG was 2.29 and he scored 2. So really, he’s pretty much on point with the number of games so it’s not particularly telling.
The split still makes logical sense.
So I took another tack and looked at regular open play versus fast breaks. Ferreira’s fast-break xG was 0.51 with 1 goal scored and his regular open play xG was 7.82 with 7 goals scored.
Again, that’s pretty much right on point.
I was starting to get a little annoyed as the stats weren’t giving me anything to sink my teeth into. The xG was letting me down.
Then, while digging into two of ASN’s newer stats, xPass and Goals Added (g+), I found what I was looking for.
Goals Added
First came the Goals Added which breaks down Ferreira play by six categories.
Action | Goals Added | FCD Rank |
Receiving a Pass | -0.88 | Worst |
Passing | -0.54 | 5th Worst |
Dribble | 0.48 | 5th Best |
Shooting | 0.2 | 3rd Best |
Interrupting | 0.48 | 5th Best |
Fouling | -0.24 | 4th Worst |
So, what conclusions can we draw from that?
Receiving a pass – Bad. That means playing as a high 9, back to goal, and having balls played into him isn’t good. Probably also checking back or playing in midfield and trying to combine too for that matter. Frankly, I feel like this stat isn’t as bad as it seems, he’s mostly playing in the danger areas of the field under high defensive pressure.
Passing – Bad. He’s not a creator, he’s not a 10. We’ll come back to this.
Dribble – Good. Ok, he needs to go at defenders. I frequently talked about Ferreira splitting the gaps in the defense on the dribble last year. So he needs to be a touch deeper and not high.
Shooting – Good. Pull the trigger, we’ve talked about Jesus being a good finisher. Getting him shots is good.
Interrupting (aka defending) – Good. He’s a quality presser. You want him defending up the field to create goals. He is a quintessential modern pressing player.
Fouling – Bad. Right up to the point he fouls the defender and lets them off the hook from the press. So don’t foul.
When I took it one step further and broke Goals Added down by the same two-thirds season split I’ve been using, it got even more telling. Receiving (+.78 g+), dribbling (+.56 g+), and shooting (+.69 g+) were all three better from a deeper position rather than as a 9 and the defense/pressing was better from a higher position (+.78 g+).
So we want him higher to press, but deeper to get the ball, recover it, go at people, and shoot.
xPass
And the last stat I looked into is another of the new ASN stats, expected passes. In this category, I chose to directly compare him to Pomykal.
Attacking Third | Ferreira | Pomykal |
Passing % | 76.9% | 74.9% |
Expected Passing % | 75.9% | 71.2% |
Passes over Expected | 2.84 | 11.03 |
Middle Third | Ferreira | Pomykal |
Passing % | 83.3% | 89.1% |
Expected Passing % | 88.0% | 71.2% |
Passes over Expected | -25.01 | 19.12 |
Takeaways?
In the attacking third, both Ferreira and Pomykal are better creators than expected. But Pomykal is more better.
In the middle third things geta a little ugly for Jesus while Pomykal excels. Paxton passed way above expectations, doing better in tougher spots (lower xpass) with an almost 18% passing rate over expected, and adding almost 20 more passes then the numbers say he should.
Jesus, on the other hand, had the opposite effect. He gets into good spots (high xpass) but his passing in the midfield third is under what the metrics say they should be by about 5% and he’s 25 passes under expectations.
So in both phases, Pomykal excels at passing, keeping Luchi-Ball working and the team flowing. Pomykal is the choice as a playmaker.
Jesus creates in the final third but nearly as well in the middle.
Analytics Conclusion
What does it all mean?
It means getting Jesus up into the front line, into the attacking third, and into the box is important.
So how does FCD do that? Here’s my idea.
Seeing Paxton’s passing numbers in the middle 3rd is impressive. He’s wildly outperforming the expected.