Backpass: Three Numbers
Some early advanced metrics give us a useful snapshot of what the Rapids are doing right, and what we might want to worry about.
Roughly one-quarter of the way through the season, and on an international break when there isn’t the constant impetus to ‘review the last game, preview the next game’ we’ve started to reach the point where numbers can be useful in discerning patterns about MLS and the Colorado Rapids. Those numbers can tell us as supporters whether we need to recalibrate expectations. They might also tell the front office what this team is going to need going forward in order to be successful.
I’m going to be leaning heavily on an advanced metric crafted a few years ago by American Soccer Analysis, for whom I have written for the past two years, called G+, or ‘Goals Added.’ Goals Added “measures a player’s total on-ball contribution in attack and defense. It does this by calculating how much each touch changes their team’s chances of scoring and conceding across two possessions.”1
For reference, you might want to check out my 2021 Colorado Rapids Season Preview as a useful comparison of ‘where we thought we’d be’ to ‘where we actually are’.
One: Team G+ says we ought to be 11th. Yeah, that tracks.
Here’s the top twelve teams in MLS on G+ to this point:
Colorados 11th-place standing here, with a +1.33 G+ Differential, is roughly consistent with how they have actually performed. Here’s the top of the league table to this point for comparison:
This kind of evaluation is really important, in my humble opinion. That’s because a G+ that doesn’t correlate at all to a team’s won-lost record implies they’ve been lucky… or unlucky. SKC has the highest G+ Differential in the league; they also sit second in the Western Conference on points. Match! Houston have a G+ Diff. of -3.34, 23rd out of 27 teams; and Columbus has a G+ Diff. of -5.96, second-to-last in the league, but their performances to date have them 11th and 10th on points, respectively. No match! If I could make goofy flash animations or memes, it would be of a sea captain passing rocky shoals with a giant sign saying ‘Arrrgh! Regression ahead!’
The Rapids are 6th on points and 11th on G+ differential, and that’s pretty consistent. They finished 10th in the league table last year and 18th in the league on G+ Diff. with a -2.74; last years correlation may have been a little screwy because the normal sample size of 34 games was smaller due to the shortened seasons, and the Rapids missed more games due to Covid than any other team in the league. Statistically, they over-performed expectation in the matches they played.
They’re doing well. The underlying numbers say that makes sense. As long as they keep doing what they’ve been doing, we can start to feel confident about this teams chances for a post-season.
Two: Our MVP is Danny Wilson. No, seriously.
…
The team’s overall leader on G+ is Danny Wilson. Here’s my reaction:
He’s so good, he’s in the top 10 MLS players - at *any* position - for G+ Overall.
With exception of getting beat in a pair of one-on-one situations that resulted in supporters gasping earlier in the season, Danny Wilson has been solid defensively. His Interrupting G+ of +0.54 is the highest on the team. In addition, he’s an extremely valuable Center Back pivot. He serves as both the starting point in the attack, and the first option deep-lying player when opposing teams put on a high press. We can see that in his Receiving and Dribbling numbers (+0.13; -0.05) and in his Passing number (+0.23). If you watch the Rapids develop out of the back, things often start with Danny Wilson - starting on the wings is less preferable, and Austin Trusty is the third best option, and ‘kick it long’ is only utilized if all other options are eliminated, or the team is sitting on a comfortable lead and does not need to take on unnecessary risk in possession.
Danny Wilson’s resurgence is definitely a happy surprise. He was pretty lousy in 2018 and 2019, but in 2020 he was better. In December I wrote: “Strangely, Robin Fraser coaxed an improved performance out of Wilson last year. Wilson was 17th amongst all MLS center backs in Goals Added, with a + 0.71 g+, mostly on the back of exceptionally good interrupting and passing numbers.”
Wilson’s emergence as the team’s best CB is pretty remarkable - it’s like thumbing through your childhood Magic The Gathering card collection during the pandemic only to discover a card you forgot you had is worth three months mortgage. He’s found money. My take on ‘why is Wilson suddenly good?’ is threefold, all of which are interrelated. First, he was bad under Anthony Hudson. Hudson played three different systems - a 5-3-2, a 5-4-1, and a 4-4-2 diamond, all of which resulted in defensive weaknesses at the back *inherently* as a part of the system. He sent forward wingbacks who couldn’t get back, leaving Center Backs isolated. He sent midfielders up who couldn’t get back to defend. That demands a lot of your central defenders, and to be honest, I’m not sure anyone would look good in that system. Second, confidence. Hudson played him at defensive midfielder as well as CB, and he struggled at times in both. And then he got benched. I think it likely that as the coach lost confidence in him, he lost confidence in himself. Third, Wilson’s simply in his prime. He’s a 29 year-old Center Back, and he’s been in MLS for three years. He knows the opposition, he knows the refs, he’s physically able to do the job.
Matt Pollard has said he’d like to explore this further in interview. I welcome that exploration. I also welcome the new, awesome Danny Wilson.
Three: Some highly valued guys are off the mark
If one of your players is going to be disappointing, you ideally want them to not be the most expensive guy on the roster. Taking a look at the worst guys in MLSin terms of Overall G+, that’s not the case.
Gyasi Zardes has scored double-digit goals for Columbus for the last three years. Right now, he has 1 goal in 2021, and every one of his component G+ metrics is in the negatives. Cincinnati’s expensive DP acquisition Brenner has the second-worst G+ in MLS. Former Golden Boot Winner Josef Martinez is also on this list; if Atlanta want to be competitive this year, that has to change.
For the Rapids, Kellyn Acosta’s numbers are off, once again. Acosta was underwhelming in his first two years for Colorado before rebounding in 2020. That -0.43 Passing G+ has me worried. If your main central connector is a below average passer - and he goes flying across the globe on international duty a bunch - you now have an underperforming guy who is also overused. It makes me nervous. As I said earlier this year, he’s the key ingredient. Hopefully, this is a classic ‘small sample size’ issue and it corrects quickly. Hopefully.
…
My disclaimer with Acosta is the disclaimer for everything - the small sample size here for all players means G+ needs to be taken with a grain of salt - a big one. But we have learned some things, and it hopefully informs what we looks for over the months to come.
That’s from the guru of G+, John Muller (@johnspacemuller). To read more, click here.