Welcome to The Graphical, wherein I mine some of Opta’s publicly-available data to glean some insights about Nashville SC’s most recent result. Today, a frustrating road loss to Portland Timbers.
The passes are key
We already know Nashville SC didn’t score. Everyone has expressed some question or another about the finishing (about which more in a moment). But where did Nashville actually generate its offense from against the Timbers? Here’s the map of the team’s key passes (passes that lead to shots – would-be assists) in the game:
They break down like so:
- Two from corner kicks (one Daniel Lovitz, one Hany Mukhtar)
- Two from within the penalty area to shooters also inside the box (one from Walker Zimmerman, one from Mukhtar)
- One from Zone 14 into the penalty area (Dax McCarty)
- Two from within Zone 14 to shooters in Zone 14 (Mukhtar and McCarty)
- Crosses from the left side (one from Jimmy Medranda into the box, one from Randall Leal to Zone 14)
- One from Mukhtar near midfield, in which Randall Leal dribbled into Zone 14 before taking the shot.
There’s not a whole lot to dislike there. Nashville is generating offense from dangerous areas, and getting shooters in reasonably dangerous positions. Well, somewhat reasonably: 1.11 xG – per Opta’s model – on 14 shots is not great! Average xG per shot in MLS for the entire dataset 2011-19 is 0.097 (per American Soccer Analysis’s slightly-different model), so with MLS-average shots, you’d see around 1.351 expected goals.
So, the shots that Nashville SC is getting are a little below-average (four assisted by crosses, some from headers, both of which are lower-value categories), but at this point, the meager scoring is more attributable to two causes – a new team getting used to each other, and some simple bad luck – than underlying stats indicating that this will be a poor offense. Some teams consistently underperform xG over the course of a season, but more often than not, this would regress to the mean.
It’s still possible that an elite finisher is needed in the transfer market (I’d like to get Daniel Ríos a longer run-out before demanding it). There’s not enough evidence through two games to assert that.
Pack it in
Here are Portland’s defensive actions (defending the bottom goal) both before and after Diego Valeri gave his team a lead:
They didn’t really change much schematically, so much as they continued doing the same thing: drawing a low line of defensive confrontation just on the far side of midfield, and concentrating on preventing good chances, rather than creating turnovers or playing a positive game.
There’s no shame in that with a lead. It’s certainly still a context that applies to Nashville’s scoring difficulties as outlined above. Portland was a mediocre xG defense but a good strict goals-against defense last year, and it’s easy to see how their scheme is designed to make that happen: pack the box and above all else, protect the keeper.
Final third woes
So we’ve seen two sides of a coin with one noting that Nashville probably could/should have had more effective offense, and another that shows how Portland worked to prevent that, despite giving up a high volume of shots.
Here’s the piece that ties it together: if Nashville wasn’t creating an assist in the final third, the Boys in Gold were hardly connecting their passes at all:
Some of those are speculative lumps in from the wings, and certainly as the full-time whistle approached with NSC sporting a 1-0 deficit in the game, you can see why that became the gameplan (over-reliance on crossing and set-piece play while down a goal is one of the very legit criticisms of Gary Smith’s tactical approach through two USL seasons and one USL game, though).
Hopefully, some of the imprecision we’re seeing here will also be worked out with more time on the field together at game speed. Nashville is having to deal with a roster turnover that only Inter Miami shares. The more quickly NSC can overcome that season-starting disadvantage, the more quickly things will turn to the positive.
Nashville is creating a fair number of scoring chances, but the individuals have to be better to create even more of them, and to earn a shot out of moving the ball into dangerous positions on a more consistent basis.
From an expected goals perspective, holding Atlanta United to 0.24 in Week One is probably more impressive than limiting the Timbers to 0.18 in Week Two. I’d still say we’re getting some indications that the Nashville defense is elite.
Once again, it took something of a moment of individual brilliance (and maaaaaybe some sketchiness on that single play – but it’s a 15-second stretch in a 90-minute game still) to beat Nashville. “It took X to beat Nashville” necessitates “you beat Nashville,” though, and neither team will come away embarrassed about taking all three points.
Portland in particular wasn’t even really trying to advance into the attacking third once the Timbers too the lead.
They got on Nashville’s end of midfield and passed it around the horn, or lumped a forward pass into the attacking zone. Some of that is attributable to Nashville’s defense. Some of it is attributable to the Timbers not even trying to score (which ultimately proved to be the right approach in that situation). Contrast that with Nashville’s final-third passing noted above: while they weren’t successful in completing the last pass frequently enough, they concentrated on getting into deep positions before attempting that final ball.
Toronto FC has averaged nearly 2.00 xG per game through two matchdays. This weekend should tell a much better picture of just how strong Nashville’s defense is on its own merits (for what it’s worth, Atlanta and Portland both racked up more than 1.50 xG in their non-Nashville game).
For now, signs are good. There doesn’t appear (with extremely limited data) to be anything structural in Nashville’s defense causing opponents to outperform their expected goals on a consistent basis.
What did you see in the data? Poke around and feel free to share your observations in the comments or on social media.