Welcome to The Graphical, wherein I mine the Opta data for insight into Nashville SC’s most recent result. In this edition, a disappointing draw against the Pittsburgh Riverhounds.
87 minutes of dominance
The narrative of the game has been about how good Nashville was for the first 87 minutes (and how poor for the final nine). At the risk of beating a dead horse… that’s the narrative because it’s true.
Take a look at the Riverhounds’ heat map over the course of the game. There’s practically nothing (and certainly nothing meaningful) until the very end:
You can argue whether that’s the Riverhounds finally turning it on, or Nashville sitting back – or a combination of both – but certainly it’s a sudden onslaught to end the game. Even if the two goals were reasonably fluky (and both were), Pittsburgh was rewarded for pouring it on to close the contest.
Aside from last season’s playoff draw against FC Cincinnati, Nashville SC hasn’t regularly used a three-man (or five-man, depending on how you want to classify wingbacks) formation since their home draw against Pittsburgh 13 months ago. Even the 2018 victory in Highmark saw them go with a standard 4-4-1-1, which was the nearly-exclusive formation in even game states from there on out.
That was… not the case Saturday. Take a look at the average positions. Your eyes didn’t deceive you, this was a pure 3-5-2:
Bradley Bourgeois was pushed up just a bit higher than Liam Doyle, but from Kimura, across the back, and up to Washington, that’s unmistakeable.
More interesting to me? A switch-up in the midfield. Previously when using this formation (and also with even backline formations), Gary Smith has opted to go with a double pivot of two defensive midfielders and a No. 10. Instead, he went with a single defensive midfielder in Bolu Akinyode, and a pair of box-to-box players in Matt LaGrassa and Michael Reed.
He even mentioned it after the game when I asked him:
“I thought we needed a little bit more steel and discipline at the back there,” he said. “A three-man midfield of really genuine midfield players, rather than creators. Like I said, for long portions of the game, it turned out the right way.”
That allows NSC to continue having essentially the same balance of numbers in the defense and attack (three centerbacks and one defensive midfielder versus two and two), as you’d see in a 4-4-2. The width of the formation is shifted to be more focused on the back-end, given the lack of wide midfielders, but conceptually you make up for that by freeing the wingbacks to get up the pitch.
And get up the pitch they did
The space available to cross when facing a three-man backline (it’s worth noting that Pittsburgh had shifted from a 3-4-3 formation to a more traditional 4-1-4-1 over the course of the year, but went back to the 3-4-3 in this one) has been a regular topic around here.
It came to fruition with both teams’ fullbacks (or wingbacks) able to get upfield into crossing positions and send them in.
That’s two crosses for Kosuke Kimura and three (one obscured by Matt LaGrassa’s No. 20 down there) by Taylor Washington. The only open play cross that was completed is a mis-coded short corner by Justin Davis – you can tell it’s a corner because it’s on the right side of the pitch, where he wouldn’t otherwise play – while eight were left wanting.
On the other end, former NSC player Ryan James (7) completed 1/3 crosses, his counterpart on the right Jordan Dover (5) went 0/1, and more traditional midfielder/winger Kevin Kerr hit one from either side.
Crossing is traditionally a poor way to generate offense (the conversion rate is extremely low), but if the opponent forces you into using the cross – as both attempted to do with their defensive-minded formations – you’d certainly prefer to complete them than not.
Formation, substitution, and what changed
Gary Smith swapped Taylor Washington out in favor of Justin Davis in the 75th minute. With Davis in the game, the formation was much more defensive:
Look at how much closer the No. 2 is to his own goal (on the right) than No. 23. That’s not a matter of Washington or Davis making a conscious decision, for what it’s worth, but rather two outcomes chosen by Gary Smith at the same time, to accomplish the same goal: He wanted to be more defensive, so he put in a better defensive option and shifted to a more conservative defensive posture with that sub.
Your opinion on the difference in defensive acumen between Davis and Washington may vary (I think the pure quality difference negligible, but stylistically, Davis is more a risk-taker so it’s an odd sub to make to close out a 2-0 lead). Still, you can at least tell what the purpose there was.
Nashville dominated this game, especially when you take into account that two of the three Pittsburgh shots came within a couple minutes of each other (and that only one of them went in):
Of course, dominating the game doesn’t matter if it doesn’t lead to your scoring more goals than the other guy. Nashville put the ball in the net twice, Pittsburgh once. The problem is that Nashville also put a ball into its own net (not really Kosuke Kimura’s fault, by the way – James probably should have just fired into the open net in front of his face anyway).
I am beginning to wonder if there’s a problem with NSC taking shots that get blocked by defenders. Four of nine shots didn’t get a chance to test the keeper, and only one of those came from a position where you shrug and move along.
Last season, I was critical that the team wasn’t taking speculative shots to loosen up the defense a bit. Now, I’m concerned that there’s been a bit of an overcorrection, and reining in some of those shots outside the box that get blocked within a couple yards of being taken would be helpful. A happy medium needs to be found.