Week 2: MIN vs ATL (cold)

I’ve heard plenty of people warning against analyzing the snow game, so I’m going to ignore them.


Visiting ATL found through ball after through ball to Martinez, Almiron, and others and they mostly finished what were generally high quality chances, having bypassed Minnesota’s back line. Minnesota created some decent chances of their own in transition and on set pieces, which were on the whole more chaotic, scrambly but high percentage type chances. Fortunately for ATL, they were not converted. Atlanta pressed early and throughout the match, and did not change tactics all too much except for playing the ball long more often as the game started to settle into its natural state of Atlanta leading by multiple goals comfortably.


I found Atlanta’s shape to be interesting, especially how similar it was to the general structure in last week’s loss to NY. @11tegen11’s pass maps look similar for both games. Check them out below (first the minnesota game, then the NYRB game):

Gressel looks a bit more connected to the whole network, and Villalba and Almiron are slightly deeper, but on the whole that’s similar, especially the heavily linked up play on the left between Asad/Garza/Almiron and the rather empty looking central midfield. You can see in the pass map that Kann was much more likely to play the ball long against Minnesota given the conditions, and perhaps because of the early leads. But having figured last week’s emphasis on width was specifically dialed up against a narrow NYRB, I was surprised to see it show up visually again like this.

But, if the shape was the same, the chance creation was vastly different. Against NYRB the ball went wide (the only place NYRB players were not standing) and then it was crossed in. Against Minnesota, the visitors neglected crosses in favor of more potent, penetrating through balls. This week, if the ball was played out wide, it often came back inside or was played directly forward. Below is a map of crosses, key passes, and assists per MLSSoccer.com. Left is vs NYRB with all the unsuccessful crosses, right is vs Minnesota, with few crosses and plenty of vertically successful attacks. That’s what I like to see.

This is what I find interesting about the snow game though. My very limited tactical wisdom would tell me that crossing would be better in the wild conditions, as you’d want to create as much chaos in the box as possible. And Minnesota did do this to decent effect, just missing the conversion. But to me, Atlanta playing slick through balls through the heart of defense suggests that to them, the conditions did not make this an invalid game at all (and perhaps people should stop discounting it). They played the game how they wanted to, aggressively and with abandon – the fullbacks often leaving the central defense exposed against Minnesota counters. Sure, the pass %s are lower across the board, but the best players on the field shined (87% passing, a brace, and an assist for Almiron, 78% passing & a hat trick for Martinez) If there’s any reason to discount this game, I’m not sure it’s the weather. It might be the quality of the opposition, maybe? But if that’s the case, we should address this:

expected goals
Source MIN ATL
11tegen11 1.64 2.34
American Soccer Analysis 2.04 1.79
fivethirtyeight (shots) 1.6 1.7
fivethirtyeight (non-shots) 0.7 1.1
MLSSoccer 1.7 2.14

Why the hell, do all of the expected goals models have the game close, or even giving the edge to Minnesota. I see some grumbling about the weather again, but I think mainly it’s a combination of Minnesota getting some “extra credit” in some cases for multiple chances (rebounds and the like) in the same possession, and Atlanta’s chances not being rewarded enough in the models for how truly open many of them were.

Two of Martinez’ goals were 1-on-1 with the keeper, and on another, he dribbled around the keeper to shoot at an empty net: https://twitter.com/MLS/status/841116619657093122

Almiron’s first was a decent chance also, even at that angle, as he was in free on goal. I would say all 4 of these goals involved #finishingskill, which is perhaps what the lower values of the models are showing. At the same time, each of those 4 chances were the kind of chance that made me as a fan stand up on my feet before the shot was taken – and to me, that’s the definition of a “finishable” chance even if the actual percentages are lower than I realize while watching. And the Answer’s chance was high % as well.

Chances like these may not rack up the xG, but they seem like genuinely great chances to score for a competent finisher. @ me

Highlight: Almiron’s first goal

I did this briefly on twitter already, but I wanted to highlight how beautiful this goal was, even if the defense fell asleep. This was not a typical press, recover, attack at pace goal (like the 3 very nice Martinez goals). Almiron’s first was generated out of basic buildup/possesion play, and the fluidity and interchange between Garza/Asad/Almiron and Gressel to pick it out was sublime, even in the snow and against the frozen statues of the Loons backline. First, the ball is thrown in on the right side and cycled around to switch play to the left with Garza. He looks up to play the ball centrally to Gressel, but importantly, at this very moment, (#1 below) Asad sees acres of space in the center and decides to move into it. (#2) Asad moves into the space and drags the left back with him. Almiron sees this and runs into the space vacated by Asad’s movement – this happens as the ball is on its way from Garza to Gressel. (#3) Garza makes a wide vertical run, which freezes the fullback, who had just started to realize he’d been had by Asad and was thinking of inching back to his home to prevent something like what’s about to happen. Garza’s run forces him to recognize it, and the space remains for Almiron, and Gressel (who I thought had a phenomenal game) picks up on it immediately and slides the deep through ball in for Almiron to power it into the roof of the net.

I think the expected goals tallies for Minnesota shouldn’t be totally discounted either. Atlanta seems prone to giving up those types of chances, namely getting countered and forcing its centre backs to make difficult plays, and getting set-pieced (due to lack of height). I think that’s something they’re willing to live with if they can continue creating 1-on-1 chances with the opposition keeper.



Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s