Week 1 (ATL v NYRB) Analysis

When feasible, I will try to break down the matches into game states to see if any helpful patterns emerge. For this particular game, because both teams came out to press early and often, I spend some time focusing on turnovers below (unsuccessful touches + times dispossessed as presented at whoscored.com). Stats below come from whoscored.com and MLSSoccer.com (both from Opta I believe).

Game State: 0-0 (0-26′)

Both teams attempted to press early. Watching in person at the time I felt like Red Bulls enjoyed the opening 20 or so minutes more, but watching the replay, it seems more even or slightly in ATL’s favor. During this opening level game state, Red Bulls turned the ball over 12 times (3 in their own half – mostly on their left flank) to ATL’s 8 turnovers (with 2 in their own half – 1 centrally located). Red Bulls successfully tackled 3 times in ATL’s half to ATL’s one time in RB’s half. Both sides were disruptive with their pressing. New York however had the best chance at the 10 minute mark after Kann’s poor attempt at jump-starting a counter with short distribution. Zizzo collected and dribble into the box on Atlanta’s left flank to find Aaron Long in space near the penalty spot, only for him to miss the goal left with the shot. From my vantage point, it felt like a great chance and I was terrified, although I could not identify the player at the time (a day later, knowing he’s a CB, it feels less significant). Below are the shots. Left is ATL, right NYRB from MLSSoccer.com (Opta):

Game state changed when Asad scored at the back post, unmarked, on the end of a good cross from Tyrone Mears in the 26th minute after a cross from Almiron on the left deflected out to Carmona who found the assisting right back easily. If we reach back into preseason, an overwhelming majority of goals scored by Atlanta United are coming from crosses. We will need to monitor this as the regular season progresses, as this isn’t always the most sustainable chance creation machine – especially for a team of short fast guys…. That said, this narrow Red Bulls team did seem ripe for wide play, so perhaps this was very purposeful and specific to this game. Also, to my eye there were plenty of attempted through balls by Atlanta players even if Opta did not register them. Early turnovers:

NYRB turned the ball over sightly more, but Atlanta gave the ball up in dangerous spaces as well in the opening 26 minutes (0-0). Image from whoscored.com

Game State: 1-0 (26-76′)

Atlanta enjoyed the next 50 minutes, spanning halftime. During this period the trailing Red Bulls turned the ball over in their half 6 times (3 of them centrally) to ATL’s 3 (all on the left wing). Atlanta countered often at pace. Total turnovers during this period were ATL 16 to NYRB 19, most of them innocuous, the result of attacking moves in the opponent’s half. Yamil Asad had a striking 9 of these for ATL though, including all 3 of those in the ATL half — to my eye he and Garza were unfairly tasked with taking on the entire NYRB’s right side to play the ball out of the back on the left flank.

During this large stretch of the game, ATL turned over NYRB frequently in their own half. ATL’s own half turnovers came from attempts to play the ball out of the back on the left (through Asad/Garza). Screenshot above from whoscored.

Anyhow, these ATL turnovers were somehow mostly not punished. During this game state, Atlanta had a handful of good chances to extend the lead, most noticeably Miguel Almiron’s surging break off the interception from Pirez ending in a chip attempt which was impressively saved, but also a delicious low cross from Garza which found an outstretched Martinez at the goal mouth. One wonders about the parallel universe in which either of those chances go. Atlanta led key passes during this period 7-3 (though I counted a handful of ATL countering through ball attempts narrowly missing their targets), and led shots 11-7 but they were lower quality shots to RB’s (edged 3-5 in shots on target). See below for shots during this period:

For what it’s worth, Matt Doyle at MLSSoccer.com will tell you a couple things about when this match turned. He points to the 70th minute substitution and smart formation switch by Marsch to a 4-1-4-1 which Atlanta failed to adapt to. With all due respect, I believe he is wrong. I did not see this at all. Watching the replay, the Red Bulls continue to play with 2 clear forwards after this point, even more “2-forwardish” than what they’d been playing with for most of the match, with one of those forwards being a more free Klestjan. If anything I’d say this became a wider 4-4-2. But most importantly, he seems to forget that Atlanta still enjoyed the match for another 5-6 minutes including a nice attacking move, even if it ended with a blocked shot by Martinez and a distance reflex shot from Garza. When I watch the goal that changed the game state (Royer’s 76th minute, well headed runner off of a corner), the leadup to the corner which ended in a bungled save from Kann and a weird pinball moment took several deflections and odd busted heading plays to even turn into an attacking move for the Red Bulls. Fair play for sure, but I didn’t see an Atlanta confused by any sort of brilliant formation shift from NYRB. Seemed more like the kind of noise that’s bound to happen a couple times a match, particularly late in games. @MLSAnalyst, if i’m missing something here, let me know. Watching the replay it seemsed somewhat fortunate for Red Bulls.

Game State: 1-1 until end (76 – 94′) *(includes 2-1, and red card)

From 76 minutes on, it’s easy to remember nothing going right for Atlanta, but this would ignore one last decent attacking move where Villalba megged the defender and played Mears through ending with his cross put out for a corner. After this however, it went bad, starting with what looked like a hamstring injury to starting CB Pirez. Walkes on for Pirez, Larentowicz on for Gressel. The buildup to New York’s winning goal was nice, with an overlap from Lawrence not picked up by the Villalba/Mears combo and a dangerous ball into the box. From my vantage point at the stadium, it looked like a well converted big chance for BWP, but it appears to have gone off Walkes last. One other thing I didn’t realize until watching the replay: BWP is offside and the goal should not have counted. It’s close and happened quickly, so I do not fault the line judge, but it’s offside.


Shortly thereafter, Carmona was red carded for a stamp. Aside from the own goal, there were no shots on target, and only one blocked shot from 76 minutes on – NYRB killed the game effectively.

How I think about this result: For a fan, this was a bit of a gut punch, mainly considering the order of events, and the final portion of the match featuring an injury to a starter, an own goal which probably should’ve been disallowed and a red card. The further context of being anxious for the club to start its history off on the right foot in order to maintain the mostly good fan momentum to date, does not help. But slightly more objectively, was the result fair? Maybe. In terms of refereeing, the own goal looks offside, but there was also a penalty shout for NY for a Carmona handball (neither of which I noticed live in the stadium). There was a NY goal kick that I was adamant should’ve been a corner, but that’s fine. Taking an objective step further, given the chances created in the match, was it a “fair result” the way a manager might talk about that sort of thing after the match? I believe so. Looking at a few Expected Goals outputs floating around the internet, a draw would’ve probably been the most fair but a 1-2 loss was not a freak occurrence based off the run of play and chances created. @11tegen11 had the xG battle at 1.21 to 0.71 in favor of Atlanta with the home team winning 48% and drawing 30% of the time, based on chance quality alone. AmerianSoccerAnalysis.com had a slightly different tally at 1.02 – 1.36 in favor of NYRB, which might be adding a couple chances in the Royer/Kann pinball sequence. Ben Baer at mlsoccer.com had it at 1.13 – 0.68 in favor of Atlanta. And lastly five thirty eight had Atlanta head in its shot-based expected goals model 1.2 – 0.7 but behind 0.5 – 0.9 in its non-shot based model. But the consensus of all the models, by my estimation is that a draw was most likely. Given “own-goals” are difficult to capture, and given the own-goal in this game was the result of a menacing chance from New York’s left back (but maybe offside?),  I think either a draw or a win for Red Bulls would make sense, but 3 points for Atlanta would not have been shocking either. On balance playing relatively level to a very good (if over-travelled) New York Red Bulls team is a positive development as a first entry into Atlanta United’s ledgers — just pretend you didn’t watch it in real time.

For further reading, there’s a great bit of tactical analysis from Mark Thompson (@etnar_uk) which you should read here, mainly cuz he’s really smart, and only partially because he stayed up late in the UK to watch it, and so you owe him.

Addressing other questions posed in last week’s preview post – lest I forget them

  • On formation:
    • As most expected, Martino went with the speedier Martinez at striker instead of Jones. Hard to say that was wrong, but Martinez ended up doing real work battling for headers most of the game. For what it’s worth I loved his intensity and work rate (6 shots, 7 fouls ;). Timing of runs/through balls to him from midfield will improve.
  • On buildup out of the back vs long goal kicks:
    • Ok, this one was fascinating. For the opening game state (0-0), Kann completed 100% of his passes as ATL focused on playing out of the back, primarily through Pirez. NYRB weren’t overly troubled during this period for the most part. But then, following the goal from Asad, and as Atlanta moved into its more comfortable stretch of the game, Kann played it long, almost exclusively:
    • (left: 0-0 for first 26 minutes, right: 1-0 until 76th minute)

      Something to think about and track going forward. I’m not against playing it long if we’ve got a quick and aggressive outfield ready to press the living hell out of the opposition once the ball is knocked down.

    • Also on centre back distribution (left is ATL, right is NYRB):

      So, I really hope Pirez (#5 on the left) isn’t out for any length of time. He seems the tidier of our two centre backs (90% passing vs Parkhurst’s 76%). Also, look at the distribution from the redbulls centre backs (sub 60%) – that’s…direct looking isn’t it? Is that how they always look? Or is that what happens when ATL attackers are down your throat with 55K screaming?

  • On the narrowness of the pitch at Bobby Dodd:
    • Happy to say, it didn’t seem embarrassingly narrow to me from my vantage point, and it looked great on the TV replay in my opinion. I’m sure it favored the more narrow 4-2-2-2 Red Bulls pressing outfit and took away some of the sting of Villalba’s pace out wide, but all in all it seemed fine.
  • On set pieces:
    • Well, they got us on a set piece as I’d expected, although I didn’t see anything too crafty in the design. They also beat man2man marking against one of our tallest guys so that was unexpected. There were a few other setpieces that looked scripted to me, one free kick and at least one corner kick, but they did not result in danger.

One thought on “Week 1 (ATL v NYRB) Analysis”

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