Chance Creation Report Cards
- ATL create low volumes of great shots. Concede high volumes of good shots.
- RSL create high volumes of meh shots. Concede average volumes of good shots.
Atlanta remains towards the bottom of the table in shots – 19th with 10 per game. RSL is an average (or slightly above avg) shot volume team, shooting 12.4 times per game. Atlanta is shooting 4.5 times on target per game (good for 7th in the league) compared to RSL’s 3.7 per game (17th in MLS).
Atlanta is the most accurate team in the league with 44% of shots testing the keeper. RSL is second to last with 30%. Atlanta is also first in the league at converting shots on target into goals (48%) compared to RSL’s 31% (15th in MLS).
Analytics people will tell you that the shots on target conversion percentage will most certainly regress to the mean, and the shot accuracy figure should regress a bit as well, although it can persist. I’m torn on how to interpret Atlanta’s chances question. It is almost certain that Atlanta will not finish the season converting 48% of their shots on target (it’s rare for the best teams to convert higher than 40%). That said, the chances Atlanta are creating seem to be of the “fewer but higher quality” type (top of the league at 0.21 xG per shot) compared to RSL’s 0.06 xG per shot (18th in MLS). This matches the eye test, as plenty of Atlanta’s shots are on counters/transitions, or better yet, on the end of defense-splitting throughballs. This doesn’t exactly show up in the broad shot location numbers as Whoscored.com has 46% of Atlanta’s shots coming from outside the box though the volume of shots from outside the box is not high (4.7 – slightly below league average). My gut says the team is very focused on turning good chances into great ones when they’re in and around the box, which puts downward pressure on the volume of in-the-box chances (I think of failed dribbles or missed final balls at the end of a quick counter – which is a choice a player makes over taking a lower quality shot).
Defensively, Atlanta are allowing the most shots per game in the league (15 per game) and at a relatively high quality rate of 0.129 xG per shot allowed (4th highest in MLS). That this has not manifested itself in giving up loads of goals is due in part to Alec Kann (who, in a league of his own, continues to beat his expected goals against metrics by a notable margin) but also due to some amount of luck (Gio skying open chances). RSL are allowing just under 12 shots per game at similar chance quality levels as Atlanta.
It’s an odd combination for one of the most intense pressing teams in the league to be creating low volumes of shots and conceding high volumes. If I think of the prototypical pressing sides in Europe, one of the goals is to suppress opposition shot volumes (often at the risk of giving up higher quality chances when the press is broken). Liverpool, Man City, and Tottenham concede some of the lowest volumes of shots per game in England. Same goes for RB Leipzig and Dortmund in Germany. The fact that Atlanta is giving up very high volumes of somewhat high quality chances could be due to a number of things: too few games played so far for these things to normalize, unique situations (red cards and/or an unbalanced schedule), purposefully open and exciting play (Eales advertised this before the season started), or perhaps it’s something unique about MLS that reacts with Martino’s tactics in an odd way. Watching some more ATLUTD games should help us get to the bottom of this. Stay tuned.
One thing I will continue to keep an eye on is throughballs. Of the throughballs that are actually counted by Opta, Atlanta are completing twice as many key pass throughballs (1.2 per game) as the next best team (which worryingly is RSL). These are some of the highest quality chances a team can create. I continue to question why many of Atlanta’s assists are not categorized as such when they seem to be splitting the opposition defense and putting a player free in on goal. This may be a pervasive throughout the sport, and I just notice it because I watch all the Atlanta games closely.
RSL Scouting Report
RSL under Petke play a 4-3-3 with Brooks Lennon on the right and Plata on the left (when healthy) flanking Movsisyan in the forward line and Beckerman anchoring a midfield three along with Rusnak and Mulholland. Petke’s comments when he came in referenced wanting to play a higher line than RSL were previously and wanting to play out of the back more.
Here are RSL’s even game state defensive actions for the last two games. 54 minutes in the snow against Vancouver before taking the lead (left). 29 minutes against Colorado before conceding (right).
These seems adventurous enough to suggest to me that Atlanta will be pressed.
Nick Rimando is one of the better MLS goalkeepers and can play the ball out of the back through his centre backs, which he’s done fairly often in the three most recent games.
If he does so, Atlanta will likely be waiting. For what it’s worth Rimando appears to be conceding goals at the rate one would expect based on the shots on target he’s facing.
We’ll see if Petke is committed to having RSL continue to pass out of the back against a high-pressing Atlanta or if they will play it more conservatively and hit long. Without watching much of RSL this year, I don’t see anything glaring in the pass maps to suggest a weakness in either centre back to exploit in terms of Atlanta’s pressing. Wingert has spent most of his MLS career at fullback so perhaps he’s a bit tidier.
RSL seem like a very dangerous attacking team. While Plata has only played a couple 90s worth of match time, he’s averaging 1.3 expected goals + expected assists per game (per 96 minutes) according to the folks at AmericanSoccerAnalysis.com. Movsisyan is one of the leading MLS attackers to date, averaging 0.75 expected goals + assists per game, which puts him just ahead of Fernando Adi at the moment. Brooks Lennon is putting up Almiron-levels of attacking threat (0.34 xG+A per 96). The back line will have its hands full and Plata can show up anywhere similar to Piatti or Lodeiro.
What should Atlanta do?
This all makes for an intriguing match. Should Atlanta continue to play on the front foot, the way they’ve done every game on the road so far? Should they push the fullbacks high and play out of the back? Will RSL press Atlanta the way Seattle and Toronto did in their respective homes.
Well, I do not think RSL will sit back so there should be plenty of counter opportunities. This points me towards Villalba at forward instead of Jones. I would flank him with Asad and Almiron on either side and allow those guys to to all interchange to their hearts’ contents. Given Plata can pop up anywhere and Brooks Lennon is killing fools I would play Larentowicz, Carmona, and Gressel in midfield with plenty of positional restraint. Perhaps someone man-marking Plata and the other two shielding and running. I would focus on pressing and transition more than possession and building from the back, and I would aim to have the front 3 plus Gressel bother Beckerman often.
Looking at the extended highlights of the RSL COL game, it looks like the Rapids’ chances came on direct balls out to the wings or up and over the full backs. This does seem like something that could be replicated by Villalba if he’s playing wide or central.
I think a draw is a good result. It’s very hard to predict with RSL having a new manager and Atlanta travelling long distances (again) after running furiously while down a man (again) for 45 minutes last weekend. Happy to have Asad back and for the rescinded Pirez red. Let’s just see if Almiron does something special this weekend.
Prediction: 2-2 (submitted with anxiety)