It's time to check in on Atlanta United's underlying numbers (Pt. 1: The Basics)

It's the most magical time of the year

This one is free. The next article will not be. The next article is focused on individual players and is coming out later this week, so…Please subscribe.

Atlanta United have played nine matches this season. That’s typically not enough for your record to reflect who you are as a team, but it is about the time of the season that the smartest folks in the room say underlying numbers start to become predictive. Basically, about eight to 10 games in is when you can start looking at things like expected goals with some degree of confidence. 

MLS caveats do apply though. Injuries, playing through international breaks and extreme home-field advantages skew things around here a little more than in other leagues. Those things even out (some) over time, but it’s still best to take numbers for exactly what they are: Data points. Just like what you see with your eyes is a data point. The idea here is to take as many data points as we can and get somewhere close to the truth in a sport that lies constantly. 

Anyway, let’s look at what the numbers are saying about Atlanta United so far. 

The Basic of Basics: Expected goals 

The nerds have a messaging problem. They always have in sports. They didn’t really have a breakthrough until they started calling a stat something metal and turned boring batting average arguments into discussions over WAR. 

The point here is that people new to the statistics side of soccer hear “expected goals” and immediately dismiss it because goals were goals during the war and you got them whether you expected them or not (or something). But it’s the (bad) descriptor that won out and we have to keep using it for clarity’s sake. 

All we talk about when we talk about expected goals is how often a team creates chances and what the quality of those chances is. The team that creates better chances than their opponent is more likely to win. It’s just giving a name to one of the few things you already know to be true about soccer. 

Well, FOLKS…Atlanta has consistently been creating better chances than their opponents. 

Per American Soccer Analysis, Atlanta United is currently sixth in MLS in expected goal differential per game. The only teams out-creating their opponents by more are LAFC, D.C. United, Red Bulls, Philadelphia¹ and Minnesota United. That’s very good company this early in the year. 

Unsurprisingly, the attack is leading the way here. Atlanta’s 1.71 expected goals for per game is the fourth-highest in the league. Only D.C. United, LAFC and the Galaxy are creating better chances each match. Suprisingly, the defense is getting the job done. The Five Stripes’ 1.4 expected goals allowed per game is 17th in the league. They aren’t quite as stingy as their current one-goal-allowed-per-game pace would suggest, but considering the chances they’re creating and the very important fact we’ve seen the team’s two starting center backs together for only three games this season, 1.4 is just fine. 

It’s actually the same level of chances last year’s team allowed. The big difference here—In addition to starting backup center backs for the majority of the year—is that 2024 Atlanta United is creating 0.33 xG worth of chances more per game than 2023 Atlanta United. 1.71 xG per game is the highest mark for an Atlanta United team since 2018. They aren’t near the level of what’s statistically the second-best attack in MLS history², but we are seeing as effective an attack as we’ve seen since the healthy Josef days. And, again, it’s worth considering how injuries have affected both sides of the ball. 

Which real quick, just to lay that out…

Health: Atlanta’s “9+2” status

You can read a full breakdown of “9+2” here, but we’ll try to be short. It started in Seattle with longtime Sounders analytics legend (and current Atlanta United analytics partner) Ravi Ramineni. The idea at its core is that once teams began playing without three experienced starters in their lineup, they began to struggle. If you only lost two, you could see pretty typical results. 

It’s really a framework for finding the best opportunities to integrate young players into lineups, but we like to co-opt from time to time to assess the relative/health and strength of a team. Here’s what Atlanta’s lineups have looked like so far.

Anyone…uh…notice anything about the terrible, no-good, winless April? 

Atlanta United have played four of their nine games with a lineup that could be called 9+2 or better. They won three of those and lost the other on opening day to the reigning champs and the team set to take down CF Monterrey in the Conca Champions semifinal. 

As for the other games, the only one they didn’t play well enough to win came in a 5+6 lineup on the road against a healthy Toronto side. 

Which lines up relatively well with this…

A more advanced metric: Goals Added 

Goals Added (G+) is American Soccer Analysis’ all-encompassing metric that assesses the value of every action on the pitch. While xG is great at telling you about chance creation, G+ is here to do the same while filling in the blanks. At the most basic level, G+ is about what areas of the pitch you work the ball into and which areas you keep your opponent from getting into. 

It stands to reason then that if you’re working the ball into your opponent's 18 more often than your opponent, you’re probably doing something right. Which, hey, by the way…

(thanks to Ben Wright for the graph he doesn’t know I stole yet, go check out Backheeled.com please and thanks) 

For a more detailed breakdown of G+ you can head here, but you can just trust me when I say that G+ is the most predictive MLS metric we’ve got. If your team is near the top of the league in G+, odds are higher than ever that good results are coming your way. 

(thanks to Tiotal Football for the graph he doesn’t know I stole yet, go check out AbsoluteUnit.substack.com please and thanks) 

Well, FOLKS…Atlanta has been consistently working the ball into better positions than their opponents…and the rest of MLS. 

The Five Stripes are first in the league in goals added differential per game. They’re fourth in MLS in attacking G+ and sixth in MLS in defensive G+. 

When comparing this group to past Atlanta United teams, they’re just behind the all-time great 2018 side and well ahead of Atlanta’s third-best side, the 2019 team. 

Right now, at this very moment, there’s an argument that Atlanta United is as effective as we’ve seen them at any point besides 2018. Without two terrible blown leads against Philadelphia and Cincy, we’d be talking about a Supporters’ Shield contender sitting on 1.89 points per game and putting up Tier One underlying numbers. 

Health and numbers 

We’ll dig into some more numbers in the coming days, but for now, my takeaway from the stats and my eye test is that Atlanta United is one of the best teams in the league. They’re putting up these kinds of numbers without key starters and healthy lineups.  It’s honestly impressive. 

My gut feeling is that these numbers will improve as the team gets back to full strength. And it seems like 9+2 lineups will be back soon with Stian Gregersen getting healthy. 

It’s not perfect right now and Atlanta is about to face its toughest stretch of the season, but there’s reason to believe they’re going to handle it just fine. 

1. Philly may have even dropped below Atlanta after last night’s performance against Seattle. Still waiting on those numbers to come through.

2. 2.09 xGF per game lol