When a soccer player and a ball love each other very much: How Austin FC goals are made

Over the course of a season, xG gets more meaningful, but to truly understand it, you have to see how a team selects its shots, how many shots it gets, and then what it does once it gets those shots off.

Thanks to a new project from Sebastian Bush, my favorite MLS and NWSL data visualizer working today (and the force behind the MLS Analytics Bluesky account), we can now get a better sense of how Austin FC is getting its 2025 goals compared to its 2024 goals under a new head coach and a new system — though as we're reminded after each game in this young season, it's still early.

Let's start with, per American Soccer Analysis, reviewing what's a great chance or good chance vs. an average or poor one.

If you're getting off a shot in that Great xG range, you should be converting a little more than every other one, whereas if you're in that Good xG range, you should be getting a little less than one in four to go in. Only 19% of MLS shots have an xG of 0.15 or above, but those shots account for 58% of all goals.

Just over half the shots taken are in the poor zone — often calculated as such because of distance, but sometimes for other reasons.

Per StatsBomb: Each xG model has its own characteristics, but these are the main factors that have traditionally been fed into the large majority of Expected Goals models: distance to goal, angle to goal, body part with which the shot was taken, and type of assist or previous action (throughball, cross, set-piece, dribble, etc…). Based on historical information of shots with similar characteristics, the xG model then attributes a value between 0 and 1 to each shot that expresses the probability of it producing a goal.

Sometimes those poor shots go in — 13% of all MLS goals started as shots with an xG of 0.01 to 0.05. But only 2% of those shots go in ... and as we'll see from Austin FC's 2024 season, sometimes it's even less than that for a team in a given season.

(Hint: You probably remember both of 2024's poor xG goals pretty well.)

Analyzing 2025 (so far)

Let's look at what Austin FC's done so far in 2025.

With only five goals through six matches, we're dealing with a small sample size, and the lone poor xG shot resulting in a goal so far this season was Guilherme Biro's goal against LAFC. Four of the five goals, by contrast, had an xG of at least 0.15, and three of them had an xG of at least 0.33 to put them in the Great xG. Here's what the goals look like.

Biro's goal is a tiny red dot all-but-obscured by the yellow circle marking Myrto Uzuni's goal against St. Louis on Sunday. The closest-to-goal circle in Osman Bukari's face-goal against Sporting Kansas City (roughly 0.4 xG), the one at the corner of the six-yard box is Brandon Vázquez's goal (also roughly 0.4 xG), and the one farthest back is Jon Gallagher's goal against SDFC (close to 0.43 xG).

(We should also note here that American Soccer Analysis uses its own xG system with some modifications on what Opta puts out. That's explained here, and that's what Bush uses. As I was telling a friend who checked in with me backchannel after this published, "xG is in the eye of the beholder.")

Here's what all of Verde's shots for 2025 look like so far.

This includes one odd-looking Good xG chance – the one from Austin's own 18-yard line. That's the last-minute Jáder Obrian shot that went wide after CJ dos Santos came up on the deep-into-stoppage-time corner kick to try to salvage a draw.

Nothing else is particularly outlandish on the chart, though Verde are woefully underperforming their xG so far, and they're clearly unproductive with shots outside the box — as well with the average xG shots they're securing inside the box. They're actually only overperforming expectations on poor shots, but that's something that is likely to come back to the average as they take more poor xG shots over the course of the season — unless one of those 1-in-50 chances do go in.

Here's something instructional: I've grouped the two losses together into one graphic, which shows shots and goals accumulated up to that point in the season, when they were 1-2-0 and a bit of worry was setting in.

There's one great chance in there — Uzuni's flubbed and rued attempt against the Rapids, and an overall xG average that's less than what they've accomplished all season. Twenty-seven of the 33 shots are average or poor, making up 1.56 of the nearly 3 xG they amassed in those two losses.

Now, measure it against the SKC win.

Without Bukari's goal, it's 11 shots getting about 0.94 xG total, but there's a higher proportion of Good/Great xG shots.

Compare that with the San Diego match.

The only shot from outside the box was Obrian taking a chance on an empty net. Weirdly, that's the only "Good xG" chance, but there are four Great xG shots in there, and Verde went two-for-four there, good enough to secure the win. As a result, Verde got nearly 28% of their xG in just this one match.

It's not exactly radical analysis to say that it's better to get inside the box before taking a shot, but a typical distribution over the course of a season will have a fair number of shots ripped from outside the 18. Here's 2024.

Those two Poor xG chances that Verde converted? The Biro wonderstrike in the thrilling 4-3 home win vs. SKC, and the Dani Pereira against-the-run-of-play goal at Portland late in the season that kept Verde's fingernails clinching to the lip of the playoff cliff.

And here are the Great xG chances from that snakebitten season.

That's only nine goals from 21 Great xG chances, less than the expected average. Converting just three more of those in the right spots might have been the difference between making and missing the playoffs.

So, how are these goals made?

Looking at xG's only one piece of the puzzle. It's still early in 2025, but here's a chart for what's leading to the goals so far in 2025.

So, "set piece pass" is doing some work for Bukari's goal, as the deflection off Julio Cascante from Owen Wolff's corner kick is being counted as a pass here, but so far, Austin's only registered goals in three of the five categories.

Contrast that with what happened in 2024.

Set pieces, taken together, accounted for just 18% of Verde's non-penalty goals. Individual play accounted for more, which makes sense given that Sebastián Driussi was on this team.

This is how all MLS teams made goals in 2024 — Verde was a bit more dependent on progressive passes and individual play, and short of the 26% of goals scored from set pieces.

Looking at 2022 and 2023, it's evident that Verde held to a similar game model with similar goal distribution — albeit with more success on set pieces in 2022 — even if they had drastically different results.

This is something I'll be monitoring throughout the season, now that the tool's live and the offense is churning into being fully operational. I expect that over time the numbers will shift into more varied distributions — but will Verde continue to be as set piece dependent as they were in their first matches? And when will we see some individual play lead to goals?

Verde All Day is a reader-supported online publication covering Austin FC. Additional support is provided by Austin Telco Federal Credit Union. You can comment here if you’re a subscriber, or reach out via Bluesky.