top of page

What went right against Canada East?

Everything. From top to bottom, this team destroyed Montreal. The post game scores are enough to show you how dominant this performance was.

Courtesy of FotMob

But numbers don't do this justice. Even these scores seem too low. I want to dive into Expected goals today, because this number is not, and SHOULD NOT be the conversation topic it is.

Apple TV is using expected goals as a staple of their producttion, and If you'll remember, there's been a near constant echo from the MLS media scrutinizing Austin for their performance against expected goals. Specifically, that Austin outscoring according to expectation.

What is an expected goal?

By googling, I came across this definition:

An xG model uses historical information from thousands of shots with similar characteristics to estimate the likelihood of a goal on a scale between 0 and 1.

By definition, this is a pretty narrow stat, specifically looking at a shot that was TAKEN, and not necessarily an opportunity. So, when a shot is taken, it should register somewhere between 0 and 1 based on HISTORY.

But this stat seems unhelpful for judging a team's performance. Let's take Rigoni's xG for Example. All chances considered, he finished with a .3 xG. Now, maybe for Rigoni, .3 was correct. He seems to have no skill in finishing right now, which is a concern. On overall statistical basis, how did his two wide open clear chances only achieve .3 goals expected? Frankly, an opportunity like both of those should have netted AT LEAST one goal. Let's see both to remember:

In essence, xG is saying that each of these shots from this position on the field only have a 15% chance of going in. For comparison, Zarde's two contested headers tallied ~.16 xG. A player in rhythm and on his foot with just the keeper in front of him has got to be a higher xG than two contested headers.

Beyond that, this stat completely ignores potential chances. There were two shots taken in this game that scored goals (Pereira's shot that called off because of a foul and Zardes' offside goal) and then later, another goal Maxi put off the keeper that almost scored and also was barely offside. None of these counted as xG, and yet, when you take the game in as a whole, those are incredibly telling moments. 3 times, we nearly scored, completed ignored by the analytics.

Austin's Expected Goals in this match tallied to only 1.9. On eye test, that game felt like 3-0 or 3-1 Austin.

Expected Goals (xG) is a faulty stat. Although it's a nice model to have to compare games, it's a flawed to use it to try and express how well a team played or didn't. Opportunities don't always end in shots, and those shot opportunities can be misjudged.

For NBA fans, this stat acts very much like plus/minus (+/-). This stat tells you how the team does while a particular player is on the court. So, if the team goes on a run and extends the lead while a player is on the court, you see him garner a higher plus/minus. This stat ignores who you're on the court with, and who you're on the court against, and therefore will over inflate a players value depending on when they're playing in the game (against backups, or playing with a great player). It's something that can quickly be refuted by the eye test, and and xG is the same.

A quick eye test tells me Austin should have been up 1-0 after the two Rigoni chances. A wide open DP winger with a backup keeper in front of him. That's at least worth half an expected goal. This is why I don't rate stats as highly as watching the game overall.

Speaking of rating the eye test:

Owen Wolff

Owen rated as one of our worst players on Saturday, and although I can't deny that he made mistakes, what I'm seeing are bursts of ability that are revealing a superstar. One thing I've noticed about him is that he's incredibly accurate with his passes. It's rare that he plays a ball off target, but he's prone to ignoring the defender. So, although the ball is on point, they're getting intercepted. Avoiding the defense will come with experience, but his underlying skills are there, and they were on full display in the last 15 minutes of that game Saturday. Come, join me on this journey to Mt. Wolff.

His tear lasts for 8 minutes, but leads to numerous chances for us, and this stretch triggers the last 15 minutes in which Austin dominates.

This stretch here is great. Collects the ball from Leo and plays a long progressive ball forward and on target. It circles back to him, where he decides it's time to place a ball no Gallagher's foot on the sidelines. That pass alone should be at least .2 xG.

Wolff is really good at drawing fouls. He drew 6 in this one alone (3 against ST. LOU). He baits the defender, then immediately fires it to Pereira to get the break started. He really has an old school game about him, which is why the team (and I) feel comfortable with him out there.

Here he is dribbling through defenders and drawing a foul in a great spot. It's moments like these three that tell you this kid doesn't care the situation, and that he knows he can be the catalyst. Turning it up 75 minutes into a game is gamer shit, and he's going to be that for us. He also threw himself in front of two shots at the end of this game as well.

Staring at the post-game stats don't show you this. Soccer specifically is not driven by stats, mostly because stats are hard to address. There's few points. Rating pass, shot, and chance creation is subjective at worst, and analytically bare bones at best. But watching a player corral control like this in a must win situation is evidence enough.

A Few More Notes
  • Everyone's talking about Ring at CB, and although I thought he was phenomenal back there, I worry that we weren't tested too hard by this Montreal side. His go to was to boot it out of bounds almost every touch. Although effective, It's leaving turkey on the bone when he could have just grabbed possession. I think he might slot in there long term as his career carries on, and I liked his forward ball skills and quickness in this role.

  • Rigoni's and Zardes are getting into spots, and it's only a matter of time before they get going. It wouldn't surprise me if it happened at the same time in a blow out win where we just unload on a squad. I called for Urriti to start on Saturday, and I'll continue to call for that on Tuesday against the Haitians.

  • The fullback play in this game was off the charts. I think we've gotten an internal upgrade there in the off-season with Kolmanic coming in cooking and I think it's increased competition. I hope this continues, because I have been proponent of upgrading at this position, even over CB.

Recent Posts

See All


bottom of page