I’ve seen a lot of discussion around Austin FC that focuses on differences between 2022 and 2023. The roster is not significantly different from last year though the differences have increased throughout the year.
When I think of soccer, I think of it in terms of stats and tactics, maybe to my own detriment. As such, I wondered what demonstrable, measurable ways the team has changed from the previous season to the current one. A lot the national discussion surrounding Austin in 2022 was xG over performance, something I was well aware of, being the kind of person that keeps FotMob open while I’m watching most matches.
The tactics have been relatively similar to my eye, with even the occasional difference in shape still including the same principles of play including maintaining possession, keeping a high defensive line and playing a lot of passes sideways, all in the name of Disorganizing The Opponent With The Ball. I’ll leave further analysis of tactical differences to someone with a better eye than mine.
I like stats and the resulting analytics because it’s all rooted in fact; these are demonstrable measures that cannot be argued over, unlike body language, motivation, locker room cultures, etc. The conclusions may vary, but different conclusions from the same data set can be valid if they rely on the data to make their conclusions. I’m reminded of a moment in high school English where a friend asked our teach if he could write a paper about some silly topic from Tom Sawyer. My teacher replied “you can write whatever you want as long as it’s supported by the text.”
So I ran some numbers and wrote a bit.
FBRef has 191 distinct measures in team statistics. (Please forgive any counting errors there; several measures appear in multiple tables) and I compared Austin’s 2022 to Austin’s 2023 in every one of them.
HUGE Caveat: I have never taken a statistics class. I am entirely self-taught from playing fantasy sports and reading FiveThrityEight (RIP, sort of).
Brief methodology: I compared each statistical measure on a per-90 basis since the 2023 season is ongoing. I then analyzed for percentage change from year to the next, whether positive OR negative. Each change was categorized in five buckets: change of 10% or less, 10-20%, 20-30%, 30-40% and 40% or more. The total average change was 19.46%, so I considered anything above 20% change to be significant change. Of those 191 measures, 54 (28.7%) of them had this significant change from 2022 to 2023. I separated those out for further discussion and analysis. Small caveat: I do not have the tiem or bandwidth to see how consistent soem of these measures would be year over year, so I am assuming that 10% change or less is probably not significant, especially in a league like MLS with its roster rules and consistent parity.
FBRef has ten different statistical tables on the league stats page. I’ll use those tables to organize the rest of my discussion, which will rely heavily on referencing the table itself, available here: Austin FC Stats Year over Year (yellow highlighting shows the categories where I have identified some kind of meaning)
Standard Stats
Obviously, Austin has scored fewer goals and had fewer assists from 2022 to 2023. We’ve also had fewer penalties awarded. Biggest news here is that red cards have decreased 41.7 percent over 2022 and 2021. Great job boys!
Goalkeeping
We know Stuver is good. We think Las might be good some day. The people who want Las now are crazy. Biggest changes here are also obvious, with the wins and losses and clean sheets all worse than last year. One interesting item is penalty kicks against have gone down significantly, which makes it better that Stuver hasn’t saved one this season.
Advanced Goalkeeping
Stuver really shines in Post Shot Expected Goal Plus/Minus, which is Expected Goals on Target minus Goals Allowed. A higher number indicates great goalkeeping skill (or luck, I suppose). This has actually been Stuver’s worst individual season by this measure (he could catch up to last year’s number, as it is cumulative), but it has been the best Austin team season of the three in existence. (Which says soemthing unfortunate about Andrew Tarbell). He is 11th in the league in this measure, but Austin’s GKs as a whole (just Stuver) are 5th in the league, which shows the value of his consistency (and the value of him not getting sold to Chelsea).
Shooting
Here it comes…..the xG over performance! Fun fact no one is talking about: Austin is still over performing xG in 2023. The difference: this year’s over performance is about 3 total goals. Last year it was about 14. This means that the total drop in over performance was 75 percent! That means that all things being equal, the 2023 team is 75 percent worse at being lucky (or finishing well, depending on how you feel about this philosophical debate) than 2022.
There’s a big difference on the defensive side too. Last year, they over performed on goals against by about 4 goals (gave up 4 less than they “should” have). But this year it’s under performance: giving up 46 goals on about 40 xG.
This is pretty wild considering that Stuver has OVER performed by about 2 goals. Which means that Austin has given up about EIGHT more goals than would normally be expected, all other things being equal. I interpret that as some combo of the defensive players being below league average in talent and teams scoring weird/crazy goals against Austin. Checks out!
This category is basically the season in a nutshell, in my opinion.
Passing
There’s been a pretty significant drop-off (25%) in passes into the penalty area from 2022 AND 2021 (shocking, to be honest). I suspect it is related to something below and definitely passes the eye test for me. The reduction in xA is pretty big as well, but that's died directly to xG, more or less. The drop in xA being bigger than xG might mean Austin is taking more unassisted shot, but I'm not totally sure that's the case.
Pass Types
The people love through balls. Josh Wolff does not (OR Austin plays a number 10 that is mostly a second striker). Austin has had a 40% decrease (averaging less than ONE per match) is next to last in this category after being 17th last year (not great but much higher) but also next to last in 2021. Fun fact the only team behind Austin in 2023 is St. Louis CITY. (Funner fact: STL CITY are over performing they expected goal differential by TWENTY-FOUR (24) GOALS.
Switches have also decreased by about 25%, an interesting outcome from the Berhalter coaching tree, especially after watching the Oman match last week. (Speaking of the Berhalter tree, teams coached by Luchi Gonzalez and Nico Estevez have a lot more through balls over the last couple of years.)
Now, the one you’ve been waiting for: crosses have increased by 15%. Austin is now 5th in crosses after being 16th last year. The decreases in passes into the penalty area and through balls paired with the increase in crosses is the first significant difference found my this analysis. It also passes the eye test, which we love to see when we crunch the numbers. (Crosses were outside of my initial parameters for change but this is a noticeable one for anyone watching the matches, so I wanted to include it here.
Austin has also had a decent increase in blocked passes. I don’t know what that means but it’s infuriating to watch. It could also be an indicator of below league average talent, but I’m not certain.
Goal and Shot Creation
We know Austin is generally worse at this in 2023 than in 2023, but there are a couple areas of interest. There have been a lot more shots coming as a result of dead ball passes (free kicks, goal kicks, and….throw-in) and take ons (I was shocked). However, the goals scored from these shots has increased (slightly) from the dead ball actions, but not from the take ons. More on take-ons later. The other thing of note is that shots and goals that result from defensive mistakes have both decreased significantly.
Defensive Actions
If it feels like the pressing is aimless, it might be! Tackles in the attacking third have dropped off while tackles in the middle and defensive thirds have increased in a significant way. You may have heard that our strikers are old and slow and injured. Clearances have also increased, while (shockingly!) errors that lead to an opponent’s shot have decreased by by 28 percent. (I might just be uninformed but I can’t seem to find Passes Per Defensive Action [PPDA] for MLS anywhere online. If you know where to find it please let me know.)
Possession
Take-ons! Who knew Austin was so aggressive in one on one situations? It probably has a lot to d with the majority of our midfield minutes coming from two guys who prefer to dribble in Pereira and Wolff combined with a team that starts most offensive potions with the keeper or center backs. BUT that doesn’t explain the increased number of takes leading to shots. AND the other part of this is that the success rate has remained more or less the same. The take-on success chart is so funny I felt like I had to include it.
As far as actual possession, it’s not included in the significant changes chart, put its down from an average of 55% in 2022 to 51% in 2023, a drop from 5th to 9th in the league over all.
Misc States
Fewer red cards! And they’ve both been for second yellows, which is also down from 2022! Hooray! Also, more aerial duels have been won, but more have ben lost as well. I don’t know that means, aside that Austin’s players are more aggressive on headers this year, or that we are playing more teams that like to play long passes.
Conclusions
The biggest differences in these measures don’t seem to tell us much about how the team is intentionally playing, aside from hitting more crosses and attempting more dribble take-ons. The season really comes down to two different things, as far as these measures can tell us: unlucky/bad defensive play mixed with unlucky (bad) finishing. I’m more willing to call the defensive play bad than unlucky, though Austin has certainly given up some wild goals. I’m more willing to call the finishing unlucky than bad because kicking a moving soccer ball where you want/need to with world-class athletes bearing down on you is hard.
This is obviously just one perspective and part of the entire puzzle that is 2023 Austin FC, but I think this data and analysis support the idea that the team is not functionally different in the actions that take place on the field.
Editor’s Note: I am the editor, so please excuse any typos or inconsistencies in style. I want over performance and under performance to be single words but I got tired of fighting autocorrect.
Comments
Post a Comment