Synergy
09.03.2026 20:07
👍 6
🔁 0
💬 0
📌 0
Synergy
Screenshot of dashboard: "How your team scores: MLS xg data" various stat breakdowns and pitch graphic with shots shown as dots below
It took me a sec, but the xG dashboard is back for 2026!
For those who weren't here last season, the dashboard uses @jamonm.bsky.social & @carloncarpenter.bsky.social's WGCF principles to break down MLS shots.
Built using @americansocceranalysis.com data.
public.tableau.com/app/profile/...
ha mls over time
Looks like about 2015/16 for the peak and downhill since.
home advantage in MLS over time
Looks like down a bit in recent seasons.
So that insane home advantage didn't survive week 3 of the 2026 MLS season (46th percentile now).
Is Celantano moving in the summer? Seems to be the only way this makes sense.
And this assumes shootouts did not happen.
Adding in all MLS games, 2026 has the highest home advantage to start a season - just edging out 1997 and 2000.
I pulled some data via @americansocceranalysis.com's {itscalledsoccer} and while home advantage has never started out so high in an MLS season since 2013, it is not quite the highest ever (only 98.41 percentile).
I suspect this is an aberration and it will bounce around just like it always has.
Atlanta United could get worse
A graph showing what percent of 2025 minutes are returning in 2026 for each team in the NWSL. Seattle Reign has the most, with 90.5%, and North Carolina Courage has the least, with only 58.8% of 2025 minutes returning. The league average is around 75%.
With only two weeks left before the 2026 NWSL season kicks off, check out which teams will look most similar to and which teams will look most different from their 2025 counterparts!
Returning minutes graph for @americansocceranalysis.com.
If you want to catch up, you can find all the previews here
www.americansocceranalysis.com/home/categor...
It's actually the half season loan of Ayo Akinola for a half season spot.
While there could be a few more changes, this incorporates all the recent moves (Salloi, RBW, Dorsey, Carmona, Elie).
This season has the highest roster turnover since I started tracking this prior to the 2019 season, but there are no statistically significant differences in any of the seasons.
Truck with a "Fuck Andy Beshear" decal on the back window
I saw this truck in Nashville in June 2020. Based on that, I assumed that Beshear must have been doing a good job with Covid.
It was from a campaign video. The image file I have was created on July 1 2035 so probably sometime around then.
Zohran: Just take a look at this graph
Still haven't found a use for this, but it is a great template.
Dema Kovalenko!
Since I wrote this @michaelarace1.bsky.social interviewed with Henrik Rydström where he answered:
- #Crew96 will probably play 3 CBs because that's what the team knows- but may go to back 4
- Zawadzki is CB for now
- Habroune can play as AM, 8, or 6
creators.spotify.com/pod/profile/...
@akeaswaran.me made this neat graph of a quick strength of schedule model. Really interesting to see how relative difficulty between the front and back parts of the season may drive narratives. Neville might find his seat pretty hot, while Varas is my pick for COTY winner w/ a likely hot start.
I’ve written some articles that took months of research and writing on nitty gritty soccer analytics topics. But nothing has ever come close to being as popular as my transformer-based QSMNT tweet generator shitpost from back before ChatGPT was a thing. www.americansocceranalysis.com/home/2022/6/...
A big ass truck
A big ass truck that hates Andy Beshear
The British mind cannot comprehend.
You'd think that hating soccer would disqualify someone from running Major League Soccer. Given their actions it appears that not only it doesn't, but it is a qualification for getting the job.
I do hate that Seattle invented the MLS fandom tSNE neighbor cluster.
That said, I also love the neighbors of the American Soccer Community.
The thing with tSNE dimensionality reduction is that it preserves local structure in the data (dots in a cluster are close to each other in high D), but global structure are not (the larger cluster positions don’t really mean anything).
More modern dimensionality reduction techniques try to fix.
Fancy Stats commissioned a bunch of articles about the 2018 World Cup from American Soccer Analysis writers.
I will not apologize for any contribution I have made to the shift from open play to set piece scoring.
www.washingtonpost.com/news/fancy-s...
Pod drop: podcasts.apple.com/us/podcast/a...
We talked to @armankafai.bsky.social about the ASI Summit and some of the top tier research going on there. Then we got weird on MLS roster rules.
Available wherever you get pods.
Engel, Kamara, Kubo, Orellano, and Yedlin are the big ones. I did miss that Powell is coming back so it will increase a bit once I fix that.
