When Wondo came in in 2014 he won every ball and created havoc on Belgium. In extra time, he created almost as much xA (0.75) as the US (0.81) had xG tonight. There was *no one* on that bench who could be be brought in to do that tonight. But we focus on his miss. And so we miss what matters.

Honestly would rather lose the way we did in 2014 to Belgium.

Trying to find a positive not named Malik.

Berhalter. Ball striker.

Win a second ball, I'm begging.

The first 25 minutes made it look like the US hasn’t progressed since 2014.

1.93 to 1.93 xG is insane. The margins on this game were tiiiiight.

Apparently Google considers my Wondo article “famous”. Vimeo locked my account over hosting the videos. It’s nearly impossible (or was) to find the full minute or so before the miss or to watch the incredible work rate Wondo showed in his few regulation minutes. The pass to Dempsey was gorgeous.

This game has turned into a MLS midweek After Dark special.

I always believed. bsky.app/profile/stev...

Thank you, Cape Verde

Unbelievable. Cabo Verde played Argentina till the end in an amazing game, and this wasn’t a fluke. This wasn’t haram ball. A team tried to play soccer toe to toe with the World Champs and did it. This tournament is anyone’s

Pretty, pretty good. Red card bad. Must rescind.

For a team whose capital home stadium is 9000+ ft in elevation, Ecuador's struggling ... (Checks notes) Oh they haven't played in Quito in nearly three years? They play now at Guyaquil at (checks notes) 13 ft elevation? Because (checks notes) their Euro based players struggled at elevation? Oh.

Brendan Aaronson is the exact opposite of this. Guaranteed mishits.

You can say a lot of things about Sebastian Berhalter, but one thing you better say is that man strikes a ball clean as hell. Dead ball, from the run of play, for an assist or goal, he's going to hit it clean.

Berhalter understands ball physics. So nice to have that this WC.

A reminder that Tim Howard's saves record was set in a match that had 30 mins of extra time. To even come close to that in under 90 minutes is completely insane. www.theguardian.com/football/liv...

Ecuador isn’t exactly doing a Türkiye impression here either. Lots of quality chances. Courtesy @futi.live

The answer to the number of Poor xG shots by Türkiye last night is was 23. Just following up on this, courtesy of @mlsstat.bsky.social.

How do you get to 1.07 xG on 21 shots? One terrible choice at a time.

Early warning signs: bsky.app/profile/eric...

Can’t wait to see the number of Poor xG shots from Türkiye in this game. It’s going to be a lot. They have had some terrible shot selection, even up a man.

Ochoa could never.

After watching Morocco play for a draw for an entire half in the last game, this Scotland / Haiti game looks to be much more entertaining.

Only two “good” xG chances for the US tonight, including a goal on one, but the real difference was getting eight average quality shots (2 goals) while limiting Paraguay to mostly poor xG shots (7).

I’m significantly more concerned with the fact that US defenders continue to be easily befuddled by Germany’s off-ball movements than they were easily befuddled on a single set piece.

No Friday Stat Dump today due to family activities, but I do plan to post during the World Cup break. Stuff to come soon.

This is the highest single-game combined xG (8.80) in the @americansocceranalysis.com database (2013-present). Opta has it a little higher at 9.20, but regardless, one for the record books. Previous record was 7.13 (!!!) in 2019 between, get this, SKC and NER.

5 of the top 10 highest combined xG games in MLS history per ASA have come in *THE LAST MONTH.* Yesterday's 4-4 match between DC and Montreal ranks 6th. 7/10 have come since last September. Something is happening in MLS and I am here for it.

After some pandemic-induced cageyness, we're back to blockbusters. 2026 has the highest average single-game combined xG of any season since 2013. Enjoy it. Data: @americansocceranalysis.com

Judd scored in second-half stoppage at Providence last season. He had a stretch there of three goals in his last 15 min of playing time at Providence Park.

It was something like this bsky.app/profile/drif...

Brad Stuver just made the most Kung-fu looking save I’ve ever seen.

To provide a bit more context (proof) on ground crosses, in full MLS seasons between 2017 and 2025, teams in winning game states used ground crosses 23% of the time and teams in losing game states used them 17% of the time. Tied game state avg was 22%. You can look at the average lines here.

Friday Stat Dump: Crosses It's been 3-5 years since the whole debate about crossing good vs. crossing bad was waging, and, to some level, it's been going on since around 2017. To me, there are times when crossing can be good, and times it's clearly not good. First, who crosses a lot? 1/

Friday Stat Dump: Crosses It's been 3-5 years since the whole debate about crossing good vs. crossing bad was waging, and, to some level, it's been going on since around 2017. To me, there are times when crossing can be good, and times it's clearly not good. First, who crosses a lot? 1/

I like Recoveries Differential over Duels Won Differential for reasons I've previously explained: "duels" in data are not really the 50/50s we've been conditioned to think they are. But since some thought that's what the original visualization was, here is Duels Won Diff by game state. Enjoy.

So @johnspacemuller.com asked me what I'd do if I was suddenly appointed MLS commissioner, and somehow I skipped over the fact that I would do THIS. MLS teams insufficiently committed to doing media would be fined real money + GAM. And if that doesn't work, the CSO would be sent to the gulags.

For the absolute stat sickos out there, this is one of my favorite charts so far with the new game state ability: showing Recovery Differential by game state. Why do you think your team is positive or negative for a particular game state?

Friday Stat Dump: Passes per Defensive Action (PPDA) while Winning or Losing PPDA has become the de facto method of measuring pressure by those with event data. For those using ASA's Viz Hub, it's often combined with showing average defensive action height. Useful! 1/

Y'all are the best proxy QA team. The new "state" code wasn't implemented fully on a set of fields for passers, including "Intentional xA", causing some data replication in the joins. Behold the corrected Intentional xA numbers for Even player state (which re-emphasizes the players I highlighted).

That was a transparent image. Here it is with a white background to make it easier to read.

Friday Stat Dump: Game State / Player State This is obviously not a "stat", but I believe that deeper, and, honestly, better, analysis of teams and players can be unlocked by "slicing" and filtering data when winning, tied or losing (game state) and even, up, or down players (player state). 1/

Working extra long today, so Friday Stat Dump will be out tomorrow morning with a new dimension to my weekly analysis. I'm looking forward to sharing it!

This RSL / Portland game is all about RSL’s attacking patterns just dismantling Portland’s defense and then (usually) Pantemis making a great save to bail them out.

Portland does nice work here to take advantage of a little bit of sluggishness in RSL's rotations, get it between the lines and... waste it on an 0.02 xG rip from 30 yards. Minutes later they're down 2-0. This is what the fans mean when they complain about a lack of attacking patterns.

Been seeing some chatter about San Diego's poor defending this season, and so I took a quick look into it. They have the fourth worst xGA in MLS 2026 on Defensive Action Sequences, the biggest category. A closer look shows they give up the 2nd-most xG from opponent Loose Ball Recoveries. 1/

Bonus post: flip this around. Who is giving up the most xG off of clearance recoveries, and who is not getting pinned back? Nashville, Colorado, and others looking pretty good here...