2026 WC USA (et al): Worldwide Opponents Watch [all R's]

I’d be interested in seeing an analysis of how well ELO (or PELE) does at predicting WC success compared to how it does in other contexts. Seems to me the variables in rosters and competitions over time would make it much less predictive than with club play or individual sports.

 

I think the US is the hardest to predict because we have a lot of home games but don't always or even usually have a home crowd. It's not the same as being on the road in a hostile environment but it's also nowhere near the the home crowd almost (or maybe every other) country has. Will the US even have a home crowd at their WC games next month?

 

It won't be an away crowd....................................

If its mixed, that provides a nice environment.

Also, the US players are used to that. It shouldn't really affect them one way or the other.

 

So you're agreeing they won't have the home crowd behind them that every other nation would.

 

Australia will feel like a home crowd. Paraguay will be mixed. Turkiye will have more support than we do.

 

I'm really intrigued by what the crowds will be after FIFA totally screwed up the prices and availability. It's hard and expensive enough for US citizens to get last minute tickets and then have to pony up last minute expensive flights and hotels but to have to do that from another country and even continent would be ridiculously expensive. Had they done it more orderly and cheaper from the start I think would have allowed many more true fans to participate.

 

Y'all underestimating what the U.S. support is going to look like.

Paraguay is one of the smallest and poorest countries in South America. To think the WC home opener in the United States is going to be "mixed" (whatever that means, every crowd is "mixed") is silly.

It will be massively pro-U.S.

 

For sure. Elo is really a ranking system, most famously for chess, which is probably the most individualized sport out there. It's definitely gonna be better for that, and probably club soccer over international soccer too. I don't have any data handy to test that hypothesis.

However, that doesn't mean Elo as a predictor falls apart in the World Cup. Here's the Elo predictions plotted against actual results using the 448 World Cup matches since 1998.


Each blue dot represents 28 matches. The Elo win expectancy is the x-axis and the actual results are the y-axis. The orange dot would represent the expected points from those 28 matches given a "perfect prediction".

You can see r-squared is 0.91 for linear, so that's quite a solid record. The tails are messy and where things get a bit rough. Some of this is the 3rd group match, where favorites are already through and under perform (Brazil-Cameroon, France-Tunisia from the 2022 WC for example). I ran this excluding the 3rd game, but there's still enough whacko tail results to not improve it much (Saudi Arabia beat Argentina in the first game last World Cup, that's wild).

Elo of course is blind to everything except results, it's not really trying to be a predictor. That's where Silver's PELE should be an improvement. I'm sure these new variables do not got tremendous power, but here's his stated additions to what is basically an Elo model (PELE stands for Predictive Elo with Lineup Equilibria)...

• Transfermarkt data on player market values with their club teams for what’s essentially each nation’s best 23-man roster.
• Average player ages5 for this roster. Younger teams are usually projected to improve.
• A country’s region. We use 12 custom-built regions rather than FIFA’s six confederations.
• A country’s aggregate GDP, adjusted for the cost of living, inflation and global economic growth.6
• And a country’s soccer legacy, usually based on the year it or one of its predecessors joined FIFA.

 

The USA has vastly OVER performed at home relative to expectation historically, despite the crowds. I ran analysis for WC qualifying a few years back and our implied home advantage in those games was around twice what the 100 points Elo gives. It's on BigSoccer somewhere.

For PELE, Silver has calculated the HFA for every nation, and yep, USA is about twice the average.

First a graph which is kinda fascinating...

But on to the US.

PELE has the base HFA set at 50 "PELE Points". For the USA it's plus 89, around twice the base. We OVER index at home. The USA has the 45th best HFA of 228 nations in the soccer world here.

We're no Bolivia + 230 (good god!) or Bhutan +155, but historically our home advantage is very strong.

And, of course, the one and only most applicable example is the 1994 World Cup. And we over performed.

 

You have to factor in the diaspora as well but even then it's not as big as some other Latin American countries. I think Turkiye will probably have the biggest contingent in the group stage, if our home friendly against them was any indication.

 

I am factoring in the diaspora. It's not very big. The Paraguayan diaspora in the United States is estimated at somewhere around 30,000 people. It's tiny. We have seen Paraguay crowds in the Copa America. They aren't that big.

Turkey will for sure have the most fans but I think people are underestimating the pull for U.S. fans to travel from all over the country for this.

American fans will be at the U.S. games in force. We actually do have a lot of fans. They are just spread across a huge country. They will be flying in for these games.

This isn't a CONCACAF NL Semi-final vs. Jamaica on a Thursday at 5:30 p.m. in Dallas.

This is the World Cup.

 

Good stuff

 

Where PELE doesn’t seem to really improve on ELO is around injuries and the US’s somewhat unique dynamic of B teams for about half of our games. Did Silver try to adjust for that at all?

 

Based on leaks and suggestions from sales trends, the consensus seems to be that the Australia match in Seattle is selling very well, the Turkey match in Los Angeles is seeing mediocre sales, and the Paraguay match in Los Angeles is seeing relatively poor sales. At least that is what I remember from several weeks back.

I wonder if the USSF is going to regret not pushing back on FIFA about pricing for our group matches.

Because there was a 2 month period in late 2025 where fans could purchase World Cup tickets if they got through the lottery, with only Mexico, US, and Canada fans having the advantage of knowing their team's schedule. If prices had been more realistic (low hundreds for all but the best tickets), then maybe we would already have a home field advantage guaranteed with 50,000+ US fans for each match.

This may turn out to be a lost opportunity, since everything over the last year suggests that the United States holds enormous negotiating power over this tournament.

 

For the World Cup, they plan to use the actual 26 man rosters for the Transfermarkt data, so that will capture some of the injury absences. Maybe they will dynamically adjust as the tournament goes for each game as well as players get injured/suspended...here's what they say so far...

Two things we plan to account for are using actual World Cup rosters once they’re announced instead of our algorithmically generated rosters, and accounting for the incentives in each match. In matches where both teams would advance with a draw, for example, teams have shown a remarkable aptitude to conspire to secure one.

That last piece about the 3rd game is a good addition as I stated it's a drawback with pure Elo.

The B-team roster/injuries should have been accounted for to some extent with the Transfermakt data when they were building the model.

They also weight matches like Elo does of course. Elo has 5 levels of competitiveness for the point exchange. In PELE they seem to be a bit "smarter". Ther'e still 5 general levels...but there are levels within the levels it seems...
Friendlies: 0.5-0.7x9 multiplier
Minor and friendly tournaments: 0.7-0.9x
Regional tournaments and Olympics10: 0.7x (qualifiers)-1.0x (main tournament)
Continental tournaments (e.g. the Euros): 1.3x (qualifiers)-1.4x (main tournament)11
World Cup: 1.5x (qualifiers)-1.6x (main tournament)

So that accounts for some of the B-team stuff as well, though those intentional B-team Gold Cups probably do hurt us more than they should.

 

I think the Paraguay match is selling especially poorly because it’s priced especially high as the first match.

 

I haven't heard a single person on BS say that they've got a ticket for a US game. Does anyone?

 

If USSF had any say in ticket pricing with FIFA, then this outcome (obscene prices) would fall right in line with USSF's ticket pricing for every other match for the last, what?, decade.

 

I do. I have Paraguay and Australia tickets. I am going with one other person vs Paraguay and 2 others vs Australia.

 

France World Cup roster

🚨 OFFICIAL: France World Cup final squad. 🇫🇷

Any changes you’d make? 🏆👀 pic.twitter.com/46f3BD3AoN

— Fabrizio Romano (@FabrizioRomano) May 14, 2026

And some of the players left out

🚨⛔️ Some of the main players left out of France squad for the World Cup. ❌🇫🇷

❌ Eduardo Camavinga
❌ Khéphren Thuram
❌ Corentin Tolisso
❌ Randal Kolo Muani
❌ El Junior Kroupi
❌ Florian Thauvin
❌ Esteban Lepaul
❌ Pierre Kalulu pic.twitter.com/K2FfNXQOYc

— Fabrizio Romano (@FabrizioRomano) May 14, 2026

 

Americans on Mexico's provisional World Cup roster.
Mexico preliminary Squad for 2026 FIFA World Cup, Full Roster, Player List - Soccergraph

Julian Araujo
Richie Ledezma
Brian Gutierrez
Obed Vargas
Efrain Alvarez

It'll be interesting to see how many make the actual squad.....................................

 

Haiti has announced their World Cup squad.
US based players here, including US-born players. We can feel good for Derrick Etienne. Born in Virginia (while his father was playing for the Richmond Kickers). Former homegrown signing of RBNY, and currently on Toronto.

Duke LaCroix: Colorado Switchbacks
Deedson Louicius: FC Dallas
Carl Sainte: El Paso Locomotive
Derrick Etienne: Toronto
Danley Jean-Jaques: Philadelphia

No Fafa Picault, who'd switched to Haiti and been capped multiple times in 2025.

2055318290591805808 is not a valid tweet id

 

Bummed Picault didn't make it, always thought we could've used him as a spark off the bench in the 2018 and to a lesser extent 2022 cycles.

 

We'll see. Araujo is injured, right?

I believe that Mexico has never had an American-born player on a World Cup squad. Right?
So we're looking like we'll have at least one.

[We've had American-raised like Funes Mori.]