SEBA Projections

Updated SEBA international forecasts through June 8

Below are the current forecasts using data from matches through June 8, 2017.

Hexagonal Power Rankings

The “Power Rankings” we concoct are the actual “strength” of the team according to competitive expectations. The value represented on the X-Axis is the average amount of points we expect that team to receive when playing against every national team that participated in the previous World Cup on neutral territory:

The following chart shows the average points bonus for a particular nation’s home field advantage. This is important in explaining why teams may seem to have favorable odds to qualify even when ranked lower.

World Cup Qualification

The US’s chances of qualifying for the World Cup have increased from 77.49% to 86.30% from the last published forecast.

The next chart shows the probabilities of each of the USA’s potential Hex finishing spots.

The following chart shows the likelihood of the AFC team for the inter-confederation matchup (4th place CONCACAF vs. 5th place AFC).

The following chart shows the “Snub Index.” The idea is based on, if the World Cup were to be a tournament for the 32 truly-top teams in the world (at least according to SEBA), then this chart is sorted by how unfortunately damaged the top 32 teams’ chances are in qualifying for the World Cup.

The USA is not in this chart because they are not currently quite among the top 32 teams according to SEBA.

On the other side, this chart shows the nations outside of the top 32 who, perhaps unfairly, benefit from an imperfect qualification system and are ranked by their benefit (the table shows only the top 29 teams but the graph shows all teams with a chance).

The next charts show the probabilities of other confederations qualifying for the World Cup.

World Cup 2018

So… I went a little crazy over the MLS off-season and I decided I wanted to forecast the 2018 World Cup as well. The difficulty with this is that seeding for the groups depends on the FIFA ranking system, which means I would need to forecast every team’s rank to determine the “Pot 1” teams.

For those who don’t know, Pot 1 in the World Cup refers to the host and the top 7 teams according to FIFA rankings. These will represent the Group leaders for the first round of the World Cup. The remainder of the teams in the groups are placed semi-randomly with attempts to avoid over-concentration of teams from the same confederation.

For the last World Cup, the October 2013 rankings were the ones deemed ‘final’ before the World Cup draw so I’m now assuming October 2017 are the ones to shoot for. I also don’t have the forecast system perfectly aligned with reality, although it is quite similar. I have not built in the different point outcomes for PK-wins/losses in my FIFA ranking projections and I, obviously, do not have a mechanism for forecasting currently-unscheduled friendlies. However, I have forecasted every scheduled tournament between now and October.

Below shows the simulated ranks for the USA. Summary statistics are below the chart. USA’s most common forecasted rank is 25, but could go as low as 13 and as high as 60. USA’s current rank is 23.

Below is the probability of obtaining Pot 1 status. In the table, we also include the team’s qualification probability (if they don’t qualify, they can’t be in Pot 1), their current FIFA rank, and their confederation for helpful information.

USA did not not make Pot 1 in any of the 10,000 simulations. A lot of this is due to not having enough opportunities to ascend further in FIFA rankings. Mexico has a better chance in part due to currently being ranked higher, but also due to their qualification for the Confederations Cup, which gives a large importance multiplier in addition to top-ranking competition.

The following then manages to forecast the 2018 World Cup. Again, we accompany the odds of victory with qualification probability to help understand why some teams may be lower than expected.

The following shows the USA’s most probable group stage Opponents.

The following shows the various odds of each national team progressing out of the Group Stage.  USA has increased from 29.15% to 32.42% of making it through the group stage of the World Cup.

The following shows the probabilities of teams advancing through the Round of 16. The USA has increased from 12.15% to 13.54% in their chances of making it through the Round of 16.

The following shows the probabilities of teams advancing through the quarterfinal. USA has increased chances from 4.94% to 5.24% of making it through the quarterfinal.

The following shows the probabilities of teams advancing through the semifinal. USA has increased marginally from 1.80% to 1.82% in chances of making it through the semifinal.

Gold Cup

The following shows the Gold Cup champion forecast. The’s USA odds of winning the Gold Cup have increased from 32.3% to 33.14% from the last published forecast.

The following shows the probabilities of teams advancing through the group stage of the Gold Cup. The USA’s chances remain at 97.24% of advancing through the group stage.

The following shows the probabilities of the USA’s quarterfinal opponent.

The following shows the probabilities of teams advancing through the quarterfinal. The USA increased chances from 70.43% to 71.08% in advancing through the quarterfinal.

The following shows the probabilities of the USA’s semifinal opponent.

The following shows the probabilities of each team advancing through the semifinal. The USA has marginally increased from 49.20% to 49.78% in chances of advancing through the semifinal.

The following shows the probabilities of USA’s final opponent.

Other Tournaments

The following shows the Confederations Cup projections. CONCACAF’s representative, Mexico, has seen its odds of winning decrease from 18.46% to 18.26% since the last published forecast.

FIFA Rankings Comparison

For a discussion on how the FIFA rankings system differs from SEBA’s, check out the article from 3/28/2017.

The following shows, according to SEBA, the top-10 overrated teams to the left and the top-10 underrated teams to the right.

The following shows, if the USA wanted to schedule its friendlies to best manipulate the FIFA ranking system, who they should schedule against. The top table shows home matches, which are inherently the best choice for manipulating the FIFA rankings system. The second table shows, if the USA needs to schedule away friendlies, who they should schedule against in that scenario. The third table shows the current USA scheduled matches.

The MatchFIFAPts are 3-times-win-probability + draw-probability. rkPts represents the average expected points the USA might receive within FIFA’s point system for their rankings.

Scheduled USA Matches

Finally, the following shows the probabilities of results for upcoming USA matches.

5 Comments

  1. Andy Muenz says:

    Interesting that despite the win, the US qualification chances went down, most likely due to Panama picking up the point against Costa Rica, although I think the Australia win should have helped since they are now more likely to finish in the top 2 in their group rather than 3rd.
    .
    I’m thinking that some of the likely group opponents are incorrect, specifically with regards to the Asian countries. For the last couple of World Cups, AFC and CONCACAF have been potted together so they can’t face each other. Have you heard anything indicating that is likely to change?

    • Chris Sherman says:

      whoops, the text that said we decreased was actually leftover text from my previous update. I forgot to replace it. It has been edited/fixed. Thanks!

      the US odds of qualification increased to 86.3%, up from 77.5% before the win.

      I’ll look into your query regarding world cup group opponents. I remember when programming it that the guiding principles were only to avoid playing those in the same region, but I’ll double check that.

      • Andy Muenz says:

        As soon as the pots are announced, the US will know they can’t be grouped with anyone from the same pot. In the past, except for pot 1, the pots are arranged by region. For the least 3 tournaments, CONCACAF and AFC have been in the same pot (CONCACAF was with CAF for the 2 tournaments before that). Given that CONCACAF and AFC have the playoff for their half teams, we know that there will be 8 teams from these two confederations. Unless Mexico makes it to pot 1, I’m thinking it is likely they’ll stay in the same pot, although it is possible that if CONCACAF only qualifies 3 teams, they could be in the same pot as CAF.

      • Chris Sherman says:

        So you’re probably right, but there’s still a lot of uncertainty about what will happen in the draw.

        If we just take the best odds and follow the same pattern as 2014, the following would be the outcome:
        Pot1: 4 UEFA & 4 CONMEBOL
        Pot2: 1 CONMEBOL & 5 CAF & 2 UEFA
        Pot3: 3-4 CONCACAF & 4-5 AFC
        Pot4: 8 UEFA

        That’s not too dis-similar from 2014, so you’re probably right, but who is to say that a 3-team CONCACAF wouldn’t get paired with CAF instead of AFC?

        Ultimately, due to the uncertainty, in my program, I took a little shortcut, where I do a pot 1 as normal, saves remaining UEFA teams to the end, and in between randomizes the other confederations while making sure not to double-up any groups with confederations.

  2. I just stumbled on this site, I really enjoyed looking through it.

    Thoughts on the pots, I was interested to see that Poland jumped into a seeded pot, essentially knocking Portugal out go that Pot. Especially since Portugal will have three very winnable group games at Confeds Cup. Plus with Belgium destroying everyone in their sights, to drop below France who still has to combat Netherlands and Sweden in their qualifiers was interesting.

    In terms of CAF vs AFC, if CONCACAF only got three teams in that would mean AFC would have 5. So they would probably just keep it as is. Though it would be nice to have an opportunity to play some of the AFC teams, instead of always seeing two European teams and an African team.

    Though as things look right now, this might be the first time since 02, that they don’t have to play Ghana.

Leave a Reply to Chris Sherman Cancel

Your email address will not be published. Required fields are marked *

*