HexagonalBlog

How did the hosts do in opening matches?

I think it might be a good idea to look at previous World Cups and find out how the hosts fared in their opening matches.  Here's the full list, with the hosts listed first:

1930: Uruguay 1-0 Peru
1934: Italy 7-1 USA
1938: France 3-1 Belgium
1950: Brazil 4-0 Mexico
1954: Switzerland 2-1 Italy
1958: Sweden 3-0 Mexico
1962: Chile 3-1 Switzerland
1966: England 0-0 Uruguay
1970: Mexico 0-0 Soviet Union
1974: West Germany 1-0 Chile
1978: Argentina 2-1 Hungary
1982: Spain 1-1 Honduras
1986: Mexico 2-1 Belgium
1990: Italy 1-0 Austria
1994: USA 1-1 Switzerland
1998: France 3-0 South Africa
2002: South Korea 2-0 Poland, Japan 2-2 Belgium
2006: Germany 4-2 Costa Rica

So from 19 opening matches, there have been 14 wins by the host team and five draws.  No defeats. There are some variances in quality amongst the hosts, of course, but the host teams' performance in their World Cup openers is not something to be considered lightly.

I agree that South Africa are perhaps the weakest host team in the history of the World Cup finals, but Mexico have a tall order if they are to score a win on Friday.  It will certainly be a historic result if they do.

New CONCACAF coefficients updated, version 2010-04-28

The last set of CONCACAF club/league coefficients for the 2009-2010 season have been updated.  Visit my other site Soccermetrics and take a look. 

I'm in Wikipedia, baby!

I had no idea about this until I checked my referral logs the other day, but there is now a Wikipedia page on the CONCACAF club coefficients that references my work and my updated statistics.  I don't know who wrote the page, but I want to point out a few discrepancies:

• The league coefficient is explained but the results of the league calculations are not presented.  There are also +2 bonus points awarded for advancing to later rounds of the Champions League proper.
• The club coefficient formula is not presented.
• CONCACAF does not use my exact formula to determine seeds for future Champions League competitions.  I don't know what they have, but their calculations just happen to match what I'm seeing.  This is what CONCACAF officials have told me, but maybe they've started to use my formula, in which case we need to talk.

If I have time I'll try to edit the page, but the way Wikipedia works I'm sure it will be edited within 10 minutes of publishing this post.

Trip to Boston for Sports Analytics Conference

If you were following my tweets over the weekend, you would know that I was at the MIT Sloan Sports Analytics Conference on Saturday.  It was a collection of experts and personalities from the world of sports and those who like to consider statistical analysis problems related to them.    I made a lot of contacts and met a lot of interesting people, including some in the soccer business.  At my Soccermetrics site, I've published a couple of posts on some of the sessions that I attended -- the Emerging Analytics panel session in the morning, and the Limits of Moneyball session in the afternoon.  They give some idea of the major topics that were discussed which turn out to be quite common across all of the sports that were talked about. 

I'll update those two posts with some pictures that I took during the day.  There are a couple of other posts in the pipeline, too.

Another item of news to report is that I have a second Twitter channel that I will use to publish items related to soccer analytics work.  That way I can keep @hexablog dedicated to news on soccer in the North American region.  The new channel is called @socmetrics and if you're interested in that kind of work, please follow it.

Some discussion of the latest CONCACAF coefficients

A couple of months ago I released the latest CONCACAF club and league coefficients after the conclusion of the group phase of the Champions League, and promised to discuss them at some point in the future.  Now that the knockout stage is two weeks away, perhaps now is a good time.

Soccernomics review posted

I had mentioned in a previous tweet that I was in the process of writing a review of Soccernomics, the best-selling book by Simon Kuper and Stefan Szymanski.  I wanted to write the best review possible without giving away too many of the details, so it took a little longer to write than I would have liked.  But I've completed the review and posted it on my Soccermetrics website.  I'm interested in your comments, especially if you've read the book.  (I'm shutting off comments here and leaving them open in my post at Soccermetrics.)

Updated CONCACAF club coefficients, 2009-10-24 version

I've updated my CONCACAF club/country coefficients with data from the now-completed group stage of the 2009-10 CONCACAF Champions League.  I'll discuss the results in a future post, but the coefficients confirm the near-total domination of the Mexican clubs in the region's top club competition.

I've posted the coefficients on my Soccermetrics blog, and you can find the files here.

How the worm turned

Now that the Hexagonal is drawing to a close, perhaps it would be interesting to show how the teams' performance has progressed during the final round.  Below is a graph charting the point history of the six teams (horizontal axis is matchday number).

 

Costa Rica's early run of good form has kept them in contention for the top three.  USA had a hot start, leveled off during the most difficult part of the schedule, and hit their second win in the latter third of the Hexagonal.  Mexico and Honduras surged during the second half of the Hexagonal, but Honduras have stalled at the wrong time.

There will be a lot more analysis of the Hexagonal here and on my Soccermetrics website.

New updates to CONCACAF club coefficients list

I've posted a new update to my CONCACAF coefficients list on my Soccermetrics site.  The new season of the Champions League is only a couple of weeks old, but there has already been some movement in the domestic league coefficients. 

Mexico has the largest coefficient by a very large margin (almost a factor of 2.5 over the second-best coefficient), but USA has the second highest league coefficient.  It doesn't make sense if you go by the performance of MLS teams in last year's Champions League and this year's preliminary round, but Costa Rica's good results in the Champions' Cup have fallen out of the five-year window and USA are still riding the semifinal appearances of Houston and DC United in the final two editions of the Champions' Cup.  It's also interesting to find Puerto Rico already ranks third on the domestic coefficient list, but Puerto Rico's "league" is composed of just one team --- the Islanders.  That coefficient would likely diminish if teams from the PRSL find their way into the Champions League qualifying round. 

There's a long way to go in the Champions League, so these coefficients will change.

Coefficients updated

I updated the CONCACAF club coefficients a few days ago and uploaded a new file, but I've had a really tough time revising my original post to reflect that.  I think the link works, but I'm not sure if the filename is case-sensitive.  If you can't get to the database from the original post on this site, you can get it from the post on my other site, Soccermetrics.