World champions

[pne]

I have a philosophical problem with world championships such as the football one running right now, at least, if their aim is to select a “world champion”.

For me, that would imply that the “winner” is, by some objective, measurable standard, “the best”.

However, the way it is measured seems unsuitable to me.

For starters, it implies that there is a well-defined “better(team1, team2)” relation that can be measured—and that doesn’t depend on things like Tagesform (I don’t know of a good English word or phrase for that; basically, the way you feel on a given day, and/or the level of energy you have/the level you perform at on a given day). So I would think that something like “best 3 out of 5” (at least), ideally varying other conditions such as playing at least one game each in the team’s home and at least one somewhere else, would be needed between any two given teams to determine which of those two teams is “better”, rather than relying on exactly one game.

Also, the selection method seems to imply that “better(team1, team2)” is transitive: that is, if Team A is better than Team B and Team B is better than Team C, then Team A is better than Team C. Otherwise, you would have to have each team play every other team, rather than being able to do things like finals, where beating “the best of Group C” implies beating every team in Group C individually, thereby eliminating them all at once.

The fact that better(team1, team2) [or, more precisely, the relation that was tested during the current games, which is a proxy for better(team1, team2)] is not transitive can be seen easily in the group Germany played in: Germany won 4:0 against Australia, and Australia won 2:1 against Serbia, so Germany “should” win against Serbia—but in fact they lost that game 0:1.

So, while the current championship might make for lots of football viewing (for people who like that), as a measurement, I think it’s a sham and unsuitable, and the “winner” isn’t a meaningful relationship to anything in particular. (Not that that probably matters to most people who watch it.)

Comments

[sophie]

It's not meant to be a scientific method. It's a game - the fun comes in playing it, and watching others play it. You're right in that winning a game - *any* game - doesn't necessarily mean you're better at it. And to be honest, that's part of the fun of it, too.

At least, that's my view on it. (Note that I don't actually like football, nor do I keep up with the World Cup or anything. I'm basing my views on other games, and assuming football is similar.)

[pne]

It's a game - the fun comes in playing it, and watching others play it.

OK, fair enough.

But then I don't see the point in a championship like this -- if you want to have fun playing and watching it, you could simply have a collection of friendly inter-country games which get televised for people who like watching Uruguay play South Korea (or whatever).

It's the "eliminate other teams and rise to the top" mechanics that bother me; they don't seem to be compatible with a "play for the fun of it".

[sophie]

I think it's an excuse to have lots of games. ;p I mean, it's the nature of being competitive, I guess.

I do agree with you; it does seem pointless to me. (I've never been particularly competitive.) However much you play, you're never going to get a scientific result, even if all the matches were 'best of three'. And when it comes to something like the World Cup, it's already quite a long affair; having every match be 'best of three' would at least double the length of time it already takes, and could potentially be up to three times longer. And the result *still* might not be accurate.

Like I say, I think it's just an excuse to play lots of games. :D

[pne]

Oh yes - any reasonably scientific method would be completely impractical in terms of the time it would take to arrive at the result.

As for accuracy, on the LiveJournal copy of this entry, chipuni suggested that matches be played until one could say with 95% confidence (statistically) that one of the teams was stronger than the other one; though I don't know the mathematics behind it, they said that If one team always wins, then it only takes six victories to reach that level - so if the other team wins occasionally, too, you'd need even more games, so this would be even more impractical than "best of three" or even "best three out of five".