After watching Russia fall into a heart-breaking and eliminating draw against Algeria, I did at least acknowledge the quality of the story that would be coming up, which would be a rematch between Germany and Algeria, offering the latter team possible revenge for the 1982 arranged German win against Austria that prompted a good bit of my post two days ago suggesting how to keep the 3rd group match competitive. However, that match should not be happening. Germany is the 6th best group winner in an abnormally competitive field (four teams went 3-0-0, which is highly unusual) and Algeria is the 5th best runner-up. They are both getting easier draw than they deserve.
Here’s the bracket as it stands in real life for the elimination games in the 2014 World Cup, starting Saturday:
But that’s nothing like what it should be. As I predicted in my earlier post, the Netherlands and Mexico are both being treated extremely unfairly, though I didn’t know then that they’d each be literally the best of their set of teams and getting the worst possible draw they could face. This is why random grouping is inferior to a seeding system.
For reference, here is how the 16 elimination rounders (on APDA, we’d call them octofinalists) fared through group play:
Yeah, I hate to break it to you new soccer fans, but the US isn’t all that good.
So here’s what the bracket should look like under my new system:
I broke the Germany/Brazil tie (the teams have the exact same stats overall) on how much they won their group by, setting up a Brazil-Chile match-up in real life and what’s most fair.
When we look at the comparison, here’s the teams that got lucky in reality and those who got remarkably unlucky:
Greece (+7) – it probably doesn’t matter that much, because Greece had a -2 goal differential and only got through because of a dubious call in extra time in their last game, but Greece should be getting walloped by Netherlands (the best group winner) and instead drew the worst group winner, Costa Rica. Who will still wallop them.
Costa Rica (+6) – now I’m rooting for Costa Rica and they impressed in a difficult group, but they are objectively the weakest group winner. They should be facing the best runner-up (Mexico), but instead get the second-worst (Greece). For what it’s worth, their second-round match-up would be the same (#1 Netherlands), so that’s reasonable.
France (+3) – France gets Nigeria when they should be getting Uruguay. All that means at this point is that they won’t be getting bitten, though FIFA ensured that it is decidedly less likely that anyone gets bitten playing Uruguay any time soon. In any case, this will probably actually be a close match even though France should be getting an even bigger challenge.
Nigeria (+3) – Nigeria should be facing Colombia and they instead get France. Colombia would probably mow them down, and like I said, I think they have a chance against France.
Germany (+2) – not a huge deal, because Germany is likely to destroy either Switzerland or Algeria, but they should have the harder match-up after not doing all that well in their admittedly difficult group. Though they should be getting mighty #3 Argentina in the second round, but instead get #5 France.
Netherlands (-7) – the top team in the draw may be out in the first round because they’re facing the best goal-keeper and a team that very nearly won a group with the host country in it.
Mexico (-7) – the best runner-up has drawn the toughest team in the draw that rolled through its group. Whoever wins this game is going a long way, but this game is a mighty injustice for both of these teams.
Argentina (-3) – Argentina should be getting the bye that is the United States (sorry folks, it’s kinda true). Instead, they have to deal with Switzerland, who still isn’t prepared to win a knockout game in the heat of Brazil.
Colombia (-3) – Colombia should be getting Nigeria, but instead will be facing Uruguay. This would be a really tough break if Uruguay still had their star player, but they don’t, so Colombia should still have little trouble going through.
Not only does that Netherlands/Mexico match stand out like crazy (yes, I’m going to keep harping on this), but the second-round looming match between Argentina and Belgium pits two teams that went 3-0-0 in their group and should rightfully both make the semifinals. Admittedly Argentina would probably choose Belgium over Germany, but Belgium would definitely pick the France-Uruguay winner over Argentina and they should have that more fair match-up.
I’ve come up with two objections to this improved system that are reasonable, but neither of them do I find sufficient to be deal-breaking. One is that the schedule would be problematic, because we’d need to add off-days between group play and elimination play to ensure that each team had enough time off. And consequently, we might have some teams who play in the later elimination games have a really long rest if they were in group A or B. I still think overall fairness of who you draw as your opponents and the avoidance of any possible rigging are of higher value than a precise amount of time off, however.
My friend Frese came up with the other objection, namely that Group G and H could really set their match-up because they’d know exactly where every other team would stand. Again, I think this compares pretty favorably to every team already knowing that because of the randomness that currently sets the elimination rounds. But I also question whether anyone would choose to be a runner-up instead of a group winner because they prefer the first-round match-up they’d get. After all, this system ensures they’d get a much harder second-round match-up in that case, so they’re unlikely to tank their seed because of the long-term implications.
I’m open to other objections, but I think they all pale in comparison to what Dutch and Mexican fans have to face on June 29th.
Now back to packing.