I'm going to do things a little differently today. Usually we figure out how good we think various teams are, and then simulate a season or a tournament to determine their odds of winning. But we can also work backwards - if we know each team's odds of winning the tournament, we can use the simulation to figure out how good each team is, and use that to determine their odds of winning various head-to-head matchups.
First, the odds of winning overall: Betbrain.com has aggregated a significant number of betting sites' odds for who's going to win the gold medal in hockey. After removing the bookie's profit from the odds, we get the following predictions for the gold:
[As of 9 PM ET, 2/22/09]
Assuming those odds are roughly correct, we can work backwards to figure out how often we'll see each matchup in the final:
And also how often each team makes the Gold Medal game:
And of course we can find the head-to-head odds for some marquee quarter-final and semi-final games:
And for the finals:
Let's just say that the people who have money riding on this game ain't givin' the US no respect - the combined odds imply a money line of +200 for the US if they come up against Canada in the finals. Just to put that in perspective, the Toronto Maple Leafs have been a +200 or more underdog just twice this season, against Washington and Pittsburgh. Toronto was on the road in both games and was playing back-to-back games and they had Vesa Toskala playing.
Think about that: take last year's Stanley Cup winner and put them up against the worst team in the league, coming off a game the night before, with the worst goaltender in the NHL starting in goal for them. And that's how much of a slaughter the bettors think a Canada-USA final will be.
I think somebody's trying to either limit the number of Canada bets so that they won't lose money if Canada wins...Or they're trying to sucker Canadians into placing bad bets. But that's just me, maybe Canada really is that much better than the US.