i'm on Canada 2 units, -197, I'd be SHOCKED if they lose

[QUOTE=Fezzik;17874]Would love to see the stats on this.......I'm sure it shows a MASSIVE HFA, part of it has to be ref bias AND HFA.[/QUOTE] I only have NBA stats since the '02-'03 season. Since then my records show only 20 game 7s. Of course, that's too few to make a good sample, but in those games, the home team has covered 14 by an average of 9.75 points. Over the same time period in game 1s and game 2s, the home team has won by 6.72 and 9.57 points respectively in 102 trials. I don't have any NHL stats. So, in too small a sample size, I've got game 7 results being better for the home team than game 1, but basically the same as game 2 (although there are good reasons to think that the series that go to game 7 aren't a random subset of all series.) Given this data and these sample sizes, I'd hesitate to conclude anything. Certainly, we can't assign a 9.75 HCA for NBA game 7s, for obvious reasons, even if the sample size were sufficient. Someone else with more data or NHL data can provide those results. In no case does it answer the question of how much extra HFA one can assign to motivation, right?

to be clear...... [QUOTE=npc;17875]I only have NBA stats since the '02-'03 season. Since then my records show only 20 game 7s. Of course, that's too few to make a good sample, but in those games, the home team has covered 14 by an average of 9.75 points. Over the same time period in game 1s and game 2s, the home team has won by 6.72 and 9.57 points respectively in 102 trials. I don't have any NHL stats. So, in too small a sample size, I've got game 7 results being better for the home team than game 1, but basically the same as game 2 (although there are good reasons to think that the series that go to game 7 aren't a random subset of all series.) Given this data and these sample sizes, I'd hesitate to conclude anything. Certainly, we can't assign a 9.75 HCA for NBA game 7s, for obvious reasons, even if the sample size were sufficient. Someone else with more data or NHL data can provide those results. In no case does it answer the question of how much extra HFA one can assign to motivation, right?[/QUOTE] Is there any value at all in the Game 1/2 comparison? You hint at it, but to clarify, and point it out overtly, the game 1 and game 2 stats would always be better than game 6 and 7 stats (assuming 2-3-2 format) because they include total mismatches where the favorite in the series is -3500 and such, correct? The strong numbers for game 7 are meaningful because the teams have proven to be matched up fairly evenly. Game 1 and game 2 vs. game 7 in series that are priced competitively? Now we're talking.

[QUOTE=RealWorldSports;17879]Is there any value at all in the Game 1/2 comparison?[/QUOTE] I don't know. I wanted to compare the Game 7 results to something, preferably something statistically meaningful This is what I had handy. I'm not suggesting anyone draw conclusions based on this data, it's just something to discuss. BTW, I looked at games 1 and 2 of those series that went to a seventh game. The average point differential for those games was 6.8 and 10.1 repsectively. Basically the same as for all playoff series. Of course, this is a tiny sample size with 20 trials. [QUOTE] You hint at it, but to clarify, and point it out overtly, the game 1 and game 2 stats would always be better than game 6 and 7 stats (assuming 2-3-2 format) because they include total mismatches where the favorite in the series is -3500 and such, correct?[/QUOTE] You'd suspect this, but it's not necessarily true. Some 1 vs 8 series go 7 games (think Atlanta/Boston), some 4 vs 5 series are sweeps (think Chicago/Miami in '07). Still, yeah, you'd expect things to lean this way, although given the data above it doesn't really appear to be true. We could do more and compare all favorite home games, but that would take quite a bit more effort on my part. [QUOTE] The strong numbers for game 7 are meaningful because the teams have proven to be matched up fairly evenly. Game 1 and game 2 vs. game 7 in series that are priced competitively? Now we're talking.[/QUOTE] The strong numbers for game 7 aren't meaningful because the sample size is too small. Period. If I had a significant sample and if they showed a significant difference from other games in the series, then we can make some conclusions. Even with the small sample size, in the same series game 7 average margins are less than game 2 average margins. This is contrary to your statement. Game 7 (and game 2) margins are considerably bigger than game 1 margins, but with such a small sample size, I think it would be foolish to think you can draw strong conclusions. I'm still waiting for someone to provide some strong data in support of this argument.