Comments on: Senate Prediction http://www.secretsofthecity.com/mnspeak/senate-prediction Minneapolis + St. Paul Sun, 22 Nov 2009 06:28:20 -0600 http://wordpress.org/?v=2.8.4 hourly 1 By: justpbob http://www.secretsofthecity.com/mnspeak/senate-prediction#comment-201742 justpbob Mon, 24 Nov 2008 14:52:00 +0000 #comment-201742 Both righty tighty and lefty loosey blogs feed on outrage, Tom. They might be secretly rooting for Franken. It will give them six years of material. Both righty tighty and lefty loosey blogs feed on outrage, Tom.

They might be secretly rooting for Franken. It will give them six years of material.

]]> By: Tom Bartel http://www.secretsofthecity.com/mnspeak/senate-prediction#comment-201740 Tom Bartel Mon, 24 Nov 2008 14:45:36 +0000 #comment-201740 And that line 54 stuff is just the sort of evidence Powerline will use to prove that Franken cheated. And that line 54 stuff is just the sort of evidence Powerline will use to prove that Franken cheated.

]]> By: Tom Bartel http://www.secretsofthecity.com/mnspeak/senate-prediction#comment-201739 Tom Bartel Mon, 24 Nov 2008 14:43:38 +0000 #comment-201739 My prediction is a small nuclear explosion at Powerline if Franken wins. My prediction is a small nuclear explosion at Powerline if Franken wins.

]]> By: jeffk http://www.secretsofthecity.com/mnspeak/senate-prediction#comment-201722 jeffk Mon, 24 Nov 2008 13:44:00 +0000 #comment-201722 The first time I read it I was confused too; it's not very well explained. But in a word, I think what he's doing is using the correlation between Franken gaining votes and a lack of challenges to estimate what percentage of the challenges will fall for Franken. It's interesting, and I suppose it's about the best statistical analysis a person could do with the information we have, but it's a stretch. The statistical error bars are such that this only gives Franken a 52-48 edge (this is mentioned later in the thread), and that's assuming the methodology is right. and everything is appropriately random, which of course it isn't. The first time I read it I was confused too; it’s not very well explained. But in a word, I think what he’s doing is using the correlation between Franken gaining votes and a lack of challenges to estimate what percentage of the challenges will fall for Franken. It’s interesting, and I suppose it’s about the best statistical analysis a person could do with the information we have, but it’s a stretch. The statistical error bars are such that this only gives Franken a 52-48 edge (this is mentioned later in the thread), and that’s assuming the methodology is right. and everything is appropriately random, which of course it isn’t.

]]> By: g rote http://www.secretsofthecity.com/mnspeak/senate-prediction#comment-201720 g rote Mon, 24 Nov 2008 13:20:55 +0000 #comment-201720 I wonder if the sports books in Vegas have an over/under line on this. If it was at Franken @ 27 votes I would take the under. I wonder if the sports books in Vegas have an over/under line on this. If it was at Franken @ 27 votes I would take the under.

]]> By: kurtis http://www.secretsofthecity.com/mnspeak/senate-prediction#comment-201719 kurtis Mon, 24 Nov 2008 12:51:24 +0000 #comment-201719 I'd check line 54 of the rules.input_evaluators.inc script and make sure the first argument in the callback to the call_user_func_array() function is valid. I’d check line 54 of the rules.input_evaluators.inc script and make sure the first argument in the callback to the call_user_func_array() function is valid.

]]> By: kurtis http://www.secretsofthecity.com/mnspeak/senate-prediction#comment-201717 kurtis Mon, 24 Nov 2008 12:47:55 +0000 #comment-201717 I really wasn't able to follow Nate's argument this time. Too much math, not enough name calling. He should just accuse both sides of cheating like the Star Tribune commenters. I really wasn’t able to follow Nate’s argument this time. Too much math, not enough name calling. He should just accuse both sides of cheating like the Star Tribune commenters.

]]> By: sandburg http://www.secretsofthecity.com/mnspeak/senate-prediction#comment-201715 sandburg Mon, 24 Nov 2008 06:35:04 +0000 #comment-201715 (That is not based on any sort of mathematical formula or simulation, fyi.) (That is not based on any sort of mathematical formula or simulation, fyi.)

]]> By: sandburg http://www.secretsofthecity.com/mnspeak/senate-prediction#comment-201714 sandburg Mon, 24 Nov 2008 06:34:31 +0000 #comment-201714 Is this where we place our bets? I'll say Franken, but with 20 votes. Is this where we place our bets? I’ll say Franken, but with 20 votes.

]]>