Comments on: Evaluating prediction market success in the 2008 election (…or, why wrong is right) http://blog.mercury-rac.com/2008/11/17/evaluating-prediction-market-success-in-the-2008-election-or-why-wrong-is-right/ A blog on prediction markets and innovation Mon, 22 Feb 2010 12:22:07 +0000 http://wordpress.org/?v=2.9 hourly 1 By: Fryzury http://blog.mercury-rac.com/2008/11/17/evaluating-prediction-market-success-in-the-2008-election-or-why-wrong-is-right/comment-page-1/#comment-2284 Fryzury Thu, 12 Mar 2009 20:46:12 +0000 http://blog.mercury-rac.com/?p=106#comment-2284 There are also hypotheses that economic conditions are an important determinant of election results, and that market prices reflect economic changes before polls do. There are also hypotheses that economic conditions are an important determinant of election results, and that market prices reflect economic changes before polls do.

]]> By: Fryzury http://blog.mercury-rac.com/2008/11/17/evaluating-prediction-market-success-in-the-2008-election-or-why-wrong-is-right/comment-page-1/#comment-1192 Fryzury Thu, 12 Mar 2009 14:46:12 +0000 http://blog.mercury-rac.com/?p=106#comment-1192 There are also hypotheses that economic conditions are an important determinant of election results, and that market prices reflect economic changes before polls do. There are also hypotheses that economic conditions are an important determinant of election results, and that market prices reflect economic changes before polls do.

]]> By: autos clasicos http://blog.mercury-rac.com/2008/11/17/evaluating-prediction-market-success-in-the-2008-election-or-why-wrong-is-right/comment-page-1/#comment-1022 autos clasicos Tue, 27 Jan 2009 16:21:30 +0000 http://blog.mercury-rac.com/?p=106#comment-1022 great article, and Obama win! yea great article, and Obama win! yea

]]> By: jedc_mercury http://blog.mercury-rac.com/2008/11/17/evaluating-prediction-market-success-in-the-2008-election-or-why-wrong-is-right/comment-page-1/#comment-751 jedc_mercury Tue, 18 Nov 2008 16:41:19 +0000 http://blog.mercury-rac.com/?p=106#comment-751 Hi, Peter. Thanks for the note.<br><br>What I believe is that you can't say right now that either InTrade or <br>the polls were more accurate back in July/August/September. While you <br>can compare the relative dynamics between the two, there's no way to <br>say one was better.<br><br>I hope to post more about this soon...<br><br>Jed<br><br>On 18 Nov 2008, at 17:58, "Disqus" Hi, Peter. Thanks for the note.

What I believe is that you can't say right now that either InTrade or
the polls were more accurate back in July/August/September. While you
can compare the relative dynamics between the two, there's no way to
say one was better.

I hope to post more about this soon…

Jed

On 18 Nov 2008, at 17:58, “Disqus”

]]> By: Peter McCluskey http://blog.mercury-rac.com/2008/11/17/evaluating-prediction-market-success-in-the-2008-election-or-why-wrong-is-right/comment-page-1/#comment-750 Peter McCluskey Tue, 18 Nov 2008 15:57:47 +0000 http://blog.mercury-rac.com/?p=106#comment-750 People have long speculated that polls produce misleading fluctuations shortly after the conventions. The results this year are consistent with the hypothesis that Intrade prices reflected that knowledge.<br>There are also hypotheses that economic conditions are an important determinant of election results, and that market prices reflect economic changes before polls do.<br>If you want to claim that Intrade's better forecasts for swing states in July through September were 90 percent luck and 10 percent knowledge, I won't dispute you. But if you're claiming Intrade's prices reflected no additional knowledge you should address the obvious hypotheses about how they might have done so. People have long speculated that polls produce misleading fluctuations shortly after the conventions. The results this year are consistent with the hypothesis that Intrade prices reflected that knowledge.
There are also hypotheses that economic conditions are an important determinant of election results, and that market prices reflect economic changes before polls do.
If you want to claim that Intrade's better forecasts for swing states in July through September were 90 percent luck and 10 percent knowledge, I won't dispute you. But if you're claiming Intrade's prices reflected no additional knowledge you should address the obvious hypotheses about how they might have done so.

]]>