[extropy-chat] US election prediction markets failure

spike spike66 at comcast.net
Wed Nov 15 05:20:16 UTC 2006


> bounces at lists.extropy.org] On Behalf Of "Hal Finney"
> Subject: Re: [extropy-chat] US election prediction markets failure
> 
> Keep in mind that if a prediction market puts the odds of an event
> happening at 70%, as was roughly the case for the Republican Senate issue,
> it should come out the other way 30% of the time...
> Hal


Hey cool.  This election site showed how the pollsters predictions compared
with the outcome.  Yesterday they did the senate predictions, today the
house of reps:

http://www.electoral-vote.com/

I counted the total number of polls and the overall predictions.  I found
that for the senate the polls were about 2.4% republican-ward.  IOW, the
democrat party did 2.4% better than the pollsters collectively predicted in
the senate.

Today they published the senate results, which came out about 0.6%
democrat-ward, or the republicans did 0.6% better than the pollsters
predicted in the house.

We might be able to make some "money" off of this info in the next elections
in Ideas Futures IFX.  I had heard that elections seem to come out slightly
to the left of the poll-based predictions.  Now for the first time I have
some hard numbers, and this seems to have come true, altho not by much.
Next time I will check out the pre-election polls, then invest 2% to the
left of the collective prediction.

I wonder if that rule of thumb works in Europe?  Do you guys have elections
coming up?

spike











More information about the extropy-chat mailing list