"there was not a single instance in the database where a candidate lost a state when he held a lead of more than 3.5 points in the polling average at this point in time. (Bill Clinton, in 1992, lost Texas despite leading George H.W. Bush there by that margin.)
It is possible to generalize these findings by means of a probit regression model, where the independent variable is the candidate’s lead in the polling average and the dependent one is whether he won or lost the state."
In Ohio, however, there are not just three polls: roughly a dozen polling firms, rather, have surveyed the state over the past 10 days.
There are no precedents in the database for a candidate losing with a two- or three-point lead in a state when the polling volume was that rich.
So Anon D in just a very few days you can look at the numbers and try and find an error on Probability theory falling short - Try and find one election where it did in USA using Silvers math - so far nada
And Intraders were far more better in 2004 -2008 then any of the polling firms but NOT perfect because they made a little error in 2010 Congress race's. Not a big deal but not perfect as 2004-08
Anonymous D replied with this 14 years ago, 1 hour later, 10 hours after the original post[^][v]#363,319
@previous (A)
Dude, you still don't know what I'm saying. The prediction market is out of line with the probability predicted by polling data. 52%/48% polling doesn't equate to a 66% probability of success.
Anonymous A (OP) joined in and replied with this 14 years ago, 25 minutes later, 10 hours after the original post[^][v]#363,326
Anonymous F joined in and replied with this 14 years ago, 15 minutes later, 11 hours after the original post[^][v]#363,329
@363,319 (D) > implying immediate polling scenario should be weighted that much more heavily than long term weighting of observations
Anonymous A (OP) replied with this 14 years ago, 1 hour later, 12 hours after the original post[^][v]#363,342
@363,319 (D) > Dude, you still don't know what I'm saying. The prediction market is out of line with the probability predicted by polling data. 52%/48% polling doesn't equate to a 66% probability of success
I do know what you are talking about. Your Mistrake is simple and its the one CNN and Fox news are using as well so don't feel to bad about your Mistrake.
Yes U would be correct IF That number you're looking at is HOW Intrade betters come up with the ODDS Result - It Ain't. The 52%/48% could even reverse to 48%/52% and get this the odds could be exactly the same or even better for Obama. Again its not about that number and who has more or less to arrive at the Betting Odds - Frankly the popular Vote has nada to do with the Odds shown.
That number means very little to less then nothing NOW - Its all about 11 States called the battle ground stages like Ohio and if you look at Ohio and Iowa and Nevada and 9 other key states as the betters are doing which includes Yours truly - Well given not one single time in History of USA has any Republican Won the Prezzzy without Winning OHIO. It could be done this time with a combo of other key states but Romney is doing Shit in far far far too many. So far behind in say Nevada & Wisconsin Dude look at the list of Key states and look at say just this one list of polls and Graph and angle of the Dangle
Click this to see each one and progress Romney ain't making anywhere were it counts to win the magic 535 Votes which is all that counts cause your popular vote numbers mean nada by law for USA as total