Snapshot Polls, Popular Vote, and Intrade

I have grown increasingly frustrated as the day has worn on. I keep seeing supporters, especially of Senator Clinton, referring to recent polls as evidence that she is more electable.  The fact is that we don't know who is more electable right now because there are simply too many variables, including what the months ahead have in store.  One thing is fairly certain, given the mixed and changing results in snapshot state and national polls, the Democrats would be foolish to select a candidate based on them.  

Further, in spite of what we hear from the Clinton people, there is no such thing as a national popular vote.  Too many apples and oranges. There are caucus and non-caucus states.  There are states that have included independents and those that haven't. Some have allowed party cross-overs and some have not.  These election results can't be combined into one figure, and especially not into a figure that will satisfy everyone. Bottom Line: if the Democrats back off from using the delegate count in nominating their candidate, and try to substitute a bogus national popular vote, they will be courting calamity.  (Few believe that such a substitution will actually take place. But as long as people keep talking about a popular vote, it pushes closure off into the horizon.)

If you really must have some numbers at this time, the odds are that the results from Intrade are probably more accurate than any one set of polls.  Of course Intrade isn't always accurate and its traders change their minds.  But it has a pretty good record. Interestingly, in the face of all of the current polls, as of 12:00 AM, May 23rd, the traders think Obama is going to beat McCain. actSearch/
Notice that there have been hundreds of thousands of trades, that is, "bets." 

The figures in order refer to:
Contract      Bid    Ask    Last     Vol       Chge

Barack Obama to win 2008 US Presidential Election    M     Trade        57.3    57.4    57.3    157381    +0.9
John McCain to win 2008 US Presidential Election    M     Trade        38.1    38.3    38.2    202761    -1.8
Hillary Clinton to win 2008 US Presidential Election    M     Trade        5.8    6.7    6.8    294285    +1.0

Tags: Caucuses, clinton, delegates, Election, Intrade, obama, polls, Presidential Race, Primaries (all tags)



I guess Bowers is shorting Obama and ...

I guess Bowers is shorting Obama and going long on McSame.

by Al Rodgers 2008-05-22 08:39PM | 0 recs
Bowers has issues with his maps

He specializes in posting maps that aren't quite based in reality.  That said when he focuses on issues he does a good job.

by Student Guy 2008-05-22 08:49PM | 0 recs
Re: Snapshot Polls, Popular Vote, and Intrade

money, meet mouth.

great diary.  I'd rec if I could.

by chinapaulo 2008-05-22 08:50PM | 0 recs
Re: Snapshot Polls, Popular Vote, and Intrade

Thank you.  I am posting a slightly modified version on my blog

by Mitchell A 2008-05-22 08:57PM | 0 recs
Re: Snapshot Polls, Popular Vote, and Intrade

hey -- I read your whole blog last night.  It's boss, dude.  I realized I had read some of it before here at mydd and agreed with and enjoyed all of it.  Cheers.

by chinapaulo 2008-05-23 11:08PM | 0 recs
Re: Snapshot Polls, Popular Vote, and Intrade

Interesting.  Clinton is trading at 6.6 to get the Democratic Nomination, but 6.8 to win the presidency.  That basically means that, according to the money, she has better odds of being elected president in 2009 than she does of winning the Democratic nomination.

I can think of several possible interpretations of that, none of which I even care to speculate about.  The bottom line though is that, if she wins the nomination, people think she is virtually certain to win the presidency.

by markjay 2008-05-22 09:07PM | 0 recs

The Intrade market is betting very strongly that Clinton is much more electable against McCain.  

Her bid price to win the nomination is 6.4, while her bid price to win the election is 5.8.

Some simple math suggests that the market is determining her likelihood of winning the election if nominated at a whopping 91%.

Obama's bid price to win the nomination is 92.7, while his bid price to win the election is 57.3.

The market thus believes that his chances of winning the election if nominated is 61.8%.

Make of that what you will.

by Fuzzy Dunlop 2008-05-22 09:11PM | 0 recs
Re: Actually

If you've followed Intrade a lot, as I have, you would realize that the markets are not efficient at the margins.

At this 6%, they are assuming she is very unlikely to win through a simple SDs overturning PD majority.  6% probably only reflects chances that he has some kind of major calamity that forces him from the race.

Reason I say that is that Hillary would probably be a big favorite as long as she could count on Barack voters to support her.  She is getting that now in polls because there is very little backlash against her right now.  Barack supporters feel their candidate is going to be the nominee, so they are not expressing a protest vote.  I think some Hillary supporters are (which is why Intrade predicts he is better off than current polling show).

If the nomination were really up in the air (as it was a little bit before IN/NC), her polling would reflect some of that backlash, and hence her nomination odds would be higher but her implied GE chances would be lower.

And, as everyone with sense knows, no matter what, Hillary would not be a 90% favorite to win, even if she entered the race clean (meaning no backlash from having SDs overturn PDs).

by sasatlanta 2008-05-22 11:16PM | 0 recs
Re: Actually

I am not sure they are efficient anywhere, and never made that claim.

Hence my "make of that what you will" comment.

by Fuzzy Dunlop 2008-05-23 03:57AM | 0 recs


Advertise Blogads