Monday, 13 October 2008

Opinion polls are under-representing the strength of the Obama vote

There has been much ballyhoo from the right wing blogosphere (including this one) regarding a number of issues that seem material to the assessment of Barack Obama including: his birth certificate; his connection to voter fraud organisations such as ACORN; his associations with a veritable Rogues' Gallery of unsavoury characters including Rezko, Ayers, Wright and others; and, recently, the possibility that he had an affair with some woman who has been paid to 'disappear' to the Bahamas.

Much as I agree that Obama is an unknown entity and therefore it is quite legitimate to investigate his acquaintances in order to better understand who he is the problem is that the mainstream media, for the most part, is not interested in upsetting the Obama apple cart having abrogated their journalist integrity through the Democratic primary phase, and Joe and Jill Public simply aren't interested in hearing negatives about the candidates. Instead, they are looking for a candidate with the best sounding answers to the current issues facing the country.

I have said it before and I'll repeat it again now - the public opinion polls are not reflecting the size of the lead that Obama actually enjoys. Current polls are showing that he has an advantage in the 4-7 point range.

I'd suggest that it's much higher than that and I'll tell you why.

Consider the following graph from Betfair's next president market:



With nearly $17 million in matched bets Obama is $1.21 to win, which is an implied chance of around 81% to McCain's 19%.

I have spent many years looking at the correlation between opinion polling and betting markets.

In my opinion, if Obama's lead is really the 51-45 that Rasmussen gives him or the 50-43 from Gallup then the prices would look more like: Obama $1.45 (implied chance around 70%) to McCain $3.30 (30%), especially given that McCain will do well with independents.

That's still a healthy gap so why do I believe that there is more support for Obama than being reflected in the polls?

Because, in my experience, a price of around $1.20 indicates a significant component of 'insider trading'.

That is, people with access to internal polling data are betting on the outcome of the election and are confident enough to have backed Obama in from $1.30 to $1.20 at the same time as the mainstream media started to ask questions about his association with Bill Ayers.

That sort of confidence can only come from people in the know.

Which is why I think that the opinion polls are wrong, that Obama's lead is probably more like 12-15% and that McCain needs to pull a huge rabbit out of his hat in order to win this election.

(Nothing Follows)

4 comments:

Bagua said...

Interesting concept, how accurate are the odds in measuring results on the betting site you refer to? Does a bet rated 80% win 80% of the time?

The idea of high confidence indicating insider knowledge at a certain level is interesting and quite true in many cases, alternatively it could in some cases result in a self fulfilling prophecy, or alternatively just reflect more just a consensus.

In the Obama/McCain race, (no pun intended,) I expect that this is mostly an informed consensus bet based upon all the favorable media.

Not to mention, most hockey mom's don't bet online. :-)

rightwingprof said...

Your thesis is based on no data. If you actually look at the evidence -- that is, that the %age of Dems, GOPs, and Indies that have shown up at the polls to vote for the last thirty years is remarkably stable, and compare it to the weights of the polls, you see that Democrats are vastly overrepresented.

Online bets are fun, I suppose, but they aren't data, nor are the to be taken seriously.

Anonymous said...

Jack

We're destined to disagree about this. One would think that the economics of the favorable odds on our Dear Leader would draw counter bets from those who like the odds based upon what they know or think to the contrary. But there is no reason whatsoever to believe that the community of bettors is representative of insider...or even informed... views.

As an example, I find the odds attractive, but I ain't buying a ticket to ride. And I'll play craps anytime I walk past an open table.

Rather, the current odds are representative of those who bet on that sort of thing. And who knows what feeds their information stream? A betting site in the UK is certainly going to be frequented by lots of Euro types whose media, if it were at all possible, is even more in the tank for The One than in the US.

This election has not been decided yet by any measure whatsoever. The Messiah has not closed the deal with the independents. And, as I have posted a few times here, a great number of working class Dems are lying to the pollsters. They did it in the primaries, and they're doing it again.

Lord Obama appears to have the wind at his back. And that is precisely what the elitists in the political class, the media and the Obama campaign want us all to think.

But the ballots haven't yet been counted. I know it's a pesky little detail, but let's wait until then before we jump off the tall bridge. We can make a party of it and have a nice group leap into darkness right at sunset.

But I'm betting against it.

--Krumhorn

lucas said...

I would not be too surprised if there was some deliberate attempts to
influence voters by some well-heeled obama supporters who might be
laying large amount of money on obama to win and purposely distort the
odds offered by these various betting sites.

From various articles is does seem that many of the poll results are
purposely distorted to obama's advantage and an obvious follow-up would
be the distort the odds shown by betting sites that might be used as supposedly independent confirmation of biased and manipulated poll results.

A significant part of campaigning is to get a physcological momemtum
going to attract the undecided to vote for what seems to be the obvious
winner. It is just one of the many ways in influence crowds of people
with a form of herd mentality.

In this campaign there is also an underlying threat of post-election
Kenya-style violence from the obama supporters if obama is defeated.
Poll results and betting odds make an easy justification for a "stolen
election" narrative that can be used as an easy way to incite and maintain violence.

Remember obama was a supporter and advisor of the recent Kenyan election loser Odinga.
Odinga's strategy following his election loss was to incite a massive wave of violence across Kenya and in fact succeeded in gaining a shared power by this tactic.
We dont know just how much obama was involved in the planning of this violence as a tactic, but it is not something that I would dismiss in the current environment.