I really like the trend on Intrade's Obama reelection chances -

[NRaleighLiberal]

http://media.talkingpointsmemo.com/slideshow/obama-up-on-intrade?ref=fpblg

"Investors Bullish On Obama Reelect : President's Odds Up On Intrade

Intrade is the on-line market house for placing bets on all manner of events. It's become a popular if unscientific measurement of people's expectations for political races, none more so than the race for president.

President Obama had spent the last months below or close to 50% chance of reelection. Buyers now put him back to well over 60% -- all at the same time Mitt Romney is continuing to stumble in the GOP primary."

[Alexander]

1. Be careful. Intrade had Romney as 97% likely to win Colorado.

Had Intrade been around in 1948, I'm guessing they would have had very good numbers for Dewey defeating Truman.

[Surya Gayatri]

3. That 97% may've been before the

heaven-sent "contraception" brouhaha.

Frothy Santorum enjoyed a splashy last minute surge thanks to that wedge issue.

SG

[Alexander]

4. It was 97% for Romney the day of the caucuses.

Really, when it was at 97% is irrelevant.

The fact that Intrade ever had Romney at 97% likely to win Colorado, and that he lost to Santorum, means that Intrade can, and sometimes does, get it wrong. Yet many political junkies, even on DU, are taking Intrade predictions and waving them around as if they are Holy Writ, seemingly based on their 2004 predictions.

Intrade also had Newt with about a 60% chance of winning Florida on the heels of his South Carolina victory, and in 2007 I'm sure Intrade had Giuliani and Clinton with good odds of becoming their respective party's presidential nominees.

In 1948, everyone expected Truman to lose, even First Lady Bess Truman. If Intrade had existed back then, a lot of people would've lost money betting on a Dewey victory.

[grantcart]

5. There was no polling in Colorado

PPP has been very accurate so far this year on polls and the others pretty close behind.

[Alexander]

6. But there were Intrade predictions - and they were wrong.

I agree, PPP is generally pretty accurate, especially this year. But not always. PPP got the 2008 Democratic primary in Pennsylvania wrong (they predicted Obama would win), and PPP also got the election in NY-23 wrong back in 2009 (they predicted Doug Hoffman, the Conservative candidate, would win).

My main warning to people is regarding Intrade predictions. Not only are they sometimes wrong, they're highly volatile, perhaps even more so than polls.

Right after Gingrich won the South Carolina primary, Intrade had him at over 60% to win Florida - that completely flipped within 48 hours, with Romney at over 60% to win Florida.

I was following the Intrade predictions because I was thinking of betting on Gingrich winning Florida. Now I'm glad I didn't.

[NAO]

2. source Intrade - Can I get some kind of "presidential default swap"

in the event of a Santorum victory?