Texas Gov. Rick Perry is tumbling on the political prediction site Intrade in the wake of a series of lackluster debate performances. The once formidable Perry stock is now selling for just .86 cents a share. Former Massachusetts governor Mitt Romney, meanwhile, just keeps going up in price — as of writing, he’s at $6.70 a share.

“Perry enjoyed a nice boost when he first announced he was running,” said Carl Wolfenden, the Intrade Exchange Operations Manager. “The initial enthusiasm for his chances has fallen away after largely being bested by Romney in the debates and being raked over the coals for his stance on immigration and the HPV vaccine.”

InTrade — and the bets it handles — have become candy to political junkies of late, a sort of day-trading service for politics. But is it an accurate predictor of the GOP presidential nominee — or any of the other political bets it takes? Maybe. Sort of. Let us explain.

How does InTrade work? There are actually two sites — Intrade.com and Intrade.net. The former uses real money, the latter fake money. Naturally, most of the attention goes to the site with real skin (or at least cash) in the game.

The system is simple. Anyone can bet on a yes or no proposition — buying if they predict an event will happen, selling if they think it won’t. All bets are priced between $0 and $10.

An example: You want to bet on the likelihood that President Obama will be reelected in 2012. You do it. He wins. All shares immediately go to $10 a piece. Anyone who bought shares makes a profit. If he loses, all shares immediately go to $0. If you short sold shares — selling without buying — you make a profit. You can also make (or lose) money by buying and then selling shares before the event is “settled.” All bettors must have enough money in their accounts to cover potential losses.

(Sidebar: The Fix is a terrible gambler.)

The site also posts a probability that an event will happen, which is just the price of shares turned into a percentage. So, for example, shares in the proposition that Romney will be the nominee are currently selling for $6.70. Thus, the site gives him a 67 percent chance of winning.

How big is InTrade? Hundreds of millions of dollars are traded each year, according to Intrade officials. A lot of the action is concentrated on whatever is in the news that day or week. Right now, over a thousand trades a day are happening on Romney, while only one person bet on the likelihood of a Category 3 storm hitting the U.S. by the end of Atlantic Hurricane season.

Where did InTrade come from? Intrade was founded in 2001 by Irish businessman John Delaney. The company is based in Dublin. Delaney was born in 1969 near Dublin, he got an M.B.A. in finance from University College Dublin and spent his early career in investment banking. In 1999, he went into online gambling. (Who didn’t?) He also founded TradeSports.com, a sports betting site that folded in 2008. Delaney died this May at the age of 42 while attempting to climb Mount Everest, a lifelong ambition.

Is it reliable? Yes and no. The site’s collective wisdom tends to be more reliable than than the cadre of professional pundits when it comes to forecasting election reults. (And, yes, we are including ourselves in the “professional pundit” category — for better and worse.) In 2008, bettors got only two states wrong — Indiana and Missouri. Bettors thought Indiana would go Republican and Missouri would go Democratic. Neither prediction was right. Those two states canceled each other out, however, keeping the site very close to the actual electoral college total. In 2004, the site got every state right.

Bettors tend to follow polls closely, and polls (when appropriately screened for bias or poor methodology) tend to be pretty accurate in the aggregate. Pundits have personal biases — affection for particular campaign consultants, a desire to not follow the pack — that Intrade bettors don’t.

“Intrade will ask ‘who do you think will win?’ while an opinion poll will ask ‘who will you vote for?’ Intrade also requires you to back up your prediction with cash, so traders tend to go with who they think will win rather than who they hope will win,” said Wolfenden. “You literally have to put your money where your mouth is.

But InTrade’s accuracy is also tied heavily to the amount of publicly-available information about a candidate or particular scenario.

Back in June, the site gave former Utah governor Jon Huntsman an 11 percent chance at winning the nomination. He had just gotten in the race; there were few polls and only a few hundred people betting. Huntsman, who has struggled mightily to gain traction, is now down to 2.9 percent.

Sometimes the market beats the polls to the punch. For example, in the 2004 election, Intrade was predicting that George W. Bush would win by August. On the other hand, the market did not see Hillary Clinton’s New Hampshire victory in the 2008 primaries coming. Of course, no one else — including Clinton — did either.

And, Intrade has successfully predicted events for which there are no polls — for example, that former North Carolina Sen. John Edwards would be the 2004 Democratic vice-presidential nominee and that Donald Rumsfeld would resign as secretary of defense in 2006. (Politics aside, InTrade has shown a knack for predicting Oscar winners.)

In the end, InTrade’s stock and, um, trade is collective wisdom. Sometimes that’s almost eerily accurate; sometimes it’s not. The more people are involved, the more likely it is to be right. But Perry’s poor standing is just a reflection of the news, not a prediction of the future.

Read more on PostPolitics.com

Obama outraises entire GOP field

Does Perry really want to leave Texas?

Romney parrots Trump on China

Herman Cain’s misleading pitch for ‘9-9-9’