One Romney backer may have lost $4 million trying to game the Intrade prediction market

By George Chidi
Saturday, September 28, 2013 18:23 EDT
google plus icon
Stock market via AFP
  • Print Friendly and PDF
  • Email this page

In the age of Nate Silver and election analytics, numbers matter more than ever. But as Megyn Kelly told Karl Rove on Election Night last year, sometimes the numbers are math that Republicans do to try to make themselves feel better.

That’s dangerous territory when the numbers have a dollar sign in front of them.

Perhaps this is an explanation for the results observed by a pair of researchers looking at Intrade options price movement during the election. The researchers — David M. Rothschild of Microsoft and Rajiv Sethi, of Barnard College, Columbia University — discovered that a single trader “accounted for one-third of all bets placed on Romney during our observational window,” they reported in a paper describing their observations earlier this month. That single trader, betting that Mitt Romney would win the election, lost almost $4 million doing so.

“This person was not naïve,” Rothschild told Raw Story in an interview. “This person was a sophisticated trader, making trades at a point in time and a method that would be most effective in moving the price.”

Intrade is a political prediction market, like a stock exchange tracking the odds of political outcomes. Participants essentially wager on who will win or lose a vote, with the volume and direction of the trades serving as a barometer for likely outcomes. For example, a contract to pay $100 if Obama won the election, sold at $80, implies an 80 percent chance of an Obama victory. Intrade successfully predicted 49 of 50 state presidential results in 2012.

The Ireland-based exchange closed to American traders in November after the Commodity Futures Trading Commission filed a lawsuit against the site.

During the campaign, Intrade presented a conspicuously large gap on Romney options between its prices and those on Betfair, a rival site based in the United Kingdom. Gaps like this allow traders to arbitrage away risk by betting in different directions in each exchange. In an efficient market, those gaps close quickly.

But the gaps persisted on Intrade. The study authors were able to obtain records of every trade made on Intrade for analysis, with dates, amounts and a unique anonymous identifier for the trader. And the pair believe that “by placing large bids for Romney and large o ffers for Obama,” a single trader “eff ectively created a firewall, preventing prices from moving in response to incoming information. This resulted in remarkable stability in Intrade prices for several hours on Election Day, and at other critical moments of the campaign, even as prices on Betfair were moving sharply.”

On Election Day, “it appears as if someone was trying to put a floor of about 30 on the Romney price and a corresponding ceiling of 70 on the Obama price,” the study authors wrote. “If so, the strategy was successful: the floor and ceiling both held firm for several hours until the fi rewall fi nally collapsed at around 9 p.m. ET.”

As soon as polls closed in Colorado, the last swing state in the race, the big Romney trader laid off his bets, causing an immediate reversion of Intrade prices to the Betfair odds, the authors said. “This trader spent over $375,000 to hold prices in place during the 90 minute interval starting at 7:30 p.m.” they wrote.

Rothschild said he believes it is plausible that this trader could have been attempting to manipulate public perceptions about Romney’s chances as a way to support the campaign. “The more viable a candidate looks, the more money they raise,” Rothschild said. “The more volunteers they draw …There’s a cascading effect of appearing viable.”

Intrade’s numbers seemed implausible to Nate Silver at times during the campaign, and large inexplicable movements on Intrade in Romney’s favor drew attention from time to time.

Intrade director Ronald Bernstein said in a note about the study on the Intrade site that some of the arbitrage opportunity may have been created by Intrade’s fee structure and regulatory hurdles. However, “It would not be appropriate for Intrade to speculate about the motivation for any individual members’ trading activity,” he added.

It’s possible that the trader in question may have been hedging against even bigger bets somewhere else, Rothschild said. But if the goal was to win hearts and minds, moving the Intrade market’s numbers had a willing audience, he added.

“Look at Karl Rove and Unskewedpolls.com,” Rothschild said. “There are a lot of people who wanted to believe.”

By commenting, you agree to our terms of service
and to abide by our commenting policy.