Feb. 25, 2011 at 5:34 PM ET
Online markets do at least as well as opinion polls when it comes to predicting election results, but how good are they at predicting the Oscars?
If the trading on the Hollywood Stock Exchange is any indication, you should bet big on "The King's Speech," the movie about King George VI's struggle to overcome his stuttering problem on the brink of World War II. The exchange's AwardOptions market works like an online futures market, but uses play money rather than real cash. Shares in Oscar-nominated movies and actors can be bought and sold. When it's time to settle up on Sunday, the "investors" who picked the winners get a $25-per-share payoff in play money. The losers get nothing.
With that in mind, here are the top prospects as of 5:35 p.m. ET Friday:
The idea behind prediction markets is that they distill the wisdom of crowds — particularly knowledgeable crowds — about a complicated phenomenon. The investors who are more confident about their informed opinion should be willing to invest more, and it's also possible to back out of your opinion to cut your losses if the market isn't going your way.
That's the way it works with the Iowa Electronic Markets, which is the only U.S. prediction market authorized to work with real money. The market operation is run by the University of Iowa's Tippie College of Business as a research project, and studies have shown that the IEM is at least as accurate as traditional polls for projecting presidential election results.
The IEM's researchers have used prediction markets not only to forecast elections, but also to predict the course of flu epidemics ... and box-office results as well. So how about Oscar awards?
"It's the sort of thing that's better suited to office pools than online markets," Tom Snee, a spokesman for the University of Iowa, told me today.
Unlike elections or box-office performance, the Oscar outcomes are decided by a relatively small group of 5,755 academy members — and they're not talking about how they voted. "There's really only a small number of people who know what's going on," Snee said. "There's no information to aggregate, it's just people speculating. So there isn't a whole lot we can do."
Although the IEM stays out of the Oscar prognostication business, it's getting more heavily into the box-office business. The university's M.B.A. students and faculty are already engaged in their most ambitious experiment yet, aimed at predicting the four-week box-office gross for "The Adjustment Bureau," a Matt Damon thriller that opens next Friday. The prognosticators are developing algorithms that account for the past performance of Damon's films, Damon's co-stars (Emily Blunt and Jon Stewart), the openings of other movies that weekend, the time of year (March is not the greatest season for new movies), the advance buzz and other factors.
"This is the same kind of analysis that professional marketers perform in their work, so it gives them an idea of what to expect on the job and helps them apply what they've learned in a class where there are real dollars at stake," Thomas Gruca, a marketing professor who is overseeing the contest, said in a news release.
As for the Oscars, one way to gauge the wisdom of crowds is to compare the outcome with the predictions from a living, breathing expert. Someone like, say, our very own "Scoop" columnist, Courtney Hazlett. Here are her best guesses in seven comparable Oscar categories as of Friday afternoon:
And just to pile things on, here are the market leaders from Intrade's prediction market for six comparable Oscar categories, expressed in percentage terms as of 5:35 p.m. ET. Intrade, which is based in Ireland, works with real money rather than virtual cash:
So it sounds like a classic human-vs.-market contest to me. Check back here on Sunday night to see how the real results compare with the predictions — and feel free to register your own forecasts as comments below. If you make a perfect prediction for the six top categories listed (best picture, director, actor, actress and supporting roles), I'll send you a free autographed pair of 3-D glasses as a reward. But the time stamp on the prediction must be no later than 12:01 a.m. ET Sunday, and I will accept only your first prediction (which means multiple guesses aren't allowed).
I realize that cardboard 3-D glasses aren't quite as glitzy as an Academy Award. On the other hand, what good is a golden statuette when you're watching a movie like this one?
Update for 12:30 a.m. ET Feb. 28: Seven out of eight ain't bad ... the only category that the Hollywood Stock Exchange's AwardOptions market didn't get right on Oscar night was best director: Tom Hooper ("The King's Speech") won out over David Fincher ("The Social Network"). Intrade also got that category wrong (although of course Hooper is at 99 percent now that the award has been announced). "Scoop" columnist Courtney Hazlett was correct about the director, but she missed the mark for the best-actress award (which went to Natalie Portman) as well as supporting actor and actress (Christian Bale and Melissa Leo). Bottom line? The big winner when it comes to predicting the Oscars is the wisdom of crowds.
More on prediction markets (and the Oscars):
Join the Cosmic Log community by clicking the "like" button on our Facebook page or by following msnbc.com science editor Alan Boyle as b0yle on Twitter. To learn more about Alan Boyle's book on Pluto and the search for planets, check out the website for "The Case for Pluto."