If you need your fix of market data this morning and can't take another word out of the mouth of a man running for president, we've got Oscar predictions for you here at The Signal. Using the same methods to predict entertainment awards as we have done to predict political elections and sports games, we are extremely confident that The Artist will take home the Best Picture laurels; we give it an 89.7 percent likelihood of victory.
Unlike the state and federal elections, we have no polling to utilize for our predictions. Recent research has determined what the Academy electorate looks like, but it would be very costly to create a randomized selection of these 5,783 potential voters; due to their rarefied stature, they are not the easiest to reach in random polling.
But as we've somewhat gleefully pointed out, polls undulate with the wind while markets react with a cooler head. Thus, we base our likelihood off the prediction markets Betfair and Intrade. On these websites, users can buy and sell contracts on future events--will this or that happen?--that expire at a multiple of $1 if they occur and $0 if they do not occur. These two markets are real-money markets, so the players have an incentive to bet wisely and collectively drive the price of a bid to the realistic odds.
In 2011, Intrade correctly picked 6 out of 8 Oscars. This may not seem spectacular, but in fact it's just what the numbers say should have happened. By adding up the confidence of Intrade's picks (higher likelihood means more confidence), Intrade actually expected to get 5.91 out of 8 correct.
The prices on the Hollywood Stock Exchange, a well-known non-money site that trades contracts about Hollywood, paints a more competitive picture. Their market gives The Artist just 59.6 percent likelihood of victory with The Help at 16.9 percent and Hugo at 8.1 percent. Although HSX has an impressive track record, economic theory suggests that real-money markets are going be more accurate than equivalent non-money markets.
Some media outlets are relying on measures of positive and negative sentiment in social media to forecast the Oscars. We are big supporters of the value of social media to understand public interest and opinion. But, as of now, there is modest predictive power in these social indexes; the data is too new to properly debias it, correlate it with outcomes, and fully understand its relative merit. For example, Ron Paul has consistently dominated measures of social sentiment, but it has not translated into victories at the polls. Followers of this social sentiment index are declaring a tight race for Best Picture, whereas The Artist has an overwhelming likelihood of victory.
The below table shows the most likely outcomes in the major categories:
Oscars_Feb22at2PMET
Sources: Betfair and Intrade
Given the confidence expressed above, we expect to get 3 out of 4 of these picks correct.
Follow the Oscar likelihoods live on PredictWise. We will be particularly interested to see how earlier winners affect the likelihood of later awards.
David Rothschild is an economist at Yahoo! Research. He has a Ph.D. in applied economics from the Wharton School of Business at the University of Pennsylvania. His dissertation is in creating aggregated forecasts from individual-level information. Follow him on Twitter @DavMicRot and email him at thesignal@yahoo-inc.com.
169 comments