New Approach to Handicapping the Oscars

I just filled out my ballot for my office Oscars pool for this Sunday’s Academy Awards. Despite being the only gambler in the office, I still have never managed to win this pool. I used to think it was because everyone who works here knows a lot more about movies than me. But now, I think it’s just that I’ve been relying too heavily on bad information.
Usually, I’ll pick my ballot for the Oscars based on a combination of my own taste and the predictions of entertainment industry publications like Variety or Indiewire. What I had neglected, until now, is the vast amount of data provided by online bookmakers who take action on the Oscars.
According to an article in today’s New York Times, betting markets are incredibly reliable in picking Academy Award winners. And compared with horse racing, there are some interesting similarities and differences to pari-mutuel betting markets that show why.
Most of the offshore bookmakers who take bets on the Oscars are setting lines themselves, same as they would for any other sporting event on which they take action. A bookie researches the available data, makes a prediction and sets what hopefully will be a price that should attract equal action on all sides.
In multiple-entry contests, like the Oscars (or a horse race), there is no point spread, just the odds of winning for each entrant. Ideally, people will put some money on the longshots because of the attractive payouts and then that money can pay off the few shekels to the people who bet on the favorites (assuming the favorites win).
As bets come in and as new information comes in, bookmakers are free to adjust the odds. We’ve seen this happen this year as “Boyhood” went from being an odds-on favorite to win Best Picture to “Birdman” taking its place after winning the British Academy Film Awards top prize and the Directors Guild of America Award for feature film. Bookmakers took in new information and adjusted the lines. As you get closer to the contest, the odds get more and more accurate at predicting winners.
BIRDMAN IS THE BOOKMAKER FAVORITE

This is similar to handicapping in horse racing. The oddsmakers (and the bettors) in these Academy Awards wagering markets rely heavily on past performances of films in other awards shows. Since the Oscars go last, there is plenty of data from various award-giving organizations leading up to the big night that can help you predict a movie’s chance of success. According to FiveThirtyEight, these past performances in awards shows are as predictive as polling and 80% accurate during the last 25 years.
However these markets differ from horse racing in an important way. In horse racing’s pari-mutuel betting, the odds are determined by how the bettors wager, nothing more. The more money that is bet on a horse, the lower that horse’s odds go, and the higher the other horses’ odds rise as a result. There are no bookmakers to adjust the odds based on how the public bets and what kind of new information comes in.
Even though horse racing’s pari-mutuel markets are good predictors - favorites win roughly 35% of the time - they are far less predictive than the odds set by oddsmakers, who are free to adjust based on information. This is why it’s so boring to bet on the Oscars (or football for that matter) and so much more interesting to bet on horses.
In horse racing, you can find opportunities to make good bets where the public has called it all wrong.
There are markets where you can bet on the outcome of the Ocars that operate on public opinion rather than bookmakers. At the Hollywood Stock Exchange, you can see prices on the different films that rise and fall similar to stock prices in an active trading market. These prices are determined by the supply and demand for the shares on the market.
If you look at their offerings for the Best Picture award, you’ll see that “American Sniper” is trading at the third-highest price and (as of this writing) has been rising in value.
AMERICAN SNIPER WAS MOST SUCCESSFUL AT THE BOX OFFICE

This is because the public support for “American Sniper” is high. The film has blown every other Best Picture nominee out of the water at the box office, taking in more revenue than all the other nominees combined. According to the above-mentioned New York Times article, in a recent survey by Google, more than 40% of respondents picked “American Sniper” to win the Best Picture award, and no other nominee got more than 15%.
The question here is, asked to put their money where their mouth is, would these same people bet on “American Sniper” to win? Maybe, maybe not. But it shows why pari-mutuel markets are so rich for serious gamblers - the public sometimes gets it really wrong. In those situations, it pays to be the guy who got it right.
So I chose to pick my Oscar ballot based on the betting markets this year rather than the voices of the chattering classes or, even worse, my own opinions about who should win. After all, “The Lego Movie,” which was far and away the Best Picture of 2014, wasn’t even nominated for Best Animated Feature, so what do I know?