Voting Theory a Factor in Best Picture Selection

September 11, 2009

Now that the number of films competing for the best-picture Oscar has doubled to ten, voting theorists are weighing in on problems with the proposed selection method. "There are certain mathematical dangers with more nominees," Bruce Davis, executive director of the Academy of Motion Picture Arts and Sciences, told TheWrap.

To forestall the chances of, say, a nominee winning an Oscar with only a plurality, the Academy has calculated that the safest approach is an instant-runoff election (or single transferable voting). Academy members will rank the nominees from 1 to 10. Overseer PricewaterhouseCoopers will sort the ballots into 10 piles, based on each voter’s top choice. If any pile has 50% of the ballots, that nominee wins hands down. If not, the ballots in the smallest pile will be redistributed among the remaining piles, according to second-place votes, and so on, until one pile has 50% of the ballots—and, presto, we have a winner.

Voting theorists have long been critical of Oscar voting schemes, and some find flaws in the new system. "Some voters, by raising an alternative from last to first place, [could] cause it to lose—just the opposite effect of what one would want a voting system to induce," Steven Brams of New York University told Carl Bialik of the Wall Street Journal.

Nicolaus Tideman (Virginia Tech) called the new method a step in the right direction, although he supported voting involving head-to-head matchups (called the Condorcet system). Davis noted that no voting system is perfect. For the Academy's purposes, he suggested, "it is difficult to point to a better system than the preferential system."

Source: Numbers Guy, Wall Street Journal, Sept. 2, 2009; TheWrap, Aug. 31, 2009.