Marshall Flores and I have been running – what he calls – sims (simulations) for the preferential ballot on Facebook. As it’s way inside the internet bubble, Roma continually came out on top, just like Call Me By Your Name kept coming out on top last year. We both wondered or figured this outcome was unlikely, based on a few things we suspected but needed to prove. Last year, when we had only five nominees and took out Call Me By Your Name, we ended up with Shape of Water prevailing. That wasn’t enough to convince me it would win Best Picture, and I did not predict it. It won, of course. What we saw with Shape of Water last year was that it kept picking up votes down the ballot while also having significant support on the top of the ballot. In other words, people loved it enough to place it either at the top or near the top of the ballot, and if they didn’t, they liked it enough to still rank it above its main competitors.
Unfortunately, our Facebook polls were clearly in a bubble again this year, so much so that it was hard to get a read on Green Book’s chances. No one on there was placing it all that high – and in fact many put either it or Bohemian Rhapsody last, as people in the bubble are wont to do. That meant the poll was useless to us, except for one thing. Marshall noticed that Roma got the most number one votes but it didn’t really place high if it wasn’t number one. It didn’t really pick up votes in later rounds. So Roma turned out to be more divisive than Green Book. I got this wrong in my estimations. I assumed Roma would place higher, but I guess there were a lot of popular movies heading in, so much so that every Best Picture nominee went home with at least one Oscar. That means it wasn’t just two movies vying for Best Picture but probably more than that. That means no film won the first round. That means there were multiple rounds where ballots were transferred and retabulated, and in that scenario, Green Book prevailed.
First, here is a good explanation from Marshall:
Two things about the preferential ballot. First, if literally no one thought Green Book was the best film of the year, it would not have made it out of the first round of balloting where the #1 votes are counted. Period. Second, from what we’ve seen with almost a decade’s worth of voting simulations (both done by Dr. Rob with AD readers and on Facebook) a film needs to finish in the top two or top three after the first round to have a shot at winning.
In other words, although Green Book likely did not lead the field after the first round, it is apparent it had enough top of the ballot support to stay in range of the leader (presumably Roma), and because it was also broadly liked by voters, it picked up enough 3rd/4th/5th/6th/7th place votes in successive rounds to ultimately capture a majority by the end. And that, my friends, is the preferential ballot at work.
But here is what we knew beyond that, and why I ended up predicting Green Book:
1) This was a year where the big guilds split three ways. The only other time that has happened since they expanded was 2015 when Spotlight won SAG, The Big Short won PGA, and The Revenant won DGA. The DGA alone could not drive the win. Why didn’t The Revenant win PGA? That was the question. Moreover, why did the Big Short not beat Spotlight at SAG when it won the PGA? Preferential ballot vs. plurality ballot. That’s why. Although I got my own prediction wrong that year, it turned out that Spotlight beat The Big Short head to head on a big plurality vote among the actors, who still rule the Academy, thus giving it the edge.
This year, Green Book won on the preferential ballot at PGA. Black Panther won SAG, but it wasn’t up against Green Book. Roma beat Green Book at DGA and BAFTA, neither of which use the preferential balloting system. Roma might have had a better shot with a plurality ballot, as it had at DGA and BAFTA.
2) The foreign language category — So, now we go back to the year 2000, the era of the plurality ballot (five nominees for Best Picture, one round of voting, most votes wins) when Gladiator won PGA, Crouching Tiger won DGA, and Traffic won SAG. That was a year a lot like this one. But Crouching Tiger did not win Best Picture because Gladiator did. Why didn’t it? The foreign language category.
The second reason Roma didn’t win, other than it not picking up enough down ballot votes, is that the Academy has foreign language, documentary feature, and animated feature categories on purpose. They put films there so they can win without also winning in Best Picture. The Academy has not broken this trend in the 70 years they’ve had separate categories. They don’t have a need to. Or at least they didn’t. Many thought this was the year the pattern would break but by the end, Roma proved more divisive than Green Book.
3) Never presume to know how the mysterious “new voters” are going to vote based on skin color or country of origin. A film can only win on a preferential ballot if it has broad support throughout the Academy. It can’t just win because “old white guys” like it. It can’t just win because the new members like it. It has to be broadly liked, with a healthy number of number one votes to place in the top three before any recounts. Then, when they start recounting, it picks up number two and number three votes, number four and number five votes, and so on. Check out this video by Ryan Casselman — it ends up with Roma winning out of 100 ballots but you can see how the votes have to go down to really get how it works:
The way it ended up proves that Green Book hit across all markers. Old people, young people, white people, people of color. Otherwise it WOULD NOT HAVE WON. You can’t tell me that a night where Black Panther wins three Oscars (including two history-making wins for Ruth Carter and Hannah Beachler), BlacKkKlansman wins screenplay, Mahershala Ali and Regina King win supporting categories, Spider-Verse wins animated, and Roma wins Best Director that this was a night where the “old white guys” decided it. No.
I wonder if people are angry enough about Green Book to push to get rid of the preferential ballot. Probably. But at the end of the day, Green Book was very well-liked by everyone except the rising hive minds of Twitter. Add to that how it was attacked incessantly and unfairly throughout the season, and you can see how voters might push it to the top of their ballots in response.