Share

Steve Pond’s Everything you Need to Know about Oscar Ballot Voting

Good thing Steve Pond covers the Oscar race. The Wrap’s awards journo no dives right in.

56 Comments on this Post

  1. Al Robinson

    Holy Shit!! That way of voting for Best Picture is really Stupid, and really Great all at the same time. Man, there may be no telling which movie will win this year. For all we know, it could somehow end up being Captain Phillips or Nebraska, cause I feel they might be a lot of people’s 3s or 4s or 5s.

    AWESOME!!! :-)

  2. Al Robinson

    This kind of voting could also open the race up to a BIG problem though. The voters could end up diluting the best movie’s chances. Say people don’t want to vote for 12 Years a Slave, but they have to. They rank it 9th then. What we could see happening is that the most deserving movie end up getting diluted because their opponents want to hurt their chances. Then what they do, is instead, less deserving movies end up ranking higher. Now is that to say they’ll win. No, but what I see could end up happening, is with all this strategy, we somehow DO end up with American Hustle winning Best Picture. Almost no one will want to hurt it’s chances, but not fully want it to win either. We’ll end up in the middle.

  3. Al Robinson

    Ugh. :-(

    But, lets hope that either 12 Years a Slave, Gravity, or The Wolf of Wall Street win. *Fingers Crossed*

  4. This is how I always thought it worked and then people insisted that it was about #1 votes. That’s how middling junk keeps winning. So if you’re betting you’re better off going with something you think a lot of people will have in second or third place as long as it gets a fair amount of #1 votes too.

  5. Preferential balloting isn’t simple by any means, but the switch to 5+ BP nominees necessitated it. Otherwise it’s certainly possible that a film could win BP with a minimum of 11% support (assuming there are 9/10 nominees) under the old first-past-the-post system – how could a BP winner have legitimacy if almost 90% of AMPAS didn’t even vote for it? The potential for vote splitting, spoilers, and other tactical voting shenanigans also would have also dramatically increased if the old system remained in place. Preferential balloting is far more resistant to tactical voting – it doesn’t make it impossible, but there would have to be **significant** voter collusion and awareness of each other’s preferences for it to occur. And that assumes a level of AMPAS voter sophistication that is really not supported by anything.

  6. Al Robinson

    IDK. I guess the problem here is, most voters know which 1 movie that want to win. After that, I think it gets mucky. So many movies get mashed between #s 3 – 7 or even 8. I mean, if I was voting, after The Wolf of Wall Street 1, and Gravity 2, and guess I’m not partial to any of the next ones winning, or not winning. The only movie I’d make sure to help not win, and I’m sorry for saying this, is Philomena. That one I would put at 9.

  7. Al Robinson

    So, in other words, in a manner of speaking my preferential ballot would look something like this:

    1. The Wolf of Wall Street
    2. Gravity
    3. 12 Years a Slave, American Hustle, Captain Phillips, Her, Nebraska
    8. Dallas Buyers Club
    9. Philomea.

    Obviously, I can’t put 3 for all 5 movies, but it’s just an example of what I mean.

  8. Al Robinson

    One last thought, before I call it a night. Just as long as we don’t get Oscar’s first TIE for Best Picture. For the love of all that is good and holy, please AMPAS, don’t let that happen!

    Please. :-)

  9. AMPAS has already come out and said there won’t be a BP tie. There are tie breakers in place, and it’s safe to assume the primary tie-breaker will be amount of #1 votes.

  10. Can we please go BACK to just FIVE Best Picture nominees? Expanding the field to 9 only dilutes the honor. After all, 5 is good enough for all the other categories (except makeup, which should also be 5).

    Best Picture is the only category that employs preferential balloting. Why isn’t the movie with the most votes the winner? That’s how all the other Oscars are determined. And it’s how we elect politicians.

  11. I guess the problem here is, most voters know which 1 movie that want to win. After that, I think it gets mucky. So many movies get mashed between #s 3 – 7 or even 8. I mean, if I was voting, after The Wolf of Wall Street 1, and Gravity 2, and guess I’m not partial to any of the next ones winning, or not winning.

    Counterintuitive as it may sound, as long as voters sincerely rank their top few choices, being indifferent about the rest of the nominees doesn’t really “break” the IRV system.

    Keep in mind that the ballots are only transferred to the next ranked film still in contention if the first choice is eliminated. Someone who hated, say, 12 Years and ranked it 9th doesn’t directly affect 12 Years chances – it just means its the last film that ballot would get transferred to if all other films that are ranked higher were eliminated. And if all other films were eliminated, then by definition 12 Years is the last film standing, and hence, the BP winner.

  12. Why isn’t the movie with the most votes the winner? That’s how all the other Oscars are determined. And it’s how we elect politicians.

    As I stated earlier, with around 10 BP nominees it’s very possible that there can be a BP winner that the vast majority (potentially upwards of 90%) of AMPAS didn’t even vote for, which would raise serious questions of legitimacy.

    And it’s been shown that IRV and plurality voting generally arrive at the same winner. If there is a difference in outcome, the IRV result is fairer because by construction it represents the consensus opinion whereas the plurality winner is likely a result of a vote split.

  13. I have an issue with how this video ended. It gives the impression that the number 3 film from the first round wins in the end. While this is a mathematical possibility, it is an unlikely one.

    Let’s take the situation where there is no bias—no polarizing films in play. This means that with each round when there’s a significant number of ballots, their redistribution should follow the percent distribution of the remaining films’ ballots. In other words the top films get the top percentage of the redistributed ballots. (Kinda the opposite of the second redistribution seen in the video.) So the top films remain on top. Now there would be some slight variation—with this film getting a little more than what they should under the distribution and another getting less—but nothing major. So the only way that a number three could make up the difference in the last two elimination rounds is if the top three were really close on the onset (like in the video). Then those slight variations would come into play. But to have three films with the same amount of love to elicit the same voting numbers—even close numbers—would be statistically improbable.

    Another possibility is some external factor (or bias) influences the numbers. Bias here would mean polarization of some kind. Assume we are not in the above (unlikely) tight three way race. Then, number 1 and 2 from the first round would have to be polarizing enough for distant number 3 to start racking up the redistributed votes. This to me is the more likely scenario over the three way race, but I don’t even think it’s that likely, as number 1 and 2 have to be polarizing in the opposite direction; i.e. Film No. 1 fans rank Film No. 2 fans low and vice versa.

    Either scenario’s likelihood is so slight from a probable standpoint, that it does not look good for number three winning. Again, it is a mathematical probability, but not likely.

    This year, it appears to be a three way race: 12 Years a Slave, Gravity, and American Hustle. Based on the DGA (historically a strong correlation), and the PGA (using the preferential ballot), I would say that the top two are Gravity and 12 Years. That tie at the PGA speaks volumes.

    Going into the final balloting, it’s going to be one of those two. Thank God, they are both awesome films. Hell, all of the 9 films are (I’m trusting Nebraska is since I haven’t seen it yet).

  14. And it’s how we elect politicians.

    I don’t know if you are in the US, but we don’t have plurality voting. First we have the electoral college, which is each state elects (a somewhat proportional number of) representatives to cast votes for president. Even if we were to eliminate the electoral college, we still wouldn’t have a plural voting system. For a full year prior to the general election, the parties have elections to elect who to nominate for candidate. Each parties’ candidates travel around the country to get support through votes. Then after a lot of backstabbing, name calling, and mudslinging, one by one the candidates drop off, kinda like The Hunger Games of Politics. Only then does your candidate win. But then they go to round two—the general election where they compete with other winners, kinda like The Hunger Games of Politics: Catching Fire.

    If it was true plurality, we would have one election to decide them all. The 2008 election (last time we had both parties engaging in an active primary season) would have been Obama, Clinton, Romney, and McCain (and others). Would Obama still have won? If Clinton would have deeply cut into Obama’s numbers, could McCain have taken it (provided Romney didn’t cut too deep)? Obama took 53% to McCain’s 46%. Assuming these voters would vote along party lines if they were to cast their votes for Obama, Clinton, McCain, and Romney, Obama could have gotten 29% to Clinton’s 24%, with McCain 31% to Romney’s 15%. Under that unlikely scenario using the plural voting structure, McCain would have won even though less that one third of the population voted for him. That wouldn’t say much for his mandate. Our primary elections are there to ensure that our general election winner gets a majority* of votes.

    The US voting system is not a IRV system, but it is most definitely not a whoever-gets-the-most-votes-wins system. It just looks that way.

    *-no comment

  15. I support preferential voting for political elections, as it produces a more fair representation of precisely what the public wants and needs within that system. But in an artistic context, it only weeds out the interesting, divisive choices and gives a boost to the middle-brow, middle-of-the-road, middling choices like The Artist and Argo.

  16. But in an artistic context, it only weeds out the interesting, divisive choices and gives a boost to the middle-brow, middle-of-the-road, middling choices like The Artist and Argo.

    I agree with this sentiment, and it’s a widely held one. Still, it does seem to imply that a film doesn’t need to have passionate support to win BP under IRV. On the contrary, a winner absolutely still needs those #1s – no film can win by being everyone’s 2nd choice. It needs love – just a broader base of it.

    IRV itself doesn’t create a new bias towards safer, middle-brow films – it merely acts as a more representative depiction of overall preferences. Again, most of the time the IRV winner and the plurality winner end up being the same. So as much as we like to deride The King’s Speech, The Artist, and Argo as being safe choices, they obviously had a lot of passionate support among AMPAS voters. In other words, the real issue that leads to the vanilla choices isn’t the voting system – it’s the makeup of the Academy itself.

  17. Ya know, it really does help to have a visual representation.
    But I totally alwayz understooood it anywayz.

  18. I have an issue with how this video ended. It gives the impression that the number 3 film from the first round wins in the end. While this is a mathematical possibility, it is an unlikely one.

    Yeah, that bothered me too Dr. Rob, for exactly the reasons you mentioned. But of course Pond was merely giving a basic primer on how IRV works, and he’s right – it’s much more effective to see a visual representation of the vote stacks and how their shuffled and redistributed after each round than reading about it. We wonks can fill in the details and give a more precise picture of what goes on later. :-)

  19. Here’s a thing about the preferential ballot example Pond gives where the movie with fewer #1 votes than the other ends up winning. Of course, that can happen, that’s the whole point of preferential ballot.

    But, do we think it’s happened since they started preferential? I honestly don’t.

    Not only was a movie like the artist and Argo and kings speech higher on most people’s ballots, they were also the most popular those years.

    As much as I would like to think zero dark thirty or the social network got more number one votes but then lost at the redistribution stage, I doubt it.

    This year, by contrast, is one where I can easily see something like Gravity having the most number one votes but a more general crowd pleaser that doesn’t have the “oh it’s a tech movie” stigma, like American Hustle, sneak in in later rounds.

    No?

  20. ‘I have an issue with how this video ended. It gives the impression that the number 3 film from the first round wins in the end. While this is a mathematical possibility, it is an unlikely one.’

    Perhaps Steve was trying to discourage people from just voting for their favourite film and not filling in the rest of their ballot. If he had implied that the film in first place all through the process ended up as the winner, they might wonder why they ought to bother writing in numbers 1-9.

  21. If he had implied that the film in first place all through the process ended up as the winner, they might wonder why they ought to bother writing in numbers 1-9.

    Well, the obvious reason is that if their first choice gets eliminated and they didn’t fill slots 2-9, their vote will get thrown out entirely and they don’t get to directly affect the outcome, other than lowering the 50% + 1 vote level needed to win BP.

    Supporters of either Gravity, 12 Years, or Hustle can probably get away with listing just their first choice, but it’s definitely in the best interest of supporters of the other nominees to fill out the rest of their ballot if they want to ensue their vote counts in the end.

  22. That was a great demonstration. Makes one understand in an instant how Best Picture works.

    But it just occurred to me watching this that it is actually set up to PREVENT a picture which a lot of people may like from winning (if there are divisive viewpoints on a film). An example of this would likely be “The Tree of Life”, which a lot of people either loved or hated. If a small majority dislike a film, they can list it as No. 9 and wipe out its chances.

    So the winner is, basically, “vanilla”. Which means “Gravity” will win the Best Picture Oscar this year (even without a screenplay nomination).

  23. ObamaWins

    “So the winner is, basically, “vanilla”. Which means “Gravity” will win the Best Picture Oscar this year (even without a screenplay nomination).”
    Nah

  24. An example of this would likely be “The Tree of Life”, which a lot of people either loved or hated. If a small majority dislike a film, they can list it as No. 9 and wipe out its chances.

    Again, the #9 ranking really only becomes a factor in later rounds after films become eliminated from contention. If Tree of Life at any point had the lowest number of #1 votes in the first couple of rounds, those #9 rankings would absolutely have no role in its elimination. A small minority flipping the bird at Tree of Life by ranking it last doesn’t necessarily kill its chances – there has to be *quite* a bit of collusion among voters, advance knowledge of how each will rank films, in order to hurt ToL that way. And its really would be quite the effort for a significant number of voters, much less AMPAS voters, to pull off.

  25. Steve’s demo showing that the original #3 ends up winning is a subtle signal that he believes that Hustle will end up winning. Or at least it can win.

    But uncynically, he probably just wanted to show the notion that through this voting system, consensus can change the original ranking.

  26. If he had number 2 win in the end instead of number 1, then I would have been fine with it.

    Our own Awards Daily simulated Ballot had number 2 overtaking number one three out of the four past years. ZDT over Argo (2012), The Artist over The Tree of Life (2011), and Inglorious Basterds over Hurt Locker (2009). Only The Social Network (2010) remained number one throughout the process. In none of those scenarios was number 3 ever a threat.

    Again, it is mathematically possible for a number 3 to do it, but the planets really need to align for it to come to fruition.

  27. Rob, it’s an example. Sometimes when giving examples of a complicated system like this it can be helpful to see the unlikely scenario to get a better understanding of the why and how. 9 times out of ten it might end up being the film with the most number one votes anyway, but this demonstration better illustrates how exactly this voting system can play out.

  28. Profile photo of Sasha Stone

    Supporters of either Gravity, 12 Years, or Hustle can probably get away with listing just their first choice, but it’s definitely in the best interest of supporters of the other nominees to fill out the rest of their ballot if they want to ensue their vote counts in the end.

    That’s really the key in voters understanding the preferential vote.

  29. Profile photo of Sasha Stone

    It needs love – just a broader base of it.

    It needs lots of number ones and if it isn’t number one it would be 2 or 3, etc. That fits all three of the frontrunners, lol. So in that case what would you think would make the difference? That the films continue to rank higher rather than lower on ballots? I know for a certainty Gravity will be ranked high. I assume 12 years will. Wolf is a number 1 movie if it’s anything. But it seems that American Hustle has fared better than Wolf at least in the bigger consensus voting so far, nominations wise. Nobody HATES Gravity. That gives it the edge.

  30. Rob Y, I’m American, so I’m very aware of the Electoral College.

    That’s why I didn’t write: ”And that’s how we elect the President.”

    I wrote ”politicians” to cover all the other races (i.e., mayor, representative, senator, governor). You win by getting the most votes.

  31. So in that case what would you think would make the difference? That the films continue to rank higher rather than lower on ballots?

    Pretty much. A BP winner under preferential needs lots of initial #1 love to win on their first ballot, and if it fails to do that, a higher than lower ranking (on average) on the rest of the ballots compared to the remaining films.

    This is by no means a perfect comparison, but in the Statsgasm BP nomination simulation back in January, 12 Years and Gravity not only led the field in #1s in the 500 “ballots” I looked at, but they were also among the top 4 overall in #2, #3, and #4 votes. Hustle was significantly behind in #2s and #3s.

    But it seems that American Hustle has fared better than Wolf at least in the bigger consensus voting so far, nominations wise.

    That’s probably correct. Although we don’t know which method the other branches use for voting on nominees (it’s either preferential or weighted ballot), both require a degree of consensus.

  32. Al Robinson

    Okay, correction to my own comments from last night.

    I watched the video again, and it seems like the only movies that can win are the 3 movies with the most #1 votes. In the end, all the other 6 will drop off. But I wonder what will be the 3rd most #1 votes, American Hustle or 12 Years a Slave.? I see Gravity and The Wolf of Wall Street getting the most number #1 votes.

  33. Al Robinson

    Heck, why don’t they just count the votes this way:

    #1 = 9 points
    #9 = 1 point
    etc.

    Why don’t they just add up all the points per movie, and solve it that way? (They can still throw out the ballots that only picked one movie.)

  34. Al Robinson

    I could already see the problem in my solution, but then again, they only need to figure out which movie wins. Not rank 1 – 9. I know that the movie that wins, will have the most #1 votes, #2 votes, #3 votes, etc.

  35. Al Robinson

    “I know that the movie that wins, will have the most #1 votes, #2 votes, #3 votes, etc.”

    Hell, I think even that’s incorrect. Gah!!

  36. “So as much as we like to deride The King’s Speech, The Artist, and Argo as being safe choices, they obviously had a lot of passionate support among AMPAS voters.”

    AND they had rave reviews behind them. Even if they were “safe” many people outside the Academy loved them as well and that ended up mirroring the decision for Best Picture as well.

  37. “I support preferential voting for political elections, as it produces a more fair representation of precisely what the public wants and needs within that system. But in an artistic context, it only weeds out the interesting, divisive choices and gives a boost to the middle-brow, middle-of-the-road, middling choices like The Artist and Argo.”

    Very interesting point.

  38. “Nobody HATES Gravity. That gives it the edge.”

    Totally agree with you Sasha on this.

    Before preferential voting, I would have thought Gravity at a grave disadvantage without the screenplay nomination.

    But I think the technical branches will definitely put Gravity over the top, listing it in their top 3. American Hustle and 12 Years a Slave may split the acting branch voting (the largest of AMPAS), while still placing Gravity in the top 3.

    So I’m fairly certain Gravity has this in the bag. (But heh, I’ve been terribly wrong before . . . I had predicted “Drive” to receive multiple nominations in 2012. Not even one was given.)

  39. This video was so great, I feel like I finally understand the system now:)

  40. WW,

    I was not trying to insult you. So if I came across that way, I apologize.

    My point is that we rarely have a true majority wins election, more so on a local level. There are so many types of elections that aren’t a majority wins: elections where a run off between the top two candidates if one does not get a majority, any general election vote where the candidates are selected via a primary election (which is a two step process), to some elections that are IRV (Oakland mayoral race leaps to mind here).

  41. Though I may be dead wrong, I really feel like American Hustle’s buzz may be/has been fading a little bit (in general, and with direct competitio with Wolf votes). But God only knows.

  42. Heck, why don’t they just count the votes this way:

    #1 = 9 points
    #9 = 1 point
    etc.

    Why don’t they just add up all the points per movie, and solve it that way? (They can still throw out the ballots that only picked one movie.)

    This is one form of weighted balloting, and this specific version of giving points based on rank is formally known as the Borda Count. It’s simpler than the preferential ballot/IRV, but it has its own share of pros and cons.

    In general, the Borda Count requires an even higher standard of consensus the the simple “majority rules” preferential ballot – its very possible one candidate could be the first choice on 50% + 1 ballots and *still* lose under Borda because of the weights it gives to lower ranked preferences. Additionally, Borda is far more susceptible to tactical voting then IRV, especially “burying” – when voters purposely downrank a certain film they consider a threat to their preferred choice in order to defeat it.

    Much more about the Borda Count can be found on its Wikipedia page: http://en.wikipedia.org/wiki/Borda_count

  43. Al Robinson

    Ah, okay. Thanks Marshall. :-)

    I guess, when I was making my first comments, I must’ve somewhat been thinking it this way. The Borda Count way.

  44. I like that video. The process is much clearer to me now. Who knew?

  45. JPNS Viewer

    I think if one democratically feels or believes that consensus is of importance then one is likely to prefer the so-called preferential balloting system regardless of one’s wish. [edited by myself]

    A no. 1 film (or no. 2, no. 3, etc.) on one’s own list might denote just, let’s say, a no. 6 (no. 7, no. 9 or so out of 9, 10, etc.) to another. And when there are a large number of people involved, and that different people have got different strokes, I could see the benefits of it whether or not I like the result in question (Oscar, etc.).

  46. Whichever way you cut it, I reiterate: “If a small majority dislike a film, they can list it as No. 9 and wipe out its chances.”

    That could explain why Brokeback Mountain lost to Crash.

  47. Profile photo of Ryan Adams

    kiefer, what you say is true but what you say is unlikely to the point of being almost inconceivable.

    You say “a small majority” can kill any movie by listing it last on their ballots. If 6000 people vote then the smallest majority possible would be 3001 voters. I can’t imagine 3001 Academy voters — half the entire Academy — would ever all dislike any movie that much. How would such a movie even be nominated?

    Naturally though, if a majority of any size were to despise a movie so much then that movie cannot win. No movie can win if the majority of voters hate it. That’s fair enough though, right?

    I would much rather envision an Academy in which Brokeback lost to Crash by 30 votes or 3 votes instead of an Academy in which 3000 voters all listed Brokeback as their least favorite movie. If I thought half the Academy total assholes I would not be giving any fucks about the Oscars.

  48. I really feel that Gravity is in the driver’s seat. The industry truly loves it. Behind closed doors, people who don’t want 12YAS to win, voters who think it wasn’t very well done, they can rank it low. But Gravity is the kind of film that no one has the heart to rank that low.

  49. So if Hustle is a strong #3 but still no better than 3, the million dollar question is where the Hustle votes go. Who knows? But if you’re inclined to vote for Hustle over Slave, especially Hustle as your number one, methinks you prefer Gravity to Slave too. Then again, Hustle is probably my favorite, but until the past month, I was a bit more in favor of Slave winning than Gravity.

  50. Whichever way you cut it, I reiterate: “If a small majority dislike a film, they can list it as No. 9 and wipe out its chances.”

    That could explain why Brokeback Mountain lost to Crash.

    The preferential ballot was not a factor when Brokeback lost to Crash.

  51. @Flores Who knows if it was a factor or not….can’t assume it wasn’t. It could have been. Unless they start releasing the results, we will NEVER know.

  52. Actually, we *do* know, because the preferential ballot was not used in voting for BP winners in 2005.

  53. Just going to clear something up: AMPAS has been using the preferential ballot to determine BP nominees since the 30’s. However, during the 5 nominee era of 1944-2008, they used the standard plurality first-past-the-post system when voting on the BP winner.

    So yes, the preferential ballot was absolutely not a factor on what I refer to as Black Sunday, 2006.

  54. You are right , I was thinking you were saying we knew the results.

  55. Claudiu Dobre

    “But Gravity is the kind of film that no one has the heart to rank that low.”

    That’s just an assumption, though… We have absolutely no evidence of this apart from the PGA tie, which is just as much evidence that 12 Years a Slave also isn’t the kind of film no one has the heart to rank low.

    And I don’t even think the assumption is particularly well-founded. Knowing the Academy’s tendencies to prefer actor-driven, well-written movies, I see no reason to assume there won’t be quite a few of them who dislike Gravity quite a bit. It’s not yet clear who the preferential system favors between Gravity and 12 Years a Slave, and it simply won’t be until the last award is given out on March 2nd.

    Other points can be debated (nominations, precursor wins), but on this we are almost completely in the dark (because of the PGA tie), save for some simulations like the one AD runs, which, however, is voted on by almost 0 Academy members. And we all know how vastly different the internet’s tastes are from the Academy’s – you can see it clearly just by looking at how strongly criticized the last few winners have been online – all chosen by AMPAS by way of the preferential ballot.

  56. Am still recommending this being crucial .com/blog/steve-phttp://www.awardsdaily.com/blog/steve-ponds-everything-you-need-to-know-about-oscar-ballot-voting/

Leave a Comment