Oscar Category Fraud, By the Numbers
Sunday, March 22, 2020 at 7:00PM by Eric Blume
A few days ago, my fellow TFE writer Claudio flagged a website www.screentimecentral.com, which tracks the screen time and percentage of all Oscar nominees and winners. Because Oscar category fraud has long been a heated debate here at the site, I thought it might be fun to play with the numbers on the site from the last 20 Oscars, to see what we might discover if, say, AMPAS instituted a time/percentage rule on who could be eligible for lead and supporting performances.
Now for those about to go crazy in the comments, settle down. This is just a fun game. We’re not suggesting there should be an implemented rule, but for the sake of stirring up a healthy, positive conversation, I’ve broken down some of the numbers, and it reveals a lot of interesting things...
One fun way to look at screen time and lead/supporting categories is to go from the percentage of screen time the actors get. This reminds you how much extra plot and character development any film must devote away from even the most lead-driven pictures. For example, it’s interesting to note that even in films where the lead seems to be “almost every frame of the movie” there’s still a lot of screen percentage lost (Timothee Chalamet is only in 83% of Call Me By Your Name; Michael Fassbender in 81% of Steve Jobs; Isabelle Huppert in 84% of Elle; Reese Witherspoon in 88% of Wild, etc.).
Therefore, it feels not right to split the leading and supporting categories to 50% over or 50% under the percentage of screen time. So I’ve taken a look to see how the categories over the last twenty years might have looked if to be eligible for Best Actor/Actress, you had to be in over 40% of the film’s running time, and to be eligible for Best Supporting Actor/Supporting Actress, you had to appear in less than 40% of the film’s running time.
BEST ACTOR
We have eight nominees since 2000 who appeared in less than 40% of the film’s running time (percentages are rounded up or down from the site):
- Geoffrey Rush, in 30% of Quills
- Daniel Day-Lewis, in 29% of Gangs of New York
- Johnny Depp, in 34% of Pirates of the Caribbean
- Sean Penn, in 34% of Mystic River (the winner)
- Forest Whittaker, in 34% of The Last King of Scotland (the winner)
- Morgan Freeman, in 35% of Invictus
- Hugh Jackman, in 37% of Les Miserables
- Steve Carell, in 39% of Foxcatcher
Excepting Rush and Day-Lewis, these numbers are very close to 40%, and they all do feel like lead performances. Since Penn and Whittaker’s victories, no winner has come close to so low a percentage number (the closest were Day-Lewis for Lincoln, in 51%; and Jean Dujardin for The Artist, in 58%). But nobody here feels like they should have been nominated in the supporting category. Would you agree?
BEST ACTRESS
We’d have fifteen nominees who appeared in less than 40% of their film’s running times:
- Ellen Burstyn, in 35% of Requiem for a Dream
- Joan Allen, in 36% of The Contender
- Sissy Spacek, in 30% of In the Bedroom
- Nicole Kidman, in 21% of The Hours (the winner)
- Samantha Morton, in 29% of In America
- Reese Witherspoon, in 34% of Walk the Line (the winner)
- Meryl Streep, in 26% of The Devil Wears Prada
- Kate Winslet, in 33% of Little Children
- Kate Winslet, in 34% of The Reader
- Helen Mirren, in 34% of The Last Station
- Viola Davis, in 32% of The Help
- Rooney Mara, in 39% of The Girl with the Dragon Tattoo
- Jessica Chastain, in 37% of Zero Dark Thirty
- Amy Adams, in 36% of American Hustle
- Rosamund Pike, in 38% of Gone Girl
The two truly surprising numbers for me here were Spacek and Streep, because their performances are so vibrant and powerful in those films, and they feel like they’re in much more of their respective pictures. Much press was written about how the three actresses in The Hours and how they would be submitted (Julianne Moore, in the supporting category, was in 24% of the movie, four minutes more than lead Nicole). While winners Kidman and Witherspoon had relatively little screen time, the three winners who had the most screen time are Natalie Portman, in 83% of Black Swan; Julianne Moore, in 80% of Still Alice; and Julia Roberts, in 73% of Erin Brockovich). Having even more screen time are nominees Charlotte Rampling, in 91% of 45 Years, and Marion Cotillard, in 89% of Two Days, One Night…which seems fitting as they are both towering pieces of acting where the entire picture works or doesn’t on their shoulders.
For the supporting categories, we go the other direction.
SUPPORTING ACTOR
There are twelve Supporting Actor nominees who appear in MORE than 40% of their movie:
- Jim Broadbent, in 43% of Iris (the winner)
- Ethan Hawke, in 61% of Training Day
- Thomas Haden Church, in 48% of Sideways
- Jamie Foxx, in 54% of Collateral
- Jake Gyllenhaal, in 43% of Brokeback Mountain
- Djimon Hounsou, in 41% of Blood Diamond
- Woody Harrelson, in 48% of The Messenger
- Christian Bale, in 44% of The Fighter (the winner)
- Phillip Seymour Hoffman, in 47% of The Master
- Christoph Waltz, in 40% of Django Unchained (the winner)
- Sylvester Stallone, in 44% of Creed
- Mahershala Ali, in 52% of Green Book (the winner)
Here’s where it gets arguable. The three biggest percentages (Hawke, Foxx, and Ali) do feel like co-lead performances. But most of the others feel right in this category, most of them having a lead actor or actress whose story they are clearly supporting while still having their own. While Christian Bale is the winner with the most screen time, winners with the least screen time are Alan Arkin, in 14% of Little Miss Sunshine; and Benicio del Toro and Mark Rylance, each in 17% of Traffic and Bridge of Spies, respectively). Del Toro’s number in particular is interesting, as he won Best Actor at the SAG Awards that year! There are quite a few gentlemen nominated for less than 10% total screen time in this category, most notably Sam Elliot with less than 7% of dazzling time for A Star is Born.
BEST SUPPORTING ACTRESS
We have only six who appear in more than 40% of their movies:
- Cate Blanchett, in 56% of Notes on a Scandal
- Hailee Stanfield, in 55% of True Grit
- Julia Roberts, in 47% of August: Osage County
- Alicia Vikander, in 50% of The Danish Girl (the winner)
- Rooney Mara, in 60% of Carol
- Emma Stone, in 48% of The Favourite
These performances all feel like leads. Meryl was in less (44%) of her movie with Julia, and Jeff Bridges, nominated for Best Actor for True Grit, was in 45% of it. Cate was in Carol for 55%, less time than Mara. Two performances that made it just under the 40% gun were Jennifer Hudson, in 39.6% of Dreamgirls and Viola Davis, in 38.6% of Fences... if either had been in the background of one or two other scenes, they’d be listed here.
The Favourite is somewhat of a unique case, and its Oscar placements have been the talk of much debate. But looking at the numbers, Emma is in 48%; Olivia in 42%; and Rachel in 36%. It’s said that the Emma and Rachel deferred to Olivia on where she wanted to be placed, and when she went for Best Actress, they both submitted for Supporting.
This article included a lot of numbers, but I hope it was still some fun reading. Did any numbers surprise you?
Reader Comments (55)
A topic dear to my heart.
I always think, if you want to go lead, you're a lead. There's no fraud in that direction and screen time is irrelevant.
But 30% and under is a good benchmark for supporting.
Is the link working for anyone? It is not working for me.
My guidelines are this:
<25%: Probably supporting, outside of extreme (Lecter) outliers
25-40%: 50-50 chance, depending on narrative structure/content/context
41-60%: Probably lead, outside of Kate Winslet in Steve Jobs style outliers.
61%+: No question lead.
You'd probably want to make any rules contingent on multiple factors. I don't think you could be a lead if you not in the 2-3 largest performances for instances. (It would be fascinating how this breaks down for a film like 12 Angry Men).
Mike in canada is right. Everyone should me able to campaing for lead. There must be rules for supporting though
That Meryl Streep number blew my mind for The Devil Wears Prada. It feels like she is in way more of it than that lol
It's interesting that a whole lot of these are villain characters who sort of upstage their blander point of view characters who might technically be considered leads: Daniel Day-Lewis in GONY, Forest Whittaker in TLKOS, Steve Carell in Foxcatch, Meryl Streep in The Devil Wears Prada, and you could probably extend that to Tom Cruise in Collateral and Denzel Washington in Training Day, whose prominence resulted in high screentime supporting turns by Hawke and Foxx. And of course looking further into the past the ultimate example of this is probably Anthony Hopkins in The Silence of the Lambs.
On the page a lot of these really are category fraud but they don't feel like they are simply because of how "big" the performances feel.
Very interesting, and all pretty surprising. I would have thought there would be many more in the supporting category with higher numbers. Thanks for writing this up - I love this kind of thing.
I'll echo the sentiment that I'm surprised there weren't more. It's the glaring examples that are frustrating. I'm interested to hear Nathaniel weigh in.
I just saw Silence of the Lambs for the first time in a long time, and actually, he seemed like a supporting character to me this time. If he hadn't been played by such a big star, I'm guessing that would have been considered for the supporting category.
The number that surprises me most is Cate Blanchett in Notes on a Scandal, I don't remember to watched her practically in half of the movie.
Is interesting to watch that many of that examples specially in the lead categories can be considered more like a co-lead roles (not only by the screen time but for the plot by itself) and these numbers confirmes what i was suspect: that most actresses are nominated for co-lead roles more often than men and i could bet that in most of these cases the movie starring by the actress nominated was ignored in the major category
That means Oscar are not too interested in movies wich story was develop through a female lead.
heh, it seems we've collectively lucana-ed the boring plot lines of 'devil wears prada' where andy deals with her judgemental friends/dull romantic travails and wisely pushed miranda and emily and nigel to the forefront
[seriously, i had to google the name of anne hathaway's character]
This is always an interesting conversation, and having some actual data is really fascinating. My biggest surprise from who's featured is how little of Gone Girl Rosamund Pike is actually in. Given the structure of the story, I wouldn't have expected 50%, but under 40% is really something. And if nearly half of Lincoln doesn't feature Daniel Day-Lewis, then it's even more of a towering achievement than it's given credit for.
In all, I don't generally get too caught up on category fraud, especially if it really does come down to actors' decisions for how to be campaigned. I actually applaud those that decide to make that jump to lead when it makes sense, regardless of screen time (who is Nicole Kidman supporting in The Hours?). But I do hate when actors don't seem to be either smart enough about it (Richard Gere could've won Supporting Actor for Chicago, but foolishly went lead), or they seem to be underselling themselves (Viola had THE scene in Fences and would've won either category I think). Even with that being said, I do find some of these supporting nominations to be completely egregious, namely:
Julia Roberts
Jaime Foxx
Alicia Vikander
Christoph Waltz (and for a repeat performance at that!)
It is worth pointing out that the percentages are from total runtime, i.e. including credits (I asked the creator of the site about this a while ago on Twitter), which certainly helps explaining why barely any performance is in the 90+ % spectrum.
Great exercise: Rank all the male and female categories in a given year by screen time, and place the 5 most in lead and 5 least in supporting. Which winners would be competing against each other?? What new winners would emerge??
Kate Winslet is the winner for The Reader
I always feel Emma Stone is intended lead in The Favourite. When they make these royal dramas in Chinese (and there are a lot of these), the one who goes into royalty and rise up the ranks is ALWAYS the lead. Glad to see the numbers backed this up.
I just saw Silence of the Lambs for the first time in a long time, and actually, he seemed like a supporting character to me this time. If he hadn't been played by such a big star, I'm guessing that would have been considered for the supporting category.
Silence of the Lambs made him a major star afterwards. He was simply an actor's actor prior to it.
I think the biggest grey area is often around ensemble casts (eg. The Hours, The Favourite) when 3+ actors share almost equal screen-time. it’s really hard to define who should edge towards lead or whether to fill the whole supporting category from one film. There are many ways to argue each case eg. Which character provides the entry point/POV for the audience to navigate the narrative? Who goes on the biggest journey? Who is the primary connection around which other characters pivot? Who plays the most famous person (in telling real-life stories)? Who has the most screen time? Which performance lingers in the mind after the credits roll? So many questions that are particularly pertinent to large ensemble casts... It definitely gets contentious!
Inject this article into my veins!
God, I love stat heavy Oscar trivia
Sally W - I think that an "ensemble" cast (ie. everyone goes Supporting) needs 4+ Leads and multiple other characters - more than 3 Leads is too diluted (unless it's the special cases of Closer or Carnage where there are only 4 characters).
The studios responsible for all 6 of those Supporting Actress placements should be ashamed of themselves (and Roberts personally - nobody's placing a star of her calibre without her agreeing to it).
In the Supporting Actor list I'm mildly shocked by the percentage given to Woody Harrelson - I never for a second considered him a co-lead to Ben Foster, the film has little to no interest in his life outside of Foster's orbit... is he simply "there" in the background a lot of the time?
A very exciting topic! As much as I hate when a lead performance goes supporting I love the fact that sometimes a short and maybe supporting role on paper , makes us believe they’r lead Like Patricia neal , valerie perrine , anthony Hopkins and as you mentioned here meryl streep
I disagree with the idea that there is no harm in a supporting performance being campaigned in the lead category. Shining from the sidelines is a different challenge than carrying an entire picture; not a lesser challenge, just a different one.
If Louise Fletcher had been in the supporting actress category for 'Cuckoo's Nest,' where she should have been, perhaps we would have had one of the most deserving lead actress winners of all time in Isabelle Adjani for 'The Story of Adele H.'
Honestly, my takeaway is that category fraud is not as clear-cut as thought. Many of these supporting-but-should-have-been-lead performances, which "feel" like they were in every scene of the movie, were... not. A few of these cases (Ethan Hawke, Rooney Mara) are clear cut but many of them are more debatable than expected.
LOL why are you people surprised for Meryl Streep' percentage in The Devil wears Prada? Don't you remember the movie is actually about Anne Hathaway? Meryl Streep just happen to be the lead's motivation for her actions, goals and mistakes. That's why she feels present all the time.
Mahershala Ali is the supporting actor with the most screen time, not Bale.
MrW's point about credits being included is well taken. Curious how that would impact some of the supporting players stats.
guestguestguest, your takeaway was my takeaway after spending so much time putting this together. if examined just by the numbers, many of the supporting performances that we perceive as being category fraud really don't have that much screen time. i felt very strongly about rachel weisz in the favourite, as i thought she gave far and away the best performance of the nominated actresses, and indeed, she's in 36% of the movie...that feels like a supporting percentage to me, or at least not egregiously over.
conversely, i was a bit surprised about julia in August. i think she's far superior to meryl in that movie, yet she does seem to be supporting meryl. i knew julia was in a lot of the movie, but i didn't think more than meryl.
also agree with Val that gere should have put himself in supporting and likely could have won that year.
also agree with Mike M about louise fletcher (in only 17% of cuckoo), as isabelle adjani should have won (not just because she's in 65% of adele, but because the entire movie rests on her tiny brilliant shoulders).
this article stirred exactly the kind of debate i was hoping it would!
I don't think it has anything to do with minutes and percentages but this was a super fun read and the site is beautiful.
Pretty surprised that Heath Ledger, too, was only in 21.8% of The Dark Knight, when he feels like he dominates the entire movie. I suppose no one ever remembers much of the Hong Kong subplot or the extended Two Face epilogue.
Streep was far superior to Adams in J and J.
Colin: I mean, that Two-Face epilogue was...sigh. They...REALLY didn't need to knock down that domino in a single film. Because what resulted was...nothing. They clearly only had three film ideas, but burned through the third film in a half hour of the second.
LMAO Rooney Mara is in 60% of Carol. Seriously, how more egregious can you get.
But honestly, my personal pick for most egregious category fraud is Hailee Steinfeld in True Grit. I haven't seen it since theaters, but I saw it right after Oscar nominations and left thinking how on earth this girl was a supporting actress. The whole story revolved around HER and HER narrative!
Aaron: There are supporting actor nominations that get above 70%.
@MikeM
So glad to see what IMO is the most unfortunate case of category misplacement. Much was made of a scarcity of quality lead performances in 1975. Other candidates were Barbra Streisand, Marilyn Hasett, Diana Ross, Faye Dunaway. Fletcher probably would have won in either category, but Adjani was the NYFC winner and in reality gave the best performance by any actress that year. Time for the Criterion Collection to release that film!
I love this article - and that site you link to is wonderful; I've been obsessing over it for the past few weeks. The site helps us assess so much.
I also feel that there's no magic formula: screen time can be big but a role can still feel (and be) supporting; conversely, a role can be small but leading. (I'm afraid I think Louise Fletcher is a lead in One Flew Over the Cuckoo's Nest, despite her small screen time and percentage; the role is so crucial, so central whenever she's onscreen, and when she's not on screen, the characters - including McMurphy - are thinking about her and so are we. Adjani is spectacular too - as is Ann-Margret in Tommy; it's funny how 1975, criticised at the time and since for being an especially weak year for actresses, includes three such fantastic performances. Who knows, maybe Glenda Jackson and Carol Kane are also great - I haven't seen their nominated performances from that year.)
Aaron: Agreed about Hailee Steinfeld: it's not about screen time - but she is in True Grit for ten minutes more than Jeff Bridges!
/3rtful: Agreed that Hopkins was an actor's actor before The Silence of the Lambs, but because the film opened a whole year before the Oscars that rewarded it, Hopkins, in that year, became a big star, which means that by the time of his campaign, they had a big star on their books - hence the leading campaign.
The site has so many revelatory things, there are too many to cover in one post. Something that surprised me was all the way back in the 1935 Oscars, when three men from Mutiny on the Bounty were in the Best Actor category (the only time that's ever happened). This was before the supporting categories existed (they came along the year after) and I always thought that Franchot Tone got his Best Actor nomination for Mutiny for that reason: he would have been supporting a year later. But the site reveals that Tone is in the film for longer than Charles Laughton, who surely no one would doubt is a lead! Perhaps Tone just stands around in the background for much of that screen time; I'd have to watch it again.
It's a super interesting and fun thing to look into. Something to mention that is not highlighted in the website (but that Matthew, the author has told me in the past) is the methodology.
Matthew includes in the 100% of the screentime the credits. Opening, and closing. Basically from the moment you play start to when the movie stops playing. There's a lot to discuss about this approach given how other than very rare scenarios (Timothée) the actual screentime for any actor to be in the film is less than that. No lead actor, even if they're on every scene would manage the 100% (or even close to it). Which is why Rampling's at 91% is the highest percentage for any performance.
Matthew also considers any time the actor is seen or heard part of the screentime. It makes sense but it surely boosts the numbers of performances that might not have as much to do. I assume it's why Haden Church's is in almost 50% of his film.
Ultimately, my contribution would be screentimes are an imperfect way of viewing placements given how much variety there are in the way stories are told. I too was surprised at Del Toro in Traffic's screentime, and while I place him supporting I think it's not as unreasonable to call him the lead (his SAG placement was odd but not exactly baffling to me.
Due to the above, I think a deadline for each might make more sense. Instead of cutting off the distinction at 40% I would have limits. To be included in the supporting category you cannot appear in more than 45% of the film. If you do, you must be in lead. If you have less screentime you can choose. And even this is imperfect as it also throws Haden Church in lead where I think he doesn't belong, and lets Gyllenhaal in Brokeback slip back into supporting.
Simultaneously, to be in lead you must be in more than 25%? But I find category fraud from supporting to lead much less egregious.
I'm kind of shocked that Kate Winslet was in MORE of THE READER than LITTLE CHILDREN. The former seems like a supporting role and the later, clearly a lead. I guess 1% doesn't count for much, though.
I don't consider screen time to be as huge a factor in determining category placement as others do, but this is fun! I'm actually not surprised by Meryl's number. I love her in the movie and have no actual complaints about the nomination, but it always felt like a scene-stealing/showy supporting performance to me.
I remember cos i'm old enough to that Hopkins won a few supporting prizes early on in the 91 Oscar season,then the GG put him as a Lead where he won over Nolte and Beatty then TSOTL became a critics darling and once the Oscar noms were announced everyone excepted his placement as the character had by then embedded itself in the pop culture.
I'm surprised to see Jamie Foxx isn't in more of Collateral.
markgordonuk: Yes, Hopkins won numerous critics' awards - some in leading, some in supporting. I think the turning point for his campaign was the Globe nomination in the leading category - but Nolte won the Globe. Going into Oscar night, they seemed neck and neck.
Kevin Spacey is only in 40% of "American Beauty"?!
Ben Kingsley is only in half of "Gandhi"?!
Celeste Holm being in 58% of Come To The Stable is one example of needing to take things on a case to case basis. She basically quietly follows Loretta Young’s character the whole time and plays a game of tennis. Plus the after tennis reaction doesn’t quite hit as effectively as it should on paper.
Okay, I’d probably still put Weisz in lead, but being best in the category and having only 36% screen time? I’ll allow myself a day to feel she was robbed.
Can’t wait for Nathaniel’s takeaways/interesting opinion articles that will surely be coming in the next couple days!
This was fun to read. One thing I disagree on though is The Last King of Scotland. Whitaker always felt supporting to me, and McAvoy lead, and I found it super weird that they chose to campaign them the way they did. Even though McAvoy got snubbed for Best Supporting Actor, I'd love to see what his percentage was.
Wow, skimming through that site leads to quite a few surprises for me. I just rewatched The Fugitive recently and came out of it thinking Tommy Lee Jones was almost definitely a co-lead, but apparently he's only onscreen for 26% of the film. Just goes to show that big performances can definitely feel like they take up more of the movie than they actually do.
For me it's not about screen time in a movie, but how much I feel the character being there. And for Hopkins in TSOTL and Meryl in TDWP, they feel like they're there the whole movie (and most people simply associate them with the movie immediately)! So I'm not against thier category placement-for me they're co-leads with lesser screentime.
Thank you, Eric, for this exhaustive research. What a great read. Co-sign Edward’s take on Louise Fletcher being lead for the reasons he stated. It’s also one of
my favorite lead actress wins.
Thanks, brookesboy - I was worried I was the only one!
I feel justified feeling for almost twenty years that Spacek was Supporting in In the Bedroom. It's Wilkinson's movie.
What an article!
Regarding Heath Ledger and THE DARK KNIGHT: The Two-Face epilogue was inspired by the brilliant graphic novel THE LONG HALLOWEEN and would have been just fine being a movie on its own. In a post-AVENGERS world, as almost every supervillain appears in it (but I haven’t read it in forever and so of course I forget if Joker is in it, but he would have to be, right?
I was actually surprised that Julia is only in 73% of ERIN BROCKOVICH. I haven’t seen it in a while, but I remember it as a star vehicle.
Never thought about Gere going for Supporting in CHICAGO as he won the Golden Globe for lead, although it could explain John C. Reilly’s nomination.