Thanks to Marshall Flores for tabulating the results.
EDIT: Our apologies for a tabulating snafu — we had six (count ’em, six) entrants with an impressive 22/24:
Arlene Brown
John Carter
Faizan Chhapra
Matt Dy
Matt Fischer
Jack Hengtgen
After both tiebreakers, Matt Fischer remains the overall winner. Again, our sincerest apologies for this mess up.
Please send an email to claim your prize.
(A somewhat longer list of) Runners up with 21/24:
judy anastos
Nick Boyd
Steven Brown
Joe C
Mauricio Calderon
Maura Carter
Nicholas Chung
Anthony Cidoni
Hugo Ernesto Collado
Lorenzo Cusano
Dan Dawson
Claudiu Cristian Dobre
David Elioff
Brenda Elioff
Sefa Emekli
Bill Fath
John Feasel
Zach Gilbert
Lisa Hengtgen
Joseph Hiegel
Lauren Huff
Jim Keller
Matt Lee
Graham Leverton
Dan Lodge
Eliseo Martinez
Christopher Martinez
Zachary Mayberry
Debby McMillin
Megan Miller
Chris Miller
Ralph Moscato
Aminadabe Nascimento
Eric Nehs
Chelsea Olsen
CM Ooi
Nikita Pavlov
Alvaro Perez
Jim Phipps
Isaac Richter
Maria Rivera
tara Roberts
Leroy Rogers
Kenn Ruby
Shannon S
Sandy Steudel
DANIEL TAN
Jorge Tenreiro
Jason Turer
iveta vo
Jeff York
Mendes cost me dearly and sound Editing. dont remember third wrong answer, had 22/24 on Gold Derby.
Updates to the all-industry Best Picture-predicting system
These were prompted by two things, mainly – my increasing mistrust of acting/SAG nomination stats (due to AFTRA, on the one hand, and the notion that, with 20 slots available, it simply does not happen enough that a serious contender for Best Picture is actually snubbed for acting – last two years notwithstanding -, so that these stats have simply not been properly tested often enough for one to be sure about their strength, which, in any case, the numbers alone reveal to be quite a bit shakier than that of most of the other major industry stats), in particular in terms of their being included in elimination rules, and the discovery I made this January that Braveheart was, in fact (pretty much beyond doubt, as far as I’m concerned), snubbed by the PGA (both Wikipedia and IMDb appear to be wrong in claiming that only the winner was announced – I’ve found multiple sources confirming that there were, in fact, 7 PGA nominees that year, and Braveheart was not one of them). Relevant links:
https://nighthawknews.wordpress.com/2011/11/27/the-history-of-the-academy-awards-best-picture-1995/
http://oscarsijmen.freehostia.com/PGA90.htm
https://www.awardsdaily.com/wiki/wiki/main-wiki/wiki/producers-guild-pga/
http://tr.wikipedia-on-ipfs.org/wiki/En_%C4%B0yi_Tiyatral_Film_Producers_Guild_of_America_%C3%96d%C3%BCl%C3%BC.html
I also wanted to strengthen my nominations ranking elimination rules, which took a hit the Green Book year, and just in general to no longer have any elimination rules with more than one exception. Therefore, I decided to do that, as well as to no longer use the SAG Ensemble snub as part of any elimination rules (I already didn’t really use acting snubs that way), not the two snub rule and not the WGA+1 rule. And, finally, out of necessity (due to the tightening of the nominations ranking elimination rules), I had to add one last elimination rule: the need to win the PGA, DGA or WGA. All of this also required a change in how I stack up contenders against one another… Again, none of these things had any influence on picking Parasite this year or not – it would have been the system’s pick regardless. Long story short, here is the final version of the system as of right now, after the updates:
1. Elimination rules:
– the two snub rule (PGA/DGA/WGA/ACE and Oscar directing/screenplay/editing nominations only: no Best Picture winner has ever been snubbed for more than one of these in different categories – so, not DGA and Oscar directing, for example -, with at least 5 nomination slots per category, except for Hamlet in 1949, for editing and screenplay at the Oscars, although that one I find to be suspect due to the only writing credit, according to IMDb, being one William Shakespeare, who I suspect could/would not be nominated for an Oscar; Hamlet was also snubbed by the DGA, but there were only 4 slots that year and one of them was even taken by a movie that competed in another year at the Oscars, so of course this cannot be taken very seriously, especially since Hamlet did receive a directing nomination at the Oscars);
– the WGA+1 rule (same guild and Oscar categories involved, except for the screenplay ones – no Best Picture winner has ever been snubbed for one of the remaining five and also lost the WGA while eligible, except for Green Book in 2019, which is another questionable exception, since that was literally the only time in the WGA’s entire history that no Best Picture-nominated movie won any of their categories, which makes one question whether any of the Best Picture nominees should be held accountable for losing the WGA that year at all… regardless, both this and the two snub rule remain over 98%, even with the exceptions counted);
– a mandatory win at either PGA, DGA or WGA (every Best Picture winner in the PGA era has managed at least one, and the last movie that didn’t was Out of Africa in 1986, which had no PGA Award it could win – nor did any of the ones before it, some of which also had no shot at winning the WGA, being ineligible);
– having, at most, 6 nominations fewer than the Oscar nominations leader, as well as no more than 6 movies with more Oscar nominations that year (both of these rules are on 100% all-time).
Acting and ensemble nominations are also on 100% when combined with a lot of the other nomination stats used above, but not with others and, in general, seem to lead to exceptions annoyingly cropping up out of the blue, and a higher number of exceptions in general. Hence, their exclusion.
2. Weakness count:
– for the surviving movies after the eliminations phase (or, if all Best Picture nominees are eliminated, for all of them), head-to-head comparisons will be performed to see which of them has performed better than all others in terms of key industry wins and nominations;
– comparisons will simply consist of determining which of the two movies performed better overall in one of the following six categories: PGA, DGA/Oscar directing, WGA/Oscar screenplay, ACE/Oscar editing, SAG ensemble/acting/Oscar acting;
– ACE only counts in terms of nominations; the ACE win is irrelevant (the correlation is just not there – Best Picture winners sometimes win an ACE Award, but they very often don’t and that win is just as often harmful to predictions as it is helpful);
– evidently, a WGA ineligibility is not counted as a weakness;
– the acting branch performance is calculated by adding a point for any win at SAG (regardless of how many) and subtracting a point for any snub, whether in ensemble or acting at either SAG or the Oscars (but not both – a double snub in equivalent categories, meaning WGA+screenplay or DGA+directing or ACE+editing or SAG+Oscar acting, is always counted as a single weakness);
– the movie that performs better than all others in these head-to-head matchups is picked as the system’s favorite – if there is a tie…
I shall of course consider whether the WGA ineligibility should nevertheless be counted as a half-weakness or not, were the need to make that decision to arise at some point (for the 31 years of the PGA era investigated so far, using this method, it does not).
3. Tiebreakers:
– more SAG wins in any category (ensemble plus any of the four acting categories, counted equally);
– more SAG nominations in any category;
– more snubs (in the categories for which snub deductions are made).
Further tiebreakers will be added (after studying the options) should a situation ever come up in which these three prove insufficient. So far, none have.
Some of the more interesting examples – to perhaps clarify a bit how these steps work in various situations:
2020: All but Parasite and Jojo Rabbit are eliminated (1917, due to its editing snub coupled with its WGA defeat, Once Upon a Time… in Hollywood and The Irishman due to failing to win PGA/DGA/WGA) and their matchup goes as follows:
– PGA is tied (both nominated and lost);
– DGA/directing goes to Parasite (due to Jojo Rabbit’s Oscar snub – both losing nominees at DGA);
– WGA is tied (both won);
– ACE is tied (both won);
– SAG/acting is tied (Parasite is snubbed for acting but wins ensemble, while Jojo Rabbit is nominated for both but wins neither – the Oscar acting snub for Parasite is a duplicate and is therefore not counted again);
– no further major Oscar snubs;
– Parasite wins by 1-0. (Note that, had I decided on an interpretation that awarded SAG/acting to Jojo Rabbit due to the extra Oscar acting snubs for Parasite, the latter still would have won the tiebreak round due to its SAG ensemble win vs. 0 SAG wins for Jojo.)
2019: All are eliminated (Green Book, due to its directing snub coupled with its WGA defeat, Roma, similarly, due to its editing snub coupled with its WGA defeat, BlacKkKlansman, Vice and The Favourite, due to failing to win PGA/DGA/WGA – and the others via the two snub rule). It’s fairly obvious why and how Green Book comfortably beats Black Panther, Bohemian Rhapsody, The Favourite and A Star is Born, given their many snubs and almost total lack of major guild wins (Malek’s win being the only one between the four of them), so I’m only going to give details for its remaining three matchups – it beats Vice as follows:
– PGA goes to Green Book (win vs. nomination);
– DGA/directing goes to Vice (due to Green Book’s Oscar snub – both were nominated for and lost the DGA);
– WGA is tied (both nominated and lost);
– ACE is tied (both nominated and lost);
– SAG goes to Green Book (both snubbed for ensemble, but Green Book won an acting award and Vice didn’t);
– no further major Oscar snubs;
– Green Book wins by 2-1.
And it beats BlacKkKlansman as follows:
– PGA goes to Green Book (win vs. nomination);
– DGA/directing goes to BlacKkKlansman (due to Green Book’s Oscar snub – both were nominated for and lost the DGA);
– WGA is tied (both nominated and lost);
– ACE is tied (both nominated and lost);
– SAG is tied (Green Book is snubbed for ensemble but wins an acting award, whereas BlacKkKlansman doesn’t, nor does it convert its ensemble nomination into a win);
– no further major Oscar snubs;
– since the score is 1-1, the first tiebreaker decides things in favor of Green Book (due to its extra SAG actin win).
And it beats Roma as follows:
– PGA goes to Green Book (win vs. nomination);
– DGA/directing goes to Roma (DGA win and Oscar nomination vs. DGA loss and Oscar snub);
– WGA is tied (both nominated and lost);
– ACE/editing goes to Green Book (both nominated and lost at ACE, but Roma was snubbed at the Oscars when Green Book wasn’t);
– SAG goes to Green Book (Roma is snubbed completely whereas Green Book wins an acting award);
– no further major Oscar snubs;
– Green Book wins by 3-1.
2018: All but The Shape of Water are eliminated (Three Billboards, due to failing to win PGA/DGA/WGA and Get Out due to being 9 nominations off the leader, The Shape of Water, 4 vs. 13). Count is not needed.
2017: All but Moonlight, La La Land and Arrival are eliminated. Moonlight beats Arrival as follows:
– PGA is tied (both nominated and lost);
– DGA is tied (both nominated and lost);
– WGA is tied (both won);
– ACE is tied (both nominated – Arrival won but, again, this is irrelevant);
– SAG/acting goes to Moonlight (ensemble nomination vs. snub, acting win vs. nomination and Oscar acting nomination vs. snub);
– no further major Oscar snubs;
– Moonlight wins by 1-0.
And it beats La La Land as follows:
– PGA goes to La La Land (win vs. nomination);
– DGA goes to La La Land (win vs. nomination);
– WGA goes to Moonlight (win vs. nomination);
– ACE is tied (both nominated – La La Land wins, but this is irrelevant to the system);
– SAG goes to Moonlight (both win an acting award but La La Land is snubbed for ensemble);
– no major Oscar snubs;
– since the score is 2-2 and the first tiebreaker does not decide (each won one SAG award), Moonlight’s extra ensemble nomination at SAG breaks the tie.
2016: All but Spotlight and The Revenant are eliminated (The Big Short, due to being 7 nominations off the leader, The Revenant, 5 vs. 12) and their matchup goes as follows:
– PGA is tied (both nominated, neither won);
– DGA goes to The Revenant (win vs. nomination);
– WGA/screenplay goes to Spotlight (WGA win and Oscar nomination vs. two snubs);
– ACE goes to The Revenant (nomination vs. snub);
– SAG goes to Spotlight (ensemble win and acting nomination vs. acting win but no ensemble nomination);
– no further major Oscar snubs;
– since the score is 2-2 and the first tiebreaker does not decide (each won one SAG award), Spotlight’s extra ensemble nomination at SAG breaks the tie.
2007: All but The Departed are eliminated (Little Miss Sunshine, by the two snub rule – Oscar directing and editing snubs). Count is not needed.
2006: All but Crash and Brokeback Mountain are eliminated and their matchup goes as follows:
– PGA goes to Brokeback Mountain (win vs. nomination);
– DGA goes to Brokeback Mountain (win vs. nomination);
– WGA is tied (both won);
– ACE/editing goes to Crash (Oscar snub for Brokeback Mountain);
– SAG goes to Crash (ensemble win vs. nomination, both nominated for acting but with no wins);
– no further major Oscar snubs;
– since the score is 2-2, the first tiebreaker decides things in favor of Crash (SAG ensemble win vs. no SAG wins for Brokeback Mountain).
2005: All but Million Dollar Baby, The Aviator and Sideways are eliminated. Million Dollar Baby beats Sideways as follows:
– PGA is tied (both nominated, neither won);
– DGA goes to Million Dollar Baby (win vs. nomination);
– WGA goes to Sideways (win vs. nomination);
– ACE/editing goes to Million Dollar Baby (Oscar snub for Sideways);
– SAG is tied (both nominated for ensemble as well as acting, Sideways won ensemble, Million Dollar Baby won acting awards);
– no further major Oscar snubs;
– Million Dollar Baby wins by 2-1.
And it beats The Aviator as follows:
– PGA goes to The Aviator (win vs. nomination);
– DGA goes to Million Dollar Baby (win vs. nomination);
– WGA is tied (both nominated, neither won);
– ACE is tied (both nominated – The Aviator won, but this is irrelevant);
– SAG is tied (both nominated for ensemble, both won acting awards);
– no major Oscar snubs;
– since the score is 1-1, the first tiebreaker decides things in favor of Million Dollar Baby (two SAG acting wins vs. just one for The Aviator).
2001: All but Gladiator and Crouching Tiger, Hidden Dragon are eliminated (Traffic, due to being 7 nominations off the leader, Gladiator, 5 vs. 12) and their matchup goes as follows:
– PGA goes to Gladiator (win vs. nomination);
– DGA goes to Crouching Tiger, Hidden Dragon (win vs. nomination);
– WGA goes to Crouching Tiger, Hidden Dragon (nomination vs. snub);
– SAG goes to Gladiator (ensemble and acting nominations vs. snubs for both);
– no major Oscar snubs;
– since the score is 2-2 and the first tiebreaker does not decide (neither won any SAG awards), Gladiator’s extra ensemble and acting nominations at SAG break the tie.
1999: All but Shakespeare in Love and Saving Private Ryan are eliminated and their matchup goes as follows:
– PGA goes to Saving Private Ryan (win vs. nomination);
– DGA goes to Saving Private Ryan (win vs. nomination);
– WGA goes to Shakespeare in Love (win vs. nomination);
– ACE is tied (both nominated – Saving Private Ryan won, but this is irrelevant);
– SAG goes to Shakespeare in Love (wins for both ensemble and acting vs. only nominations);
– no major Oscar snubs;
– since the score is 2-2, the first tiebreaker decides things in favor of Shakespeare in Love (multiple SAG wins vs. none for Saving Private Ryan).
1998: All but Titanic and L.A. Confidential are eliminated (As Good As It Gets, due to being 7 nominations off the leader, Titanic, 7 vs. 14) and their matchup goes as follows:
– PGA goes to Titanic (win vs. nomination);
– DGA goes to Titanic (win vs. nomination);
– WGA/screenplay goes to L.A. Confidential (WGA win and Oscar nomination vs. WGA nomination and Oscar snub);
– ACE is tied (both nominated);
– SAG is tied (both nominated for and lost ensemble, both won an acting award – they tied each other, in fact, in that category);
– no further major Oscar snubs;
– Titanic wins by 2-1.
1996: All but Braveheart (PGA-snubbed, with acting and ensemble snubs at SAG and the Oscars, but WGA winner, and no two snub rules apply) are eliminated (Apollo 13 due to the directing snub coupled with the WGA defeat, Sense and Sensibility due to the two snub rule – directing and editing Oscar snubs). Count is not needed – and would be misleading!
1993: All but Unforgiven and The Crying Game are eliminated and their matchup goes as follows:
– PGA goes to The Crying Game (win vs. nomination);
– DGA goes to Unforgiven (win vs. nomination);
– WGA goes to The Crying Game (win vs. nomination);
– ACE goes to Unforgiven (as The Crying Game was snubbed);
– there is no SAG;
– no major Oscar snubs – no differences there;
– since the score is 2-2 and the first two tiebreakers are unavailable, The Crying Game’s ACE snub (there are 5 slots), the only one for either, decides things in favor of Unforgiven.
1990: All but Driving Miss Daisy and Born on the Fourth of July are eliminated and their matchup goes as follows:
– PGA goes to Driving Miss Daisy (win vs. nomination);
– DGA/directing goes to Born on the Fourth of July (DGA win and Oscar nomination vs. two snubs);
– WGA goes to Driving Miss Daisy (win vs. nomination);
– ACE is not counted (there are only 3 nomination slots and a single category);
– there is no SAG;
– no further major Oscar snubs;
– Driving Miss Daisy wins by 2-1.
Any questions about any of this or anything I might have left out will, of course, not go unanswered… 🙂
Wow, thats absolutely AMAZING – thats a kind of trivia I am always looking for. Ben Zauzmer isn’t even close! I heavily rely on retrospective and stats by myself, so I would be really glad to discuss this topic further as I also prefer to use a some kind of hand-made statistical-behavioral model. Is it possible, sir?
Absolutely – always happy to discuss the stats… And I don’t think the mods would mind it if we discussed it here – Ryan, let us know if it’s somehow not O.K.! So, shoot – ask me anything! 🙂
Well, nice! Actually I wanted to share some tables/graphs to discuss and improve prediction strategy, so probably better to use more media-friendly social network, for instance this https://www.facebook.com/ivan.pupyshev.
However, there is a thing which doesn’t need visual presentation to discuss. This year I went 21/24 (minus Film Editing, Sound Editing and Director) and the most challenging upset for me was Bong’s win in directing. Despite some thoughts in favor of Bong, which I had prior to the ceremony (which are too vast to tell here), the only evident dominance Joon-ho had was in Local Critics’ Associations awards. Also, there is an acting thing – Ang Lee (Life of Pi) and Mel Gibson (Braveheart) are the only directors in 32 years who won Oscars without Acting/SAG nomination – was it the case? And all of that with SAG win led to the Oscar? Was there any countable precursors to predict Bong Joon-ho’s victory?
(sorry this comment got caught in the spam filter, Ivan)
No problems, shit happens – seems it was caused by Facebook link. Just hope Mr Dobre has received the message and we could continue the conversation.
Claudiu and I stay in touch directly so I’ll email him to be sure he sees your message.
(yes, external links will trigger the filter.)
That’s nice of you, Ryan, thanks!
I’ve sent you a friend request on Facebook, so we can do the tables thing and anything else you had in mind. I’ve got my fair share of tables as well. 🙂 I have weird sleeping patterns, all over the place, and always try to do too much during the day anyway, so we’ll probably have to work along the lines of you-write-I-answer-when-I-find-the-time, similarly to how I approach replies here at Awards Daily during Oscar season. But it should be O.K. – we’ve certainly got plenty of time until the next Oscars… 🙂
On directing, there was no precursor stat I’m aware of that made Bong the favorite but, like I said, the comparison between the main stat working for him vs. the one working for Mendes I think did – the stat that only one director had ever, in 85 or so years, since editing was introduced at the Oscars, won with both no editing and no acting nominations was working against Mendes, and that’s a 99% stat with a very big sample, as far as Oscar stats go, and based on all-Academy voters as well (two key branches). Meanwhile, the strongest stat working for him is much less convincing, at least to me, after giving it the thought I should have before the Oscars: the fact that he’d won the BAFTA, Globe and DGA, and tied for Critics Choice. But BAFTA has only been held before the Oscars since 2000 and there are some half-exceptions pre-Bong. Soderbergh lost all three of those against the same guy and won only the Critics Choice, then won the Oscar. Clearly, we were inches away from that happening this year as well, so this is something of a precedent. Even more so, Affleck winning all four but not getting nominated for the Oscar – I mean, we can’t know for sure he would have won. He clearly didn’t have the directing branch’s support and it’s not like Argo won a heap of Oscars that night anyway. It won 3, which were all expected. Could’ve easily lost director had he been in the lineup. So that’s another half-precedent. To me, that’s more than enough to make a stat based on only maybe 10 precedents (since only that many directors in the BAFTA-pre-Oscars era have won there, at the Globes and DGA, like Mendes did) look a lot less reliable than a 99% stat with a single exception over 85 years (and 70 years ago, no less). 10 precedents are just not enough to be sure of anything, and the stat is not really even on 100%, like I said. What do you think? I think it’s pretty clear which of these stats should be taken more seriously and which of the two should be considered the favorite. At best, Mendes is perhaps a very, very shaky, slight stats favorite in this situation. But, personally, I don’t really see the arguments for it. The fact that I was careless enough to not take the time to read the stats properly beforehand is beside the point. 🙂
“Also, there is an acting thing – Ang Lee (Life of Pi) and Mel Gibson
(Braveheart) are the only directors in 32 years who won Oscars without
Acting/SAG nomination – was it the case?”
Do you mean acting or SAG? Directing winners without acting nominations at the Oscars in the last 32 years:
– Life of Pi (which you mentioned)
– Slumdog Millionaire (won SAG ensemble and had an acting nomination there)
– The Lord of the Rings: The Return of the King (won SAG ensemble but had no acting nominations there)
– Braveheart (which you also mentioned)
So, yes, this (zero Oscar or SAG acting/ensemble nominations) was a stat also working against Mendes, and definitely not a bad one (I wasn’t aware of it before), especially given the Affleck thing and the fact that the DGA winner wasn’t nominated the Braveheart year either, which makes that situation weird as well – but the editing+acting stat I think was enough to make him the underdog, anyway, like I said…
Claudiu! You guys have our full support, endorsement, and encouragement to deep dive into any topic and follow any tangent that interests you.
I’m always lurking and learning from you.
Thanks! 🙂 Will do…
Stats update for Best Picture:
All stats that held due to 1917 losing to Parasite, specifically:
INDUSTRY:
– editing nomination stat (still only beaten by Birdman since 1981, despite having to fend off challenges from contenders as strong as Roma or 1917, and that’s just in the last two years);
– ACE nomination stat;
– SAG ensemble nomination stat (which isn’t doing so hot but obviously still matters – I still use it, in any case, and it helps, even though it’s only on 88% after Parasite’s win, 22/25);
– WGA+1 stat (in 1917’s case: no movie has won Best Picture after being snubbed by AMPAS – or ACE, for that matter – for editing and then also losing the WGA);
– two snub rule (in 1917’s case: no movie has ever won Best Picture after being snubbed for both acting and editing – 87/87, after this);
– zero SAG nominations stat (no Best Picture winner since Braveheart has not had at least one SAG nomination for either acting or ensemble);
– BAFTA screenplay nomination stat (the last 16 or so Best Picture winners have all been nominated for that).
OTHER:
– Golden Globes screenplay nomination stat (last 15);
– Critics Choice screenplay nomination stat (also last 15);
– Satellite Awards screenplay nomination stat (last 16);
– Detroit Film Critics Society Best Film nomination stat (last 13 – all of them);
– Washington DC Area Film Critics Association (last 12);
– and other minor critics awards stats…
ADDITIONALLY:
– a stat I discovered earlier this Oscar season (but never found the time to post before), which says that no Best Picture winner since A Beautiful Mind (last 18) has not won an award for either their director, writer(s) or one of their actors (including special awards – rarely needed) at NBR/NYFCC/LAFCA/NSFC (any of the four; A Beautiful Mind is, in fact, the only PGA-era winner to have failed to do this), as well as that no winner since Crash (last 14) has not won at least two different awards for picture, director, screenplay or acting (again, including special awards) from the same four groups (1917 only got mentioned for cinematography at NBR, nothing else, Jojo Rabbit was in the NBR Top 10, like 1917, but got no wins whatsoever, and neither did Joker or Ford v Ferrari);
– the stat that only one of the last (now) 17 Best Picture winners didn’t win screenplay at either the WGA, BAFTA, USC Scripter, Golden Globes or Critics Choice (The Shape of Water);
– the stat that (now) 13 of the last 14 Best Picture winners had at least an 85 Metascore (Green Book, of course, being the exception).
A few of the stats that held due to Parasite winning Best Picture (over the field):
– the stat that when a movie won my preferential simulation (which I’ve done for all preferential years except the ones when The Hurt Locker and The Artist won) that was neither ruled out by any elimination rules my official predicting system uses, nor too far behind in the weakness count (more on all of that in my next post), it won Best Picture at the Oscars as well (Parasite joins Birdman and Moonlight on that list);
– the stat that no movie to have come in second or tied for second in the same simulation has ever won Best Picture;
– the NBR-GG-DGA-WGA win stat (no movie has won Best Picture since Hamlet without winning Best Film at NBR, Best Picture in either category at the Globes, the DGA or one of the two WGA categories).
Some of the stats that held due to Jojo Rabbit not winning Best Picture (not already mentioned):
INDUSTRY:
– the directing nomination stat;
– the BAFTA Best Film nomination stat (the last 23 Best Picture winners apart from Million Dollar Baby all had it).
OTHER:
– the Critics Choice directing nomination stat (last 17);
– the Golden Globes directing nomination stat (last 14);
– the Satellite Awards directing nomination stat (last 14);
– the Golden Globes screenplay nomination stat (last 14);
– the stat that no Best Picture winner has not won at least one Critics Choice award for picture, director, screenplay or acting in the BFCA era;
– the stat that no Best Picture winner has not won at least one Golden Globe or Critics Choice award for picture, director or screenplay in the same period of time (and a few years beyond);
– and a bunch of less important ones…
All stats that broke due to Parasite winning Best Picture (over the field):
INDUSTRY:
– Oscar acting nomination stat (this would have also been broken by 1917, of course);
– SAG acting nomination stat (the last 15 Best Picture winners had at least one; also a problem for 1917);
– BAFTA acting nomination stat (last 18 besides Million Dollar Baby; also a problem for 1917); it can be noted that all of these are acting-related (and were partially made up for by the SAG Ensemble win), whereas the others had all sorts of industry stats to beat, showing weakness with multiple Academy and guild branches.
OTHER:
– Critics Choice acting nomination stat (last 10; also a problem for 1917);
– NBR Top 10 stat (again – The Shape of Water also broke this; this hit takes it under 90% over the last 16 years, so it will be borderline unusable for me in the future);
– St. Louis Film Critics Association Best Film nomination stat (12 of the last 13);
– and other minor critics awards stats…
Additionally:
– the late October top 3 predicted stat (1917 would have broken this as well) – this has been holding for so long it’s almost surreal to see it fail, even though it’s been pretty obvious it would for a while… but I guess it had to, eventually.
Again, the foreign film thing was not a stat. The only relevant precedents were, obviously, the foreign films nominated for Best Picture before (since what the stat aims to “prove” is that a nominated foreign film, which is what Parasite was, cannot win Best Picture). However, only two of them had done anywhere near as well as Parasite in the precursor stage (and Oscar nominations), Crouching Tiger, Hidden Dragon and Roma – for a total relevant sample size of 2. But, for argument’s sake, we can ignore that, and then we have 11 foreign language films previously nominated for Best Picture. Calculating their total random winning expectancy (while most of them were nominated alongside 4 other movies, three of them were facing between 7 and 9 opponents) we arrive at 1.94, so basically two expected wins out of a total 11 possible. Only two wins away from the total achieved (zero), prior to Parasite. Given how very small the sample size is and the fact that this is not a nomination stat (meaning the task at hand – winning – is quite a bit harder to achieve – since for nomination stats we are, by definition, discussing only those movies already strong enough to be nominated for whatever other award we’re using the stat in question to predict), I would say this was virtually unusable. Even before Parasite won.
Finally, a stat that held due to Once Upon a Time… in Hollywood not winning is, of course, the one about the Best Picture winner always having won either the PGA, the DGA or the WGA in the PGA era (and beyond, going all the way back to 1986).
Other observations on Best Picture (in fairly random order):
This was the first time my predicting system went fully against the PGA winner (unless we count Gravity, too) and was successful! (Of course, in its current form, it would have also gone against others – since the whole point is to come up with a set of rules that picks every single winner in the PGA era, and it alone, from the set of nominees, and then use that to predict future races -, but I’m talking about the system in its form on Oscar day of each of the years I’ve used it. I did also predict against the PGA unofficially the year of Spotlight, and got that right, but my system was going with The Big Short instead, at the time.) Therefore, even though this was a rather easy stats layout to read (the number and severity of the weaknesses 1917 and others showed was far too great, compared to Parasite’s), perhaps this was more of an achievement than I was thinking it would be… In any case, the WGA again showed why it’s still the most important Best Picture precursor (Aaron Reichwald is going to love reading this, if he’s still around), especially in years with unclear races. And why the PGA (which is now on 3 full exceptions and a half-exception in the 11 years of the preferential era) should not be treated as a particularly difficult precursor to beat just because it also uses the preferential system – I mean, even a foreign film beat it!… Speaking of which, my theory that changes in voter demographics and mentality should be reflected in precursor results, thus enabling one to focus on the stats alone and not need to worry about guessing whether or not this or other such factors will impact the Oscar outcome in unforeseeable ways is looking a bit stronger after this Parasite win. The voters were finally ready to pick a foreign film, and this was, as expected, foreshadowed in a lot of key places – at SAG, WGA and ACE. (And everywhere else within the industry, in terms of its getting the key nominations required.) It didn’t beat or break the stats – it simply followed one of the already well-established paths.
One last thing about the PGA: as I think I’ve mentioned before, that and BAFTA were the only places (of the major precursors) 1917 actually beat Parasite for picture, and one was an award voted on by producers only, who were always going to be far, far more likely to go for something like 1917 (which their history shows as well), and the same can be said for BAFTA, given the British connection. This was just far from enough proof 1917 would be a favorite over Parasite “on neutral ground”, so to speak, at the Oscars, in Best Picture. For directing it had beaten it several places. There was more evidence there. (But, again, the editing+acting snubs stat was, in hindsight, a bigger deal than the wins). The Phoenix Film Critics Society got Best Picture wrong again (they picked Joker). Even so, they’re doing better than the PGA in the preferential era – they only got The Hurt Locker, Moonlight and Parasite “wrong” (whereas the PGA, in addition to its three exceptions, also had a tie between 12 Years a Slave and Gravity). One place that did strongly foreshadow this Parasite win long before the real fight started was the AFI Awards, where, like The Artist, The King’s Speech and Roma (the only other feature films besides Harry Potter, which got it the same year as The Artist – and for the whole series, anyway) before it, Parasite got the Special Award that has now led to a Best Picture win 3/4 times (and DGA and Best Director wins the other time)…
I think that what wins BAFTA Best Film will only become relevant in the Best Picture race if they ever again decide to pick a movie they don’t think is the front runner (which is what people keep accusing the BFCA of doing – BAFTA is just as bad and, in terms of actually getting it “right”, is in fact worse) and which is also not a British production. They’ve picked either the PGA winner (The Aviator, 12 Years a Slave), the DGA winner (The Revenant, Roma), the winner of both (9 times) and/or a British movie (The Queen, Atonement, The King’s Speech – also a PGA+DGA winner, counted -, Three Billboards) every year but one since 2003, when The Pianist (which was a France-Germany-Poland-UK production, by the way – in any case, for the winner the year before, The Fellowship of the Ring, I can find no UK connection) won. The exception being Boyhood, which was still fairly widely expected to win the DGA at the time voting closed (I remember it well – a ton of people were still predicting a split at the Oscars at that point), because the DGA only announced that Birdman had won instead the day before the BAFTA ceremony. (Which was probably not the case in The Revenant and Roma’s years – there were eight days between when the DGA and BAFTA announced, those times, as I suspect is the case most years.) And Boyhood was the big Globes and Critics Choice winner, anyway, so that was hardly going against the favorite, regardless. Anyway, until BAFTA finally come up with an unexpected winner, even if they get Best Picture right, there simply will be no reason to assume it was anything but a coincidence. Even then, it will not be BAFTA that gets it right, it will be the PGA or the DGA. BAFTA will simply copy-paste and tell us nothing. They don’t even tell us anything when they pick a British winner that didn’t win the PGA or DGA – it’s not even an indication of weakness for the PGA/DGA winner, because a PGA and/or DGA winner has taken Best Picture every time that’s happened since 2003, anyway.
I was right about Jojo Rabbit not winning (especially the year just after Green Book) – I said more than once that I needed to see it to believe it. Not least of all because I never thought something with as silly-sounding a name as “Jojo Rabbit” would get handed the Academy’s biggest honour. (I couldn’t come up with any other silly-sounding Best Picture-winning titles when I checked the list, a few months ago.) Speaking of titles, again a short one triumphs – perhaps this wasn’t a non-factor in Once Upon a Time… in Hollywood’s ultimate demise. Apart from literary adaptations (which is obviously different), titles that long simply have not made it onto the list of Best Picture winners… I was also right when I said, early on (after it came out with an 85+ Metascore), that 1917 would do at least as well as The Revenant had. It followed almost the same trajectory, the key differences being that it didn’t lose the Critics Choice for director, it had WGA and Oscar screenplay nominations instead of ACE and Oscar editing nominations (arguably an improvement, especially since it also had the “excuse”), it had no SAG or Oscar acting nominations but it won the PGA (which The Revenant failed to do) to make up for it. I guess it could be argued it did a little worse since it lost Best Director (although it still matched The Revenant’s tally of 3 wins), but I had only actually been talking about its precursor run when I said this, not about how well it would do on Oscar night.
Speaking of excuses, yet again they don’t work out… The only time excuses work out in the Best Picture race is when the movie in question actually does enough in terms of wins and nominations elsewhere to render them unnecessary. 🙂 (Like Birdman or The Shape of Water did – and 1917 didn’t.) I also want to update the situation on the pre-PGA final announcement stats table I keep every year: after this Parasite win, the new “stat” is that, in the five years I’ve looked at, the Best Picture winner has always been in the top 5 in those stats rankings, and no more than 11 weighted points (and 9 unweighted points) off first place. (Parasite was third, 11 and 8 points off first, which was Once Upon a Time… in Hollywood.) Finally, I want to just say that this Oscar weekend went absolutely perfectly for me, starting with The Farewell’s brilliantly unexpected triumph at the Independent Spirit Awards (where Marriage Story, also one of my favorites of the year, did very well too) and ending with the big victory for the stats that was Parasite’s Best Picture win… I’ve been running really well in life in general since roughly the start of December (arguably even earlier than that, maybe September), so I kind of had a bit of extra confidence I would get Best Picture right this year, and it indeed panned out. That’s almost it for me for this Oscar season – one more post to make, the one about the updated form of the Best Picture-predicting system, then I’ll be off (apart from reading a few more articles and comments and replying here and there, which will probably be more or less done by the end of the week).
Damn it, I should have gone with my gut and picked Parasite for Best Picture, but I wasn’t brave enough to predict that precedent would be set. Oh well, if I’m going to be wrong, I’d rather be happy that I was wrong, and this time I was ecstatic to be wrong (my three misses were Picture, Director and Live Action Short Film)…
LOL I picked Parasite for best picture, won my Oscar night contest and came second here. If only I had realised Mendes would be robbed.
As always the Oscars are about politics and popularity. At least PGA, DGA, GG and BATFA got it right, hey?
But at least, as the AMPAS attempted to atone for Roma and #meToo, they chose a terrific film to award. Congratulations to the Parasite fans.
Parasite is head and shoulders above Green Book and other recent BP winners.
Congrats on winning your Oscar night contest! And you are right that Parasite is better than recent winners (though I personally love The Artist).
Thanks !
First I’ll do an update about Oscar contests and bets and such, as well as the stats in all other categories, then I’ll prepare the post about the stats that held and the (very few) that didn’t for Best Picture and so on and, finally, update those interested on the all-industry stats Best Picture-predicting system (not many updates have been made, and none of them played into the system picking Parasite this year anyway, but they will come into play in the future, no doubt).
So, my official stats predictions went 18/24 this year, which isn’t great, although it’s also not that bad, especially since two of those misses were shorts (where the stats, as I’ve said before – and as one would expect – are of the most unreliable kind) – I had Kitbull in animated short and Brotherhood in live action short. (Tariq Khan was right after all, when he said in a recent podcast that he was sure Brotherhood wasn’t winning.) The other categories missed were director (still regret not paying the proper amount of attention to the stats research in that category – it was the one I started with and I knew at that point that I had a ton of work left to do on the stats in all of the other categories, so I kind of rushed it, deciding way too quickly that the win stats would likely trump the snub stat, even though the latter was so, so strong – but more on that below), sound mixing (had Ford v Ferrari for both sounds – really wanted to pick a split, because of the stat about the BAFTA sound winner almost always winning at least one of the sound categories at the Oscars, but it turns out I probably would have split them wrong anyway, so I’m not entirely unhappy that I went with the safer option this time), makeup (the stats said something risky – Joker) and visual effects (the stats had The Irishman as a slight favorite over 1917, with the others far, far behind).
It was nice to get film editing right so easily (the stats were quite clear), even though others seemed to have big doubts about what to pick in that category, as well as production design, costumes, sound editing, score, documentary and animated feature, all of which were also categories with a lot of options, seemingly, if one paid no mind to stats. And, of course, picture, which had Parasite as a fairly clear (even if not large) favorite. My unofficial predictions (which, it should be said, are also based in large part on the stats, since they’re usually just me going with the second or third favorite according to the stats I use – which may well not be the complete set of stats available, given how little time I dedicate to categories outside of the big one each year – instead of the top favorite, for various reasons) went 21/24, the same score I got in the Awards Daily contest. I had 1917 in the two sound categories and visual effects, Nefta Football Club in live action short and Hair Love in animated short – those were the only differences compared to my stats predictions. I stuck to Joker in makeup, which felt like a mistake, but I was too lazy (and there was just too little time available – none, really, given how late I completed my research and how from there on it was a mad rush to get everything done that needed to be done before the ceremony) to make a correction post, so I didn’t… I did go with Bombshell there in the Awards Daily contest, though. (And in 2-3 others.)
As for Oscar bets, I made a profit of almost exactly 25% of my total investment. (Roughly $18 profit. I don’t bet a lot – I can’t really afford to take such risks, plus I don’t really want to.) Apart from last year, when I lost a very small amount (and percentage of the stake), I’ve always made a profit on Oscar bets. Goes to show (yet again) just how strong Oscar stats are… All of the profit this year was due to this one bet on Parasite to win both Best Picture and Best Director, which I made at the last moment because the odds were just far too appealing (17 to 1) – I also made one on Once Upon a Time… in Hollywood to get the same two awards, which gave even better odds (127 to 1). As I said elsewhere, I entered six Oscar contests and got one first place (which was actually second place, behind the site owner who, amazingly, finished first, one point ahead of me – categories were assigned various point values -, but was ineligible for the prize) for a $50 Amazon gift card. Strangely, that was, apart from one where I really went crazy and made all kinds of out there picks, the contest in which I got the fewest categories right (18/24)… But I had Parasite there for picture and, apart from director, I only missed techs and shorts, which ended up scoring me enough points. The standings are here:
https://funeratic.com/thorough-movie-reviews/predict-the-oscars?formsuccess=register
I did not, however, pick Parasite in picture at Awards Daily (went with insurances and intuitive predictions a lot in this one – and figured picture was too important for me to go with my intuitive prediction, which was still Parasite, instead of the insurance pick), which cost me at least a tie for first. I only missed picture, director and sound editing here (went with 1917 in the sounds).
A few notes on the other categories besides Best Picture and the stats situation for each:
– Picture, as I said, I will address in a separate post. The four acting categories, cinematography and international feature were all either locks or near-locks and everybody knew it (and the stats agreed), so there’s nothing to discuss there.
– Director was far from a lock (in fact, Mendes was probably not even the favorite, in hindsight), and this is very easy to prove: only 1/86 (now 1/87) Best Director winners had been snubbed in both editing and all acting categories before (the exception happening in 1950), for a 99% stat with a huge sample (as far as these things go), whereas a DGA+GG+BAFTA directing winner had lost the Oscar 1/8 times (if we don’t count the time Affleck won all of those for Argo and was snubbed at the Oscars) since BAFTA has been a precursor to the Oscars, for only an 88% stat with a much, much smaller sample – and the time the DGA+GG+BAFTA directing winner lost the Oscar was one of the only two times that director did not also win the Critics Choice (Ang Lee in 2001 – lost the Critics Choice to Soderbergh), which Mendes also kinda’ didn’t this year, at least not outright. Even factoring in that Bong only tied him at Critics Choice (as well as the BAFTA voting procedure changing in 2012 or 2013 or whenever it was, exactly), it’s actually quite clear which of these two stats is stronger, and it’s not the one pointing towards Mendes winning… (Even if we think it’s not clear… well, then it’s not clear, and Mendes and Bong look roughly equally likely to win.) It was really silly of me to rush the research and conclusions for this category – I blame the short season, honestly, because I really never had a moment to breathe, especially over the last 2-3 weeks. I could have made the greatest stats call I’d ever made (even better than Colman last year and such). The stats still made that call, though, which I’m quite happy about – I was just in too much of a hurry to read them properly. Lesson learned, hopefully…
– In the screenplay categories the stats agreed with most people’s predictions (and the eventual winners), so not much to discuss there either. Likewise for original score and documentary short. SAG+WGA wins resulting in a screenplay win at the Oscars each time goes to 12/12 – that’s the only observation worth making, probably.
– For film editing Ford v Ferrari was a clear favorite because a) BAFTA was already as good a predictor as ACE, if not better and b) all the rest had major snubs to contend with, including Parasite (no BAFTA nomination, no sound mixing nomination – the last 12 film editing Oscar winners had all had that one) and Jojo Rabbit (the sound mixing one plus no Critics Choice editing nomination and no Gold Derby Awards nomination in that category either – all Oscar winner have had it, since Gold Derby has been voting on awards).
– In production design, Parasite had no BAFTA nomination plus it was contemporary (those almost never win), and 1917 had no BAFTA costume design nomination (also an important one for production design winners at the Oscars, most of the time) and had won none of the ADG categories (which most Oscar winners here do), leaving Once Upon a Time… in Hollywood as the (again fairly clear) stats favorite.
– Costume design was trickier. Little Women’s CDG snub was bizarre (I honestly can find no better explanation than its being a late-breaker); it also had no production design nomination at the Oscars. However, that one’s not a very prohibitive stat at all, and Jojo Rabbit, whose CDG win is less predictive than BAFTA, in general, had been snubbed by the BFCA, by whom 10/10 Oscar winners in this category had at least been nominated. Once Upon a Time… in Hollywood was the more plausible alternative, but it hadn’t won anything, and 15/17 Oscar costume design winners (in the Gold Derby era) had won either the BAFTA, the Critics Choice or the Gold Derby Award in this category. Elizabeth: The Golden Age had won the Satellite Award for costumes and Fantastic Beasts and Where to Find Them had won the BAFTA for production design – Once didn’t win either of those. So, I decided that Little Women’s CDG snub was probably a timing fluke (it was also not a killer, there had been several exceptions in the past, even if not recently) and, since the production design nomination stat also has plenty of exceptions, the win stat seemed stronger. (Yes, I spent more time pondering the stats balance for Best Costume Design than for Best Director – I am that dumb…)
– A lot of the same stats were in play for both sound editing and sound mixing. The CAS nomination is important for both and 1917 (the BAFTA sound winner) didn’t have it, which is why my official prediction was Ford v Ferrari (the MPSE Effects/Foley and CAS winner), although the fact that the BAFTA sound winner had won at least one of the sound Oscars 11 of the last 12 years gave me pause, and I really wanted to predict some sort of split because of it. I probably would have picked the wrong split, though, because the CAS nomination stat was stronger for sound mixing.
– Song was fairly straightforward: all songs that had won both the Critics Choice and the Golden Globe and been nominated for the Oscar had prevailed, and Rocketman has now done just that itself. It also won the Gold Derby Award and the Satellite Award. True, it had no other Oscar nominations (although Egerton I’m sure was close), which is why Stand Up (which did, and wasn’t snubbed anywhere) and, I guess, even Into the Unknown, could also win, but that wasn’t enough reason to make it the favorite. The win stat was on 100% here and the snub stat, obviously, was nowhere near as strong as the aforementioned directing one.
– The reasons I went with Joker in makeup were a) the last seven Best Picture nominees to be included in this category had won it (although maybe Joker wouldn’t have made it with only three slots available, as had been the case until this year) and b) the last six winners of the Best Period and/or Character Make-Up guild award had also won the Oscar (including Suicide Squad). None of these were particularly strong stats, though (the first one has got several exceptions before this just-broken 7/7 streak, and the second is based on a very small sample), so Bombshell’s wins were maybe ultimately more important – I kinda’ rushed this one, too. There was just so little time!…
– In visual effects The Irishman and 1917 were way ahead of the rest, because of not just the VFXS supporting win and the BAFTA effects win, respectively, but also their screenplay, production design and cinematography nominations (none of the other three had even a single one of these, not even Critics Choice and Gold Derby effects winner Avengers: Endgame), which also correlate very well with the VFX Oscar win. The Irishman was a favorite due to 1917 not having either the VFXS supporting win or the VFXS nomination in the main category (16 of the last 17 Oscar winners had had one or the other). But it seems the BAFTA win should, perhaps, weigh more – unclear.
– Documentary was a lot easier to call, stats-wise. There has been no documentary winner not predominantly in the English language since at least 2006. Also, Honeyland had missed at BAFTA and ACE and wasn’t among the DocNYC mentions, and For Sama somehow didn’t make the Critics Choice Best Documentary lineup (even though there are at least 10 slots there – didn’t count) and also missed at both DGA and ACE. The Cave and The Edge of Democracy had similar or even worse snub issues, so American Factory, the DGA, LAFCA and DocNYC winner, was the clear favorite.
– Also fairly trivial was animated feature, where no winner ever has been snubbed by either BAFTA, BFCA, HFPA, Gold Derby, PGA, ACE or CAS. Klaus, Missing Link and I Lost My Body were each snubbed by at least four of those groups, and How to Train Your Dragon: The Hidden World was snubbed by BAFTA and ACE, which is more than enough anyway, so the easy pick was Critics Choice, Gold Derby, PGA, ACE and CAS winner Toy Story 4. The ACE winner in this category has won the Oscar every single time it’s been nominated. (I think it’s 10/10 now.)
– Some of the more useful stats that didn’t work out in the short categories this year were the longest title stat in live action short (all other stats held) and the IMDb rating/number of other wins and nominations stats in animated short. Again, compared to some of the stats one can use for the other categories, these are mediocre, at best, and I only use them when I’m not sure between a few shorts or if they all point away from a certain contender.
Suggestion for visual effects: concerning BAFTA’s effect, I think it’s a good basic rule that if BAFTA and the Oscar disagree, it’s because the Oscar scales down in terms of amount of visual effects
Yeah, makes sense – does seem to be a trend… I’m curious whether there’s a precursor results-based translation to this. Maybe something related to Oscar nominations, beyond what I’m already looking at… Feels like there should be something.
My 3 misses were because of 1917, Picture, Director, and Sound Editing. Would’ve had the first tie breaker, too with 6 wins for 1917.
HOLLA! That’s a record for me and I didn’t even need the top two to do it.
Ahhh… if only I hadn’t gone “insurance” in this contest and just picked Parasite for BP instead!… 🙂 Small margins… Oh well – got a prize elsewhere ($50 – one of the contests in which I did use my official prediction of Parasite), and at least I came very close here, so I know it can be done.
Congratulations to the winner, of course!
Congratulations!
But I got 22 out of 24 correct as well. I only missed Directing and Sound Editing. Can you please explain?
Me too actually! Haha. Missed same categories. Guess a tiebreaker is needed?
Matt, I added your name to the list of highest scores.
Awesome thanks!
My apologies, Jon. After reviewing your entry manually I can confirm you also received an impressive 22/24 (as did four other people). Your name is now up on the post.
However, Matt still remains the overall winner after both tiebreakers.
Gotcha. It was a very good year and happy Parasite won. Congrats, Matt!