2016 Oscar wrap up

What an interesting year for the Oscars. It was an upset in the =&0=& category – at least in terms of the math. Spotlight’s win surprised many and threw off our models. The movie about the Catholic church abuse scandal didn’t get either the Producers Guild Award or the Directors Guild Award. It also won the fewest Oscars of any Best Picture winner in 63 years.  Mad Max took home the most Oscars with 6 wins, The Revenant received 3 nods and Spotlight won 2 (Best Picture and Best Original Screenplay). We gave it an 8.8% probability for its  Best Director, Best Actor and Best Supporting Actress nominations, and its  94% fresh rating from audiences on Rotten Tomatoes. Some might say it won because it was everyone’s 2nd favorite movie. In case you know more about red carpet fashions than the behind the scenes accounting methods, the way Best Picture votes are tallied is interesting.  The Best Picture needs 50% of the 6,000 members’ votes to win. If a film doesn’t have enough votes to be declared the winner, the film with the fewest votes is eliminated. Here’s the interesting part – those ballots’ 2nd place votes become 1st place votes for those films, increasing their tallies. It’s something like political caucusing – eventually a film will reach 50% of votes and be declared the Best Picture. Adrienne! Adrienne! What happened to Rocky? Typically, the SAG winner is the most likely to win the =&1=& category at the Oscars. But this year’s Screen Actors Guild winner, Idris Elba (Beasts of No Nation), was overlooked by the #Oscarssowhite ballot.   Mark Rylance picked up momentum late with his British Academy of Film and Television win. In situations where there is a split between SAG, BAFTA and Golden Globe outcomes, the BAFTA winner is not typically the Oscar champ.  Back to the drawing board to update the models again. That’s it for the 88th Academy awards. Another year over, another year to look forward to what Alejandro Inarritu has for us.

Here we go. Our final predictions for the 2016 Oscars. We’ll check back in after the show.

The Revenant is expected to clean up – winning at least 3 of the 6 categories (Picture, Director, Actor). This will be the year Leo takes his rightful place at the throne & Inarritu claims modern directorial supremacy. Inarritu will move into the ranks of greats like Steven Spielberg & Oliver Stone as a two time winner. Inarritu could become the first director to win back-to-back Best Director awards since 1950 (Joseph L. Mankiewicz) and only the third director ever to do so.

Supporting Actor is the most controversial – Stallone is leading in the model, primarily because the SAG winner wasn’t nominated for an Oscar (Idris Elba – Beasts of No Nation). In fact, only Christian Bale & Mark Rylance were nominated for both the SAG & Oscar for Best Supporting Actor. Bookmakers still have Stallone as the favorite to win the award, which isn’t entirely off base, but I suspect this will be a category many could get wrong. (If I was pressed, I’d have to go with Christian Bale as my personal pic – played Michael Burry an investor with Asperger syndrome who has a glass eye in the Big Short)

read more

Objectivity of the New York Times

If a reader wants objective news coverage, shouldn’t it be available? Personally, if I want to get some objective news coverage, I go to the New York Times. It seems I’m not alone: this newspaper is considered an elite news source and has won 112 Pulitzer prizes, more than any other newspaper. It’s the most popular local metropolitan newspaper and has the most popular website of the nation’s newspapers. In this blog post, we try to explore just how objective the New York Times really is.

How can the objectivity of a document be measured?
One way is to compare the target document to other documents that are agreed to be subjective or objective beforehand. For example, movie reviews can be considered subjective. In a review, an author tries to give his viewpoint of the movie, and if the author is trustworthy, people take his viewpoint seriously. On the other hand, a movie synopsis just says what happened in a movie without offering an opinion or interpretation of the events.

read more

We did it: 6 for 6 Oscar picks!

We’re a little bleary-eyed this morning. But very proud. We bested last year’s pick percentage by one category—Best Director—one of the closer races of the night. (Insert nerdy, stat-geek fist pump).

Take a look at how our Oscar picks stacked up against Hollywood pundits this year.

Just when we thought the Best Picture Oscar® race couldn’t get tighter

How the Producers Guild Awards tie affects our pick

We hate to toot our own horn, but here goes: we correctly picked winners in 5 out of the big 6 Oscar categories last year. We did it using a complex predictive model (that’s “stat geek” speak for a giant, very mathy equation). So we’ve been feeling pret-ty good about our chances for a repeat performance in 2014.

Then, our data scientists got a big surprise (and these guys are kind of hard to surprise)

Double winners—12 Years a Slave and Gravity—were announced at the Producers Guild Awards (PGA) January 19th. This was surprising for two reasons: first, it’s the first tie in the 25-year history of the PGA. Secondly, and perhaps more importantly, the PGAs have historically been the number one predictor of the Best Picture Oscar winner.

read more

Farsite was right: 2013 Oscar prediction results

It worked.

We accurately predicted 5 out of 6 winners in the top categories for the 2013 Oscars: Best Picture, Best Actor, Best Actress, Best Supporting Actor, Best Supporting Actress and Best Director. And we’re back here in 2014 to do it all again.

We weren’t surprised, but we were pretty proud. See, here at Farsite, we use advanced analytics to solve our clients’ business questions. Being “stat geeks” our whole lives, we knew our statistical approach to Oscar predictions made sense.

What we didn’t expect was to blow anyone’s mind.

“I know there is no scientific way of predicting the Oscars®,” said Scott Feinberg, awards analyst for the Hollywood Reporter, when he spoke to the Wall Street Journal about Farsite’s capabilities. He went on to say that statistical models wouldn’t work because they didn’t account for the “underground intelligence” he and other pundits share.

read more