Statsgasm: Initial Predictions for the 86th Academy Awards!

By Marshall Flores

Welcome to our special Oscar Nominations Day edition of Awards Daily’s Statsgasm. With the announcement of the nominees for the 86th Academy Awards a few hours ago, the final leg of yet another marathon Oscar season has commenced. If you weren’t awake to catch Chris Hemsworth and Academy president Cheryl Boone Issacs reveal the nominees, a full rundown can be found here.

It’s going to be a mad scramble between now and March 2nd, with the guilds announcing their winners, BAFTA weighing in, campaigns performing a full-court press on voters, Oscar pundits bickering, etc. For some, the home stretch of Oscar season is the epitome of the Armageddon one Peter Venkman once described as “human sacrifice, dogs and cats living together – mass hysteria!” Others view this phase of the game as when Willy Wonka observed Augustus Gloop while he was stuck in the fudge tube: “The suspense is terrible – I hope it lasts” (if you’re wondering on my own thoughts – I tend to oscillate between both of those extremes).

But enough with burying the lede. We’re here to take a look at the initial set of winners predictions made by AD’s forecasting models! If you haven’t been a regular reader of Statsgasm, I do recommend you review Episode 3 and Episode 5 in order to have a better idea of the underlying methodology behind AD’s prediction models.

Before revealing the predictions, let’s keep in mind that **initial** is the operative term here. Although AD’s models have made some final predictions in a few races, the vast majority of categories are very much up in the air, their respective models still awaiting the outcomes of key precursors. There are even a couple of categories in which there is simply not enough data for the models to make a prediction at this time.

I have assembled all 21 of our models’ predictions into a series of four graphs. Let’s take a look.

These graphs indicate the predicted winner in each category and its (adjusted) probability of winning; we can alternatively interpret this number as the model’s confidence in its prediction. Asterisks next to a nominee’s name indicate that this is the “locked-in” prediction of the model. As such, this prediction will not change, although the model’s confidence in its selection may still be updated.

We will use these same graphs when revealing the models’ slate of final predictions, which will come sometime during the final week of February. However, AD’s Contender Tracker will be continually updated to reflect our models’ evolving forecasts as we progress through this last phase of Oscar season, so be sure to keep a lookout!

Let the wild rumpus start! Happy predicting!

Twitter: @IPreferPi314
E-mail: marshall(dot)flores(at)gmail(dot)com