Oscarmetrics: Book Review

7th January 2020

Once again, it’s that time of year when the collective thoughts of the film world turn to just one thing, the studios beg for the recognition of self-appointed peerages and every day gossip outlets speculate which men will be winners and what women will wear at the ceremony – yes, it’s awards season.

If you couldn’t already tell, I’m not a fan of awards, from the Oscars down. With a growing emphasis on red carpet appearances and, more often than not, those who receive the most hype seem to win, it feels more and more like another piece of celebrity news and a lot of time spent speculating on what is an inevitable conclusion.

Is this just cynicism, though, or, if you looked back at every Oscar result from the start, could you find patterns that could save us all the time and predict the outcomes as soon as the nominees are announced? Could it even be expressed as a mathematical model?

Author Ben Zauzmer, who holds a degree in applied maths from Havard university, believes he has created just a model that can forecast the outcomes on the night – one that correctly predicted 20 out of 21 winners ahead of time in 2018. He shares the makings and practical workings of that model in his new book Oscarmetrics.

In the book, Zauzmer has collected every piece of Oscars data, the nominees and winners of every category throughout the awards’ history, and analysed it all to find tendencies in how the Academy has voted over time.

It’s not just the Oscars, either, he has done the same for the BAFTAs, Golden Globes, DGA, WGA, Grammys and various other guilds to see if there are any correlations between how they all hand out their prizes and the subsequent Oscar wins.

In doing so, he brings up some interesting trends – such as the director of the eventual best film winner wins their respective award 96% of the time – and litters his books with interesting piece of information, such as the most nominations for an individual without a win and the average age discrepancies in the acting categories.

An in-depth knowledge of maths is not needed to understand Oscarmetrics, though it can get over-technical to the point of confusion at times. Some may find themselves re-reading passages a second, third and fourth time before finally understanding what is being said.

Even so, Zauzmer proves himself both a very credible writer on both maths and cinema and is able to articulate his understanding of both very well – even doing well to explain how the two can be linked when it comes to awards.

That said, the book can also be unfocused and meandering at times – sometimes Zauzmer doesn’t even discuss the category mentioned at the top of the chapter. Probably more harmful to a book about statistics is that sometimes he uses incomplete or unreliable data, such as IMDb user scores and Rotten Tomatoes ratings, as part of proving his hypothesis.

This is forms part most grating in an over-long section of the book, towards the end where Zauzmer uses the IMDb scores as the basis to work out which film, year on year, should have won the Best Picture gong. His doing so will make you wish more and more that the proposed “Best Popular Film” category never comes to light.

Possibly the biggest missed opportunity of Oscarmetrics is that Zauzmer does not put his findings into action and predict the winners of an upcoming Oscar ceremony. There may have been some legal issue preventing that, but it does give the book something of an anticlimactic feel.

Despite containing elements that don’t sit comfortably, Oscarmetrics is full of fascinating information and is a book that can be read, understood and appreciated by anyone, regardless of their mathematical ability or level of understanding of film.

It’s a book that feels to the Oscars what Moneyball was to baseball and will add to the excitement of those who love awards season, while giving greater validation to those who hate it. They can now point to the hard data in Oscarmetrics to show it’s all a predictable, elitist, corporate fuss that will all be long forgotten by the time next year’s ceremony comes around.

Facebook

Google

Twitter

We hope you're enjoying BRWC. You should check us out on our social channels, subscribe to our newsletter, and tell your friends. BRWC is short for battleroyalewithcheese.