SubscribeHey, how well did those math gurus predict the Oscars?
Thank goodness I forgot all about those math seers out West who got gads of publicity last year when they published their study of Oscar 
predictions in the Journal of the Royal Statistical Society.
Iain Pardoe of the University of Oregon and Dean Simonton of the University of California had created a prediction system after studying Oscar nominations in four major categories — best picture, director, actor and actress. When they applied it retroactively to the previous 30 derbies, it scored a success rate of 70%.
If I had remembered to investigate their predix during this recent derby, I might have been tempted to swipe them. But, hallelujah, I stuck with Marion Cotillard. They not only got best actress wrong this year while scoring their usual 75%, but they also had Julie Christie waaaaaaaaaaaaaay out front. They gave Christie a statistical probability of 71.8% to win, compared with 10.7% for Cotillard, 9.5% for Ellen Page. READ MORE
(Photos: Lions Gate/ Picturehouse)
If you look at the chart for 2005, the predictions for BEST PICTURE were BROKEBACK MOUNTAIN with an astounding 91.1%, and CRASH only 2.7% (!!!). Even Munich had better odds than Crash with 3.2%... and yet, insider politics finally prevailed and Oscar’s reputation and credibility took an irreparable blow.
Posted by: RDL | April 03, 2008 at 01:36 PM
And your choice for Best Actress (whom you "stuck to") was the same woman you turned around 3 days later and described as "dangerous" because of her 2007 French TV interview, thus causing her 10 days of a smear campaign? Tom, Tom -- you need to get your priorities in order. Destroying someone's reputation is not a pretty sight. Writing for the LA TIMES bears a certain amount of professional responsibility.
Posted by: kitvancleave | April 02, 2008 at 08:50 PM