File this post under “confirmation bias” or “reverse halo effect.” Nate Silver is the world-famous “data scientist” whose claim to fame is his ability to predict elections by aggregating polling data. Last summer (July 20, 2015), Silver explained why Donald Trump had a <5% chance of winning the Republican nomination. Last fall (Nov. 23, 2015), Silver grudgingly conceded that Trump had a <20% chance of being nominated. This week (May 4, 2016), with the campaign all but over, Silver finally recognized that Trump will be the Republican nominee. So, what should we make of this spectacular failure on Silver’s part to predict the outcome of this contest? More importantly, why did it take Nate Silver so long for him to update his priors? (FYI: Via Zero Hedge, Tyler Durden (or Daniel Ivandjiiski) explains in this excellent post why Silver was unable to distinguish between the signal and the noise.)

h/t: mangodebango (via imgur)