In doing some background reading for my recent post on gun ownership statistics, I came across a very interesting guest column on the New York Times’ “The Upshot” page. It was written by David Rothschild, an economist at Microsoft Research (PredictWise.com), and Sharad Goel, an assistant professor at Stanford University in the School of Engineering.
Because I try to stay clear of party politics here, I simply want to point out that not everyone was totally suckered by polls/pollsters into thinking Trump didn’t have a chance in hell of winning last fall.
To quote the conclusion to Rothschild’s and Goel’s column, published on October 5, 2016:
Over all, neither party has consistently outperformed the polls, but within a given year the errors are correlated between states. Even if a candidate appears to be trailing slightly in all the swing states, he or she could conceivably end up winning them all. This November, we would not be at all surprised to see Mrs. Clinton or Mr. Trump beat the state-by-state polling averages by about two percentage points. We just don’t know which one would do it.
BAM! Well done, sirs.