I still remember 2008, when an obscure writer here began publishing unbelievable diaries with tons of data regarding Obama’s/ Clinton’s chances of winning the primaries and caucuses in the many states. He wrote under the name Poblano. That writer turned out to be the Beta version of Nate Silver (cool factoid that he started here!). In 2009 he used his knowledge and skills to launch his own site- fivethirtyeight.com. In the years since, his site has been hosted by various media companies and added data-journalism coverage of sports, economics, science, lifestyle, etc.
Say what you will about fiverthirtyeight.com, and Nate Silver himself, but my memory of 2016 is that his model was the only one that took Trump’s chances seriously. In fact, he was mercilessly attacked by other modelers like Huffpost , who at the time was doing their own modeling and gave Clinton something like 98%-99% chance of winning- and later had to apologize to Nate. I wish he had been wrong, but his common refrain was that Clinton had an edge but we were just one regular presidential year polling error (4% points) away from Trump winning the Electoral College while Clinton would win the popular vote. And, that Trump had some structural advantages in the Electoral College that Clinton had to overcome. My point is he knows this stuff/ it’s not a fly-by-night operation.
His model launched this morning and currently shows Biden with a 71% chance of winning to Trump’s 29%. “We simulate the election 40,000 times to see who wins most often. The sample of 100 outcomes below gives you a good idea of the range of scenarios our model thinks is possible.” Trump still retains some advantages in the EC(!), though Biden is currently and consistently doing better in public polling than Clinton at this point in the race. These numbers will be updated as new data comes in on an ongoing basis there- check it out and enjoy the amazing and varied quantification of every factor. Just wanted to provide this for the good of the order.
Update: H/T to NorthBronxDem for pointing out that in Nate’s discussion of the model, he says 71% is all about the uncertainty and distance from the finish line...And if the race were held today Biden would have a 93% chance of winning based on the fundamentals.
