Is it November 9 yet? Nope.
Below is a plot of the two major presidential candidates probability of winning over the course of time. Clinton has plummeted in the past few days thanks in part to, and you can’t make this up, the FBI investigating Mark Weiner. Clinton is still a heavy favorite to win the election, however. (In fact, I guarantee it). This plot is based on state electoral markets on predictit.org.
Some of the state markets, however, I think are probably priced incorrectly. For instance, California isn’t at a dollar. We can all be sure that California is going Democrat. Likewise, Oklahoma isn’t going Democrat. There isn’t even a 3% chance of it. So I re-did the plot and left only toss up states with win probabilities that weren’t either 0 or 1. I gave Hawaii, Maryland, Vermont, California, Massachusetts, New York, Rhode Island, Delaware, Illinois, New Jersey, Connecticut, Oregon, Virginia, New Mexico, and Minnesota to Clinto with probability 1 and Nebraska, Wyoming, West Virginia, Oklahoma, Kentucky, Alabama, North Dakota, South Dakota, Arkansas, Kansas, Tennessee, Montana, Louisiana, Utah, Alaska, Mississippi, Texas, and South Carolina to Trump with probability 1 (though I think I’m being generous here). All other win probabilities were based on the state markets. That plot looks like this:
Clinton’s chances of winning are only improved by about half a percent by modeling the election this way. The plot based entirely on the state electoral markets looks like this:
On a sort of unrelated note, it really is bizarre how we elect our president. There are like 3-5 states that actually matter and like 75% of people in those states have already made up their mind which party they will vote for before we even know who the candidates are. So the president is really elected by a small group of people from a small group of states to represent all Americans. Very bizarre.