Electoral college predictions and voodoo

Predicting who will win the election based on the Electoral College is a bit like predicting who will win the Super Bowl six months prior — there is a bit of guesswork involved because things could change dramatically by the final day, but, at the same time, there are statistics you can use to make your prediction as accurate as possible. Depending on the source, a prediction may be as scientifically perfect as possible or it may be complete voodoo.

I hate the Electoral College, but we’re stuck with it. That’s how we pick Presidents. Suck it up and deal. Let’s move on.

In previous years, I enjoyed using Electoral-Vote.com that takes the map and updates it daily based on the most recent polls. The problem with that approach is that it treats each poll separately, and sometimes they may vary wildly.

More accurate this year is Nate Silver’s 538 map. This website is run by a bunch of math nerds. They don’t just take the most recent poll; they take them all and average them together based on a number of factors including the previous accuracy of that particular pollster, whether it was a poll of all voters or likely voters, how old the poll is, and a bunch of other things I don’t completely understand because math.

According to Silver, Hillary has around an 80% chance of winning the election, which isn’t completely surprising. The problem is that this election has already broken all the rules. Silver had also predicted in the past (like every other “expert”*) that Trump would never be the nominee.


This map is from 270toWin.com and matches Nate Silver’s current prediction

Seriously, Trump’s campaign is a classic example of what not do to in a campaign. The whole thing has gone against everything I ever learned as a Political Science major, a campaign manager, a lobbyist, and a campaign worker. It goes against everything I ever taught when I was a Political Science professor. He’s done everything wrong.

Of course, that could also be why he’s only given a 20% chance of winning.

But hey, the conventions haven’t even happened yet. For all we know, the GOP will find a way to nominate someone else and then we’re back to square one. Hillary, after all, is popular only in relation to Trump. If they nominate someone else, that 80% chance of winning would drop quickly.

*including me