September 06, 2012

Wherein I (Badly) Forecast The Election


Who here has heard the name Dan Brown before? Lots? Really? Ok who here has heard of someone named Dan Brown other than that crappy author? Well, I have! Thanks to his chanel, Pogobat, I now know of a great political website. It's called 270towin.com, and is unsurprisingly about the electoral college. It offers great maps that show past voting history and what the current outlook is. It also gives you the chance to play with an interactive map so you can try and forecast the elections yourself! Isn't political science fun?

Ok, so 'fun' is a relative term here. At any rate, I've spent the morning hanging out at 270towin. They have compiled all the polls taken this election season, which is great for figuring out which states are likely to vote which way. About half the states (24) states (including D.C.) had no polls done since 08/01/12), while 27 states did. Of those 27, only 10 had more than three polls done since August 1st.

I took the polling data from these states and averaged them. This should help to eliminate any margin of error errors, however, the smaller the sample size the greater the likelihood of error. I then took the averaged number and applied it to the 270towin interactive map to achieve my (bullshit) forecast of the 2012 presidential election! See? Told you political science is fun!

Ok, so, according to my analysis, Obama wins 305 - 221, with Iowa and Nevada still undecided (no data from one, even split from the other). Surprisingly, the swing states of Wisconsin, Michigan, Ohio, and Pennsylvania swung towards Obama, with Florida going to Romney.




So there we have it, one person's moderately reasonable prediction for the 2012 Presidential election. For anyone interested, check out PogoBat, C.G.P. Grey, Fallacy Files, and 270towin.

Share/Bookmark

No comments:

Post a Comment