Intrade is a futures market where people bet real money on the likelihood of real-world events, like political outcomes. It's eerily accurate, and I follow it obsessively to get a real gauge on the changing fortunes of political candidates.
Today, I wondered if it's possible to use a basic conditional probability formula to determine which Republican candidate is most likely to beat Obama IF they win the Republican nomination. In other words, which candidate is the most "electable."
The formula I used is the following:
In layman's terms, the "probability of A given B" is equal to the "probability of A and B" divided by the "probability of just B." So to apply this to Intrade, treat A and B as winning the general election, and B as just winning the nomination. So you take the probability that a candidate will win the general election and divide it by the probability that they will win the Republican nomination. Following this formula, here's what I came up with:
The results make sense in a way. Jon Huntsman is very competent, gaffe-proof, scandal-free, and has major crossover appeal. Ron Paul also has crossover appeal it seems; whenever I present quotes from Ron Paul to my liberal friends they generally respond favorably.
Gingrich vs. Romney is interesting. I think while Romney may have more crossover appeal, he is too much like Obama (thanks to the hinging of Romneycare to Obamacare), and so whatever appeal he might get, Obama will have already siphoned it.
Gingrinch, on the other hand, is polarizing, and so if Republican sentiment is more positive than Democratic sentiment, he will be the one to seize it during the general election.
Tidak ada komentar:
Posting Komentar