### POTUS 2008

About two years from now, the two U.S. political parties will be picking nominees. Who should they pick if they want to win the general election the following November? We can glean some insight from Tradesports.com and Bayes Theorem.

According to the betting over at Tradesports, here is the probability that the following individuals will be President of the United States after the 2008 election:

McCain 23.5

Clinton 19.9

Giuliani 8.5

Edwards 4.7

(These are the only four with active betting markets now.) From this information, however, you can't tell who would be the better candidate for a party to pick, because this probability reflects both the probability of being nominated and the probability of being elected if nominated. Fortunately, Tradesports also gives us the probability that each of these individuals will be nominated by his or her party:

McCain 39.5

Clinton 41.2

Giuliani 14.6

Edwards 8.2

Now apply Bayes Theorem. By dividing the first number by the second, we can obtain for each candidate the conditional probability--the probability that the person will win the general election if nominated. Here are the results:

McCain 59.5

Clinton 48.3

Giuliani 58.2

Edwards 57.3

Note the relative performance of Hillary Clinton and John Edwards. Although Clinton is more likely to end up President than Edwards is, Edwards is more likely to win the general election conditional on being nominated. At least that's what the market says.

According to the betting over at Tradesports, here is the probability that the following individuals will be President of the United States after the 2008 election:

McCain 23.5

Clinton 19.9

Giuliani 8.5

Edwards 4.7

(These are the only four with active betting markets now.) From this information, however, you can't tell who would be the better candidate for a party to pick, because this probability reflects both the probability of being nominated and the probability of being elected if nominated. Fortunately, Tradesports also gives us the probability that each of these individuals will be nominated by his or her party:

McCain 39.5

Clinton 41.2

Giuliani 14.6

Edwards 8.2

Now apply Bayes Theorem. By dividing the first number by the second, we can obtain for each candidate the conditional probability--the probability that the person will win the general election if nominated. Here are the results:

McCain 59.5

Clinton 48.3

Giuliani 58.2

Edwards 57.3

Note the relative performance of Hillary Clinton and John Edwards. Although Clinton is more likely to end up President than Edwards is, Edwards is more likely to win the general election conditional on being nominated. At least that's what the market says.

*Addendum*: For an introduction to the academic literature on these markets, see Wolfers and Zitzewitz.
<< Home