Tuesday, May 13, 2003
Read the Futures, Not the Polls
Don't believe the opinion polls. Bush I had higher ratings this time in his term from an Iraq war and lost badly. Approval ratings today are no measure of how people will vote a year and a half from now, and answering machines, cell phones, and decreased willingness to participate are making polls less reliable even with regard to current opinion. But there is a better measure.
The Iowa Electronics Market is running a real money futures market based on the 2004 election. Two, actually. One is the 2004 US Presidential Election Vote Share Market and the other is the 2004 Democratic Convention Market. These markets attract
gamblers investors who are looking at the future, not the present, and are willing to risk their money on the outcome. Empirical studies have shown that these markets are better predictors than polls of either the public or experts. (I once read the academic paper proving this off the web, and I will post the link as soon as I can find it again.)
Anyway go look at the current values of the 2004 Democratic Convention Market. Kerry has the best odds of winning of the four being tracked (Hillary Clinton, Gephart, Kerry, and Lieberman), at 30.6%. However, even he trails "rest of Field" at 38.4%. So we really don't know.
But the the more interesting market, and the one causing me to write this post, is the 2004 US Presidential Vote Share Market. What is fascinating about this market is that it not only tracks each potential Democrat's likely eventual share of the Presidential vote if he or she win the nomination, but also tracks Bush's eventual share depending on who he runs against.
This is what is crucial. Democrats can use this decide who to nominate, based on who is most likely to do job #1: Beat Bush. Unfortunately, the current values of this market do not, by themselves, tell us what we really want to know: What are the odds of Bush winning against any specific candidate? This requires someone to go through the results and divide Bush's value vs. each candidate by the total of that value and that Democrat's value. That is what I have done here, for values as of 7pm Pacific Time.
First, the news is not good. Currently, Bush has a higher value than all of his opponents, and his lowest value is against Clinton, who is not running. (Although that fact itself may explain why he is nearly equal to her.) I have expressed these chances as the percentage vote Bush the market predicts Bush will receive against the various candidates.
Clinton - 50.6%
Gephart - 54.2%
Kerry - 54.3%
Liebermen - 59.8%
Rest of Field - 53.7%
Parting observation - Bush does worse the more left-leaning the candidate is!