Lost Majority

I'm currently reading The Lost Majority by Sean Trende.  I cheated and skimmed the conclusion first.  It seems that he isn't making any firm prediction on the next election.  He seems somewhat critical of the 60% rating that Barack Obama had as of the time of the book's writing.  He points out that this was only slightly less than at his peak popularity shortly after he had been elected.  However, what this fails to take into account is that there is much more uncertainty the earlier such a prediction is made.  We would expect that as new information comes to light as a president's term progresses, that it would either look like he was more likely to win or less likely to win.  That is the odds would tend to diverge from 50%.

Now it looks like Barack Obama's odds are about the same, and 60% odds aren't all that high.  What that would indicate is that out of every 10 times something like that was repeated, the outcome would come out in favor of the president 6 times.

By the way, if you are thinking about betting on election outcomes, you should try it here first with fake money.  I participate in this market and it convinces me that I shouldn't try it in real life. I lose money.  Nothing ventured nothing lost.  They are giving the president somewhat better odds.

In short despite being critical of the odds given by the market, he doesn't seem to be giving odds that are all that much different.  He seems to think that both the president and his Republican adversaries have done a poor job of getting large numbers of people to support them.  Now it seems more and more likely that Mitt Romney will win the Republican nomination.  At the time he wrote the book things were much less certain.

The overall theme of the book is that political realignment is rare.  People talk of a permanent majority for one party or the other, or predict the other party's demise, but this seldom happens.  Large coalitions tend to break up due to the differing views of the various members.  When a party loses, they tend to look for loosely held portions of the stronger party's coalition to win over to their side.

Over the history of the U.S., the Whig party has gone out of existence.  That would indicate that in any two year election cycle, there is a roughly one percent chance that one of the two major parties will collapse.  In that case, there was division between the North and the South over slavery.  Since this author has an advertised accuracy of 99% over the last five elections, this prediction should be par for the course.