THE FALLACY OF PREDICTING DEMOCRATIC PRIMARIES

Somewhere or other, I came across this article saying that Howard Dean is gaining ground in South Carolina, a state where some write him off because he is viewed as more liberal than his primary opponents. Anyway, it got me wondering: Aside from the polled-to-death early states, like New Hampshire and Iowa, how is it possible to really get a feel for which partisan primaries are going to go which way?

If you superficially look at the Democratic primaries, you would assume that states like New Hampshire, Washington, California, etc, would favor more liberal candidates, because hey, those states are certainly more liberal than many of their counterparts, and would vote that way (Dean, Kerry, perhaps Kucinich). Conversely, states like South Carolina, Texas, and the like would seem to vote for more conservative candidates (Gephardt, Graham, Kerry). Right? Well, not necessarily. The states on the whole may lean that way, but that doesn't necessarily say anything about the political leanings of the Democratic population of a given state. As an example, Idaho and Montana's Democratic voters may be very liberal, but because they are vastly outregistered by Republicans, the state itself doesn't lean that way. It's completely possible that a "liberal" candidate could decisively win a primary in a state that is generally viewed as "conservative", and vice versa.

The wild-card, of course, lies in the states which have open primaries. The logical thinking would be that if Republicans and independents could cross over and vote in a Democratic primary, this would generally benefit more conservative candidates. Right? Again, not necessarily. The assumption I would make is that any voter who chooses to cross over and vote in an opposition party's primary is probably a sophisticated voter who is making a sophisticated choice. An example of this would be: Let's say the 2002 California GOP Gubernatorial primary were an open primary. In that case, I, as a Democrat, would have voted for the weaker potential opponent to Davis, Bill Simon, to defeat the more dangerous candidate, Richard Riordan. Of course, that wasn't even necessary, as the California GOP loves to slip on its own banana peels, and thus gave Simon the nomination all by themselves. (tbogg has more on their current idiocy, the effort to recall Davis)

So, bottom line, partisan primaries that haven't been polled to death are hard to predict. At least until Super Tuesday, at which point the media tries to cut costs on covering the campaign and anoint a winner.