Monday, January 14, 2008


As you may have noticed, the political polls change with about the same regularity as the weather. However, let's admit it: you believe they tell you something of value, and so do I. For the most part, we should discard that belief.

Right now (Monday night), I'm hearing "from Michigan" that the polls (which ones?) are telling us that John McCain and Mitt Romney are running "neck-and-neck" (is there any other way for politicians to run?). Does that mean we'll have to stay up late tomorrow (Tuesday) to see if Mitt Romney elects to take "early retirement" from the race? Maybe.

I think McCain will finish first, perhaps well ahead of Romney. Mike Huckabee will finish third.

But the Michigan primary, like the previous one in New Hampshire, is impossible to predict. For one thing, Independents can vote in the primary -- Democratic or Republican. No one knows how the "Independents" (a much smaller percentage in MI than NH) will vote -- or even if they'll vote. Also, Michigan is a fully open primary, and Democrats apparently can vote in the Republican primary. The Daily Kos, an extreme leftist blog, is telling Michigan Democrats to vote for Mitt Romney -- the better to screw up the Republican results.

What about the Democratic candidates? Well, for arcane reasons the national Democrat Party has told the candidates not to campaign in Michigan. So, there are only three candidates on the ballot. One is Hillary Clinton, and the other two are the ever-popular Dennis Kucinich and Mike Gravel. What about Barack and John? They're not on the ballot.

What is on the Democratic ballot is an heretofore unknown candidate known as "Uncommitted." In other words, if you don't feel like voting for Hillary, Dennis, and Mike, you don't have to vote for anyone. If Mr. (or Ms.?) Uncommitted gets a lot of support, I'm sure the pundits will tell us what that means.

Given all these factors, don't ask me exactly what's going to happen in Michigan. Anyone who claims to know -- even if their name is Gallup, Zogby, or Rassmussen -- qualifies as an idiot.

In yesterday's (Sunday's) Pittsburgh Post-Gazette, there was an article about failures of the pollsters in New Hampshire. Writer Daniel Malloy says, "Pollsters argued that the polls weren't wrong [perish the thought!] -- there was just a change in the thinking of voters, especially women, who suddenly became eager to support an underdog, emotional Mrs. Clinton."

Of course, that's all bulldoodoo. When someone presents us with a poll that has a certain "margin of error" and the results far exceed the margin, then we need a new pollster. After all, polls aren't supposed to be a crap shoot, are they?

I'll let you in on a secret. I know precisely -- right down to the percentage points -- what's going to happen in Michigan. But since you don't believe me, I'm not going to tell you now. I will tell you tomorrow night.

Tomorrow, I'll write some more about the polls -- including Gallup's classic blunders in 2004, when they had Bush winning PA (he lost) and losing Ohio (he won).

No comments: