|
With Oklahoma's smaller population and especially with its small voter pool, huge percentage swings can take place with relatively small shifts in the vote number. In the 2000 race 10,000 votes equaled one percentage point. The same kind of vote swing in a state like, for instance, California would barely register. Those kinds of shifts do happen, and they are amplified by the process of statistical extrapolation of poll samplings. This is one of many reasons polls in OK are notoriously inaccurate. Another is that polling companies with the necessary resources to conduct good polls rarely bother with OK.
Also, due in part to candidate disinterest in the state, exacerbated by "Good Ol' Boy" politics-as-usual, there are more undecideds, fewer people really worked up about the race who go all out to give a firm opinion. This has been the case with the Senate race for while, although people tend to be more firm on that than they were a couple months ago.
In any case, I've never seen a poll in OK be accurate with regard to the Presidential race, at least as a matter of percentage points. Yeah, it always goes Repug, but the spread never seems on the mark.
The latest poll for OK has Kerry at 28%. I simply do not believe that.
|