Let’s See How This Lame Excuse Polls

Republicans Couldn’t Predict Eric Cantor’s Loss – NYTimes.com

Yes, a poll is a useful tool for gaining insight and information, but it is only one arrow in the quiver. Without qualitative insight — talking with voters face to face to judge their mood, emotion, intensity and opinion — polls can be inconsequential, and occasionally wrong.

The Cantor campaign’s catastrophe is not without modern precedent, even if the size and scope were extreme. Anyone remember Al Gore winning Florida, John Kerry winning Ohio and, of course, President Mitt Romney?

The simple truth remains that one in 20 polls — by the simple rules of math — misses the mark. That’s why there is that small but seemingly invisible “health warning” at the end of every poll, about the 95 percent confidence level. Even if every scientific approach is applied perfectly, 5 percent of all polls will end up outside the margin of error. They are electoral exercises in Russian roulette. Live by the poll; die by the poll.

This is all heinoously self-serving bullshit, of course.  The poll guy is out beating strawpolls to death in his own defense.  See, I do remember Al Gore, John Kerry, and Mitt Romney.  What I don’t remember is any pollster giving them a 34 point lead a few days before the election, and then seeing them lose by 11.

That’s an error of 45 points, and there were none such in the elections he mentions.  Missing the final by two or three points (most of the polls he cites for these other guys were actually within a legitimate margin of error) is not a massive mistake.  Missing it by 45 points is.


Posted in Bullshit, Polls permalink

About Bill Quick

I am a small-l libertarian. My primary concern is to increase individual liberty as much as possible in the face of statist efforts to restrict it from both the right and the left. If I had to sum up my beliefs as concisely as possible, I would say, "Stay out of my wallet and my bedroom," "your liberty stops at my nose," and "don't tread on me." I will believe that things are taking a turn for the better in America when married gays are able to, and do, maintain large arsenals of automatic weapons, and tax collectors are, and do, not.


Let’s See How This Lame Excuse Polls — 4 Comments

  1. If 5% off is a 2 sigma error (should be better 21 out of 22 times) then 45% off in the same sample size would be expected to happen about once in a number about the same size as the number of atoms in the observable universe.

    He just found out what a black swan is.