There are a lot of things wrong with the horse-race meme as it is applied to presidential politics. (“Coming around the first turn, it’s Chump Change in the lead, with Doctor Strangelove coming up on the inside….”) We all understand, if we think about it for just a minute, that an election is not a horse race, and to describe it as if it is adds nothing to our comprehension of what is actually happening.
It accomplishes the same thing Ronald Reagan used to do, in his early days as a sportscaster, when he took the box scores of a faraway baseball game off the wire service and used them to imagine the game itself and broadcast a play-by-play description. No harm, no foul, we might say; although the broadcast was bogus, it was entertaining, made money for the broadcaster, and was, as they say, inspired by true events.
But what if the box scores were bogus?
The presidential horse race myth is not nearly so benign as the imagined ball game. For one thing, it lasts for two years out of every four, and is an enormous distraction from problems that are more serious and more immediate than who gets to ride on Air Force One in two years. The amounts of money involved to put on the fake-race show, and buy ads on it, is sufficiently gigantic to engender endless corruption. The reduction of the complexities of a presidential election to the mindless simplicity of a horse race stunts the growth of our Republic.
But wait, it may be much worse than that. What if the box scores upon which this giant fraud rests — the polls — are themselves fraudulent? The imaginary ball game was a lie, sure, but harmless because it was based on the truth, the box scores of a real game played by real players. We assume that whatever its faults, the horse-race meme is at least based on polls that tell us what the people are thinking and how their thinking is changing.
Maybe not. Have you noticed that the two oldest and most prestigious polling organizations in the country — Gallup and Pew Research — are not even participating in this year’s “horse-race” frenzy? They’re worried about their future — and what worries them about their future is their past. Responsible polling companies seem to have lost confidence in their own product.
Back in October, Gallup stunned the political-industrial complex by announcing it would not conduct any voter polling during the primary campaigns, and maybe not during the general election. Or maybe ever, if they cannot figure out why they have gotten so bad at doing what they have always done best. In 2012, Gallup’s final polls had Romney winning the presidency in a squeaker; Obama won going away. In 2010, Gallup had the Republicans sweeping Congress much more decisively than they did. And that’s the recent record of the best of the pollsters; the rest of the crowd has been even worse.
The fact is, political polling is no longer reliable.
To provide a useful picture of voter attitudes (which can change hours later), a poll must engage a large enough sample (about 1,000), must select them randomly (within universes, of, say, registered voters), and just be careful not to skew the answers with loaded questions (Republican candidates poll better when Republicans ask the questions, and vice versa).
What has crippled polling more than any other factor is the death of the land-line telephone. Once upon a time, every adult had one, its number told you where it was, and you could look it up in a public directory. Now, fewer than half of adults have one (or use it as their primary telephone). There are now more mobile phones in use in America than there are human beings; 60% of people under 45 have only a cell phone.
This is big trouble for pollsters. Automated random dialing of cell phone numbers is illegal, so polling cell phones is slower and more expensive. Land line polling is cheaper and easier, but increasingly no one bothers to answer. And in either case, when the person called realizes it’s a pollster, they often hang up. These and many other problems of modern polling mean that many polls are just flat wrong, maybe when they insist that Trump is leading or Carson is gaining.
When Ronald Reagan beguiled a summer afternoon by re-creating a baseball game, no apparent harm was done. When Wolf Blitzer calls a political horse race, assuring us that certifiable lunatics are leading contenders to be President of the United States — when they may not in fact be enjoying significant support — the harm could be incalculable.