First boiled frogs, now basic math. I hate to keep wondering whether the NYT Op-Ed page employs fact checkers, but it's impossible not to wonder after passages like this. From this morning's Gail Collins column:


Most people have never been to a caucus, even if their state happens to have them. In Washington, the caucuses last Saturday drew a little more than 1 percent of the registered voters.



Those wacky caucuses! But wait a minute....

As a former Seattle resident I recall that Washington state has six or seven million people. After investing 0.75 seconds in internet research time I see that a little over half of them, let's say 3.75 million, are registered to vote.

One percent of 3.75 million is 37,500 people. Now, let's think back to reports of those caucuses. All the stories talked about "record breaking turnout" and "unexpected crowds." Some 20,000 people had crammed in to a pre-caucus Obama rally in Seattle -- with thousands more outside, and presumably thousands of others in the rest of the state also supporting Obama, or Clinton, or McCain, or Huckabee. And among all of them, only 37,500 show up?

And... It turns out that four years ago, the Democrats alone had 100,000 people for their much less dramatic and consequential caucuses. By all reports, highly publicized on caucus day, at least twice as many turned out for the Democrats this year. But somehow, according to the Times, only one-sixth that many people showed up for both parties???

And... I hear from friends and local news reports that the Democratic caucuses in just one Seattle-area legislative district attracted 18,000 people. (This detail from a story with the typical headline, "Turnout Shatters Record.") So, that district accounted for half the total for both parties across the entire state????

Obviously something went wrong here. Let's say the Democrats had maybe 200,000 at their caucuses, and the Republicans mabe half that many. That would be 300,000 total. Not enough to legitimize what is in fact a wacky caucus system. Not enough to prove that people of every class and background were involved. But different by nearly an order of magnitude from what our paper of record reports, in a factoid that will no doubt be picked up and considered "true".

What's the explanation? (And, by the way, I wish that some other NYT columnist had committed this howler, since I am a fan of Gail Collins' columns.) Maybe the "too good to check" instinct when coming across a tantalizing statistic? I don't know. But if we're looking for job-creation opportunities in America, how about for common-sense checkers?

___
Update: Mystery may be solved! The number of precinct delegates chosen at the caucuses, who in turn vote for the state delegates to the national party convention, was in fact close to the magic 1% figure. An understandable mixup, perhaps -- unless you apply the "can this figure possibly be true??" common sense test.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.