But who will fact-check the fact-checkers?

From a New York Times article:

(Another end-of-year fact-check, while we're at it: Mr. Trump claimed during the campaign that the homicide rate in his new home in Washington rose by 50 percent. In fact, it fell by 17 percent in 2016.)

Then-candidate Trump said, in July 2016:

Homicides last year increased by 17% in America's fifty largest cities. That's the largest increase in 25 years. In our nation's capital, killings have risen by 50 percent.

"Last year" as of July 2016—2015—homicides in D.C. did rise by 54 percent, from 105 in 2014 to 162 in 2015. They then fell in 2016 to 135, which is to say they were down 17 percent from 2015, but up 28 percent from 2014.

So, really, Trump correctly stated (not just "claimed") that the homicide rate in D.C. rose by 50 percent from 2014 to 2015. "In fact," it did rise by that much. But some of that decline turns out to have gotten reversed in 2016, so that the homicide rate in D.C. rose by only 28 percent from 2014 to 2016.

There's a lot to be said for not focusing too much on year-to-year changes in homicide statistics, which can be volatile. Even a rise over two years doesn't tell us that much, though it's troubling. And we should indeed remember that homicides and other crimes have generally declined sharply from their 1991 peak (though of course we want to be watchful for any reversal of the trend). If the argument is simply in favor of caution about reading too much into yearly statistics, I'm all for that.

But the New York Times "fact-check," it seems to me, goes far beyond that: It suggests that Trump got his facts wrong (he "claimed" one thing but "in fact" it was something else), and I think it misleads readers into missing the fact that, even counting the 2016 decline, the homicide still rose sharply from the reference year Trump was using—2014—to the present.

Thanks to Ethan Epstein (Hot Air) and InstaPundit for the pointer.

UPDATE: The New York Times article has been changed to read,

Another end-of-year reality check, while we're at it: Mr. Trump claimed during the campaign that the homicide rate in his new home in Washington rose by 50 percent, apparently citing the previous year's crime statistics. At the time, though, the rate in the city was already falling, and by year's end, it was down by 17 percent.

Well, yes, as of July 2016 (the time Trump was speaking), the rate in the city appeared to be down 9% from the 2015 numbers—but even that would have meant a rise of 40 percent from 2014 to the 2016 rate. And though the rate fell further in 2016, it still ended the year, as I mentioned, up 28 percent from the 2014 rate.