Reporting today on a batch of new post-convention poll results that seem to show a minor bump in support for John Kerry, Adam Nagourney of the New York Times writes that “before the [convention], a senior Bush political adviser, Matthew Dowd, distributed a memorandum arguing that based on history, Mr. Kerry should gain at least 15 points, though most independent pollsters called that projection unrealistic in such a tight race.”

It’s true that Dowd predicted a 15-point bounce, and that many independent observers disagreed. Most assumed, no doubt correctly, that Dowd was playing the well-worn game of setting an unnaturally high bar for Kerry, so that the candidate would be seen as failing to meet expectations after the convention.

But the Times leaves out a key part of the story: One major reason why Dowd’s memo received the attention it did was because the Times itself gave it so much credence. Here’s how the Times’ Richard Stevenson and Jim Rutenberg reported Dowd’s prediction last month:

Matthew Dowd, Bush’s chief campaign strategist, said he expects the race to shift from dead even now to as much as a 15-point advantage in national polls for Kerry by the end of the Democratic convention.

Though campaigns typically seek to set expectations as low as possible and Kerry’s aides have said they do not expect to any such lead, Dowd’s public forecasts have tended to be relatively accurate. He said his prediction is based on the average upward bounce in the polls for challengers against incumbents over the last three decades at this stage in the race.

“We could easily be 14 or 15 points behind in the first week of August,” Dowd said. “How long that lasts and what happens after our convention, well, we’ll have to wait and see.”

This looks to us like a classic example of a “rowback” — a non-correction correction. Kerry’s actual post-convention bounce has, unsurprisingly, been much smaller than Dowd predicted. That’s left the Times looking vaguely foolish for having taken Dowd’s spin at close to face value. So now the paper is returning to the scene of the crime, and suggesting that no one took Dowd’s prediction seriously in the first place.

The larger point is this: When campaign operatives — even those with seemingly accurate track records — offer specific numerical predictions about the race, it’s safe to say that they’re not doing it as a public service. Treating those predictions with the skepticism they deserve would spare news outlets from having to go back and cover their tracks.

Zachary Roth

Zachary Roth is a contributing editor to The Washington Monthly. He also has written for The Los Angeles Times, The New Republic, Slate, Salon, The Daily Beast, and Talking Points Memo, among other outlets.