In their initial analysis of Friday night’s much-buzzed-about presidential debate, the consensus among the punditocracy was that the night was something of a letdown: The debate was—snooze—a tie. “McCain vs. Obama: First presidential debate ends in a draw,” declared the San Jose Mercury News. McCain won on foreign policy, the Wall Street Journal decided; Obama won on the economy. “In a debate that both candidates could ill-afford to lose Friday night, neither did,” the Los Angeles Times opined, “John McCain proved he was resolute and tough; Barack Obama demonstrated that he was smart and polished. And in this case, a tie could be said to favor either.”

Which was just a tad surprising. Because, according to polling conducted immediately after the debate, it was Obama who fared better in the debate. Much better. A CNN poll declared Obama the winner by a margin of 51 percent to McCain’s 38. A CBS poll had Obama at 40 percent, with 22 for McCain, and 38 declaring the debate a draw. Independents in a MediaCurves focus group gave the debate to Obama 61 percent to 39.

In other words, as the Houston Chronicle put it yesterday, “Friday night’s debate in Oxford, Miss., was a split decision of sorts: Most pundits after the debate declared it a draw or gave McCain a slight edge, but viewers polled immediately after the debate said that Obama was the winner.” Or, as Eugene Robinson put it on Friday night’s Hardball with Chris Matthews, “You know, I thought it was a fascinating debate. And my sense is that our initial decision on who won on points is frequently, and dare I say, always wrong in terms of how people at home experienced the debate.”

Again: frequently, and dare I say, always wrong in terms of how people at home saw the debate. Which is, to say the least, a bit troubling. The discrepancy isn’t a matter, after all, of pundits’ analysis adding levels of nuance, context, and other layers of complexity to the people’s basic reactions to the political event they just witnessed; it’s a matter of those pundits’ analysis being, to a large extent, opposed to the opinions developed by the people themselves. While pundits are, of course, under no real obligation to reflect the opinion of the people—just as the people are under no real obligation to reflect the opinion of pundits—one would still hope that the synergy between the two fields of opinion would be organic. We’re all witnessing the same event, after all. So when conclusions differ, you have to wonder: Why the discrepancy? Why weren’t the pundits’ assessments of the debate more in line with those of their audiences—and vice versa?

The answer, I’d say, comes down to a discrepancy in the approaches themselves, to a difference between the standards the people and the press use to assess the candidates’ performances in the debates. In short, in general, in this case: The people were assessing the candidates according to the substance of their oratory. The pundits were assessing the candidates according to their style.

The debate was a tie, the pundits claimed, not because the policies the candidates discussed were equally clever or viable or what have you; it was a tie, rather, because neither candidate made a gaffe. And because neither candidate fired off any good, memorable one-liners. Indeed, the overall consensus among the pundits was that the debate, besides being a wash, winner-wise, was also…Dull. Wonky. Yawn-y. BO-ring. “Like having,” David Brooks told Charlie Rose, “a lifetime subscription to Congressional Quarterly.”

Megan Garber is an assistant editor at the Nieman Journalism Lab at Harvard University. She was formerly a CJR staff writer.