As Obama speaks at a town hall in New Hampshire today, I have some unsolicited advice for him. No more facts. Let me explain:

Here's a sentence that will never appear in a history textbook: "The turning point of the health debate was when a majority of Americans suddenly realized it was necessity to bend the curve ofgovernment expenditures by at least 1.5 percent for our long-term fiscal health." Why? It's because it's not true. It's because "facts" aren't convincing, and statistics are fine condiments, but you can't make a meal out of them.

But by emphasizing cost-controls in health care, Obama has created a meal out of statistics. And it's a meal that fewer and fewer Americans want to eat. Into the news vacuum of August has rushed a torrent of debates over health care facts -- endless rebuttals of Sarah Palin's "death panels" craziness, the White House seeking out "fishy" misinformation, the new war against Betsy McCaughey. There's a palpable moaning among some left-of-center blogs and columns I read which seems to scream: Does nobody respect the sacredness of facts?

No, actually they don't. Here's what I think Americans think about facts. In Washington, in November 2002, I saw Brian Stokes Mitchell play Don Quixote in a National Theater production of "Man of La Mancha." This was in the run-up to the Iraq War, and musical theater patrons, even in the nation's capitol, are among the least bellicose group in the universe. There is a famous line in the book, and the play, where Quixote exclaims: "Facts are the enemy of truth." For the most part, that's an empty aphorism. But from the mouth of Stokes Mitchell to the ears of musical-theater-goers in late 2002, it was a revelation. The show was interrupted with ecstatic applause for about 20 seconds. Why? Because the "facts" of November 2002 were the rumors of Saddam Hussein's weapons program -- facts that, even if corroborated by intelligence services across the globe, rang false to liberal Americans.

My point is not to equate employer-provided insurance statistics with 2002 weapons intelligence. My point is that Americans aren't study panels seeking out scientific consensus and high p-values. We're moral animals and when statistics or reports don't match our moral order, we'll usually make enemies of them, or flat-out ignore them.

So why fight with statistics? I don't know. But I wonder if Obama and Orszag have made their task more difficult by making health reform about fiscal reform and stuffing their quiver with numbers and facts instead an argument rooted in values. Ezra Klein was right: If Obama's going to win this argument, he has to make it moral.