...and two people made gifts already. So nice!

Meta analysis shows modest but real effect of acupuncture in treating chronic pain.

Meta analysis shows modest but real effect of acupuncture in treating chronic pain.

Huh - that's very surprising (since no individual well-designed study has shown an effect). I'm curious what the numbers are for "real" meridian-points acupuncture vs random-needles-anywhere acupuncture. (I'm willing to believe that poking people with needles has a physiological effect, I'm really not willing to believe that it's because the needles are manipulating the flow of chi.)

I don't know how it works, but it has been mad effective for treatment of Beau's skin-related allergies. Seriously.

Today, I am a fountain pen.

Jessica goes Old School, yo.

My hand to God, that's exactly what I was going to post, and then I thought, "Way too dorky and obscure. Who else would even remember it, let alone get it?"

Truly, y'all are my tribe.

I wonder if the "sham acupunture" mentioned is random-needles-anywhere acupuncture, or what. If anyone has access, the full write up is here: [link]

Also, question for the mathically inclined:

I've now gone from nice guy and all to deluded fanboy because I don't repudiate Nate Silver for having at some point in the run-up to the 2010 midterm elections given the Democrats a 20% chance at retaining the House -- both Silver and I are clearly Dem-pandering morons because the Dems always had zero chance of retaining it.

I *think* he's confusing the real-world likelihood of capitalizing on a 20% chance of success with the statistical probability, which I don't know has ever been zero for any election ever in US history, but I'm not sure my guess is correct, and, if correct, I'm not sure I'm explaining it right.

Help a deluded fanboy, please?

unless the person with whom you are arguing can demonstrate through mathematical modeling that in ALL circumstances, the Democrats could not retain the House, he is full of shit.

Democrats would have had to lose X number of races in order to lose the House and there is no way you can be 100% certain of that eventuality unless all polling data showed EXTRAORDINARY weakness of the Democrats running. 20% chance is low enough.

If Romney projections were 20% of winning the Presidency, the rich members of the GOP would be losing their shit more than they are currently.

Thanks, le n. That's what I'd been groping toward in my non-mathy way -- in a problem with as many data points and as little actual available data (public polls for only 25% of the House races) as the 2010 elections, how on earth do you conclusively calculate an absolute negative?

My brain hurts just trying to put the words around that.

well, JZ, absent of context, Nate Silver expressly does not want to do that. He isn't a commentator or a politico. He provides data and analysis. Therefore, he shows what polling and other things may mean for race predictions.

He expressly does not come down on the side of "ain't no WAY that will happen." If said poster is looking for that kind of analysis, perhaps he needs to choose sources for his information that are less nuanced.

Then again, it is difficult to see journalistic truth in sources that are not nuanced.