Friday, December 18, 2009

75% Chance of Laboratory Detection of Dark Matter Reported

From a press release by the Cyrogenic Dark Matter Search (CDMS) collaboration:
"In [the 2007-2008] data set there are indeed 2 events seen with characteristics consistent with those expected from WIMPs... We estimate that there is about a one in four chance to have seen two backgrounds events..."
Here's a link to a news story on this release from the Daily Mail (UK).

Now clearly a 75% chance of having detected weakly-interacting massive particles (WIMPs) is different from a detection, which would have taken 5 particles according to the CDMS press release, but it's still pretty darn good. Perhaps the most useful result is that the claim to have placed much stronger upper limits on the interaction rates of possible WIMP candidates. The press release is not a journal paper or even a pre-print so it's pretty short on hard data, but overall this seems to be an important day for 23% of the mass of the universe.

[Update]: There is a pre-print on with some of the technical details. Check back here later for a slightly more technical summary.


  1. What, not a 5 sigma detection in a Nature paper as the original rumors would lead us to believe!!!

    Be honest, how many of your schools had people placing bets on how big of a deal this announcement would turn out to be?

    75% is interesting, unfortunately, the particle community has settled on a 5-sigma excess to be the rule for detection.

    Perhaps in the future we will have a discovery as I believe they still have more data to collect.

  2. Oh you picky particle people.

    In the good old days of astronomy Hubble's law was formulated based on a fit that was clearly not 5-sigma accurate. Now you cosmologists and particle physicists actually want to be careful with statistics and suck all the fun out of life.

  3. Nick it's true. I think it stinks as well. I could see the need for at least 3 sigma, and possibly even 4 but 5 to me seems too much.

    I have a hard time thinking if you had a 4 sigma excess that you haven't discovered something other than a statistical fluctuation. And even 3 should grab your attention.


To add a link to text:
<a href="URL">Text</a>