At the beginning, I thought that the experiment had a certain chance to be exciting which was a pleasant feeling.
Commercial break: try CERNland, CERN's games not only for kids.However, I have considered the CDF paper to be obvious crap since the moment when Tommaso Dorigo assured me that the hypothetical cross section of the new physics phenomenon that they were "discovering" was not 73 picobarns but rather a whopping 207 picobarns. That meant that they had hundreds of thousands or millions of "exotic events". That's a lot, especially if you compare it with the small number of new particles that the colliders have a chance to produce.
Mostly because I incorrectly assumed that the CDF people were at least qualitatively sane, I have lost an argument with Tommaso Dorigo concerning the question what the CDF team was actually saying and I had to apologize to him. My scalp made him very happy. If you remember, I just couldn't believe that they were saying such a thing because it is absurd for the Standard Model to be violated by such an intense process in such a seemingly mundane context.
I continued to study the paper and their sloppy accounting of the "unexpected" events made it pretty clear that the error margin of their calculated number is comparable to the number itself. There is no signal that they could see with their sloppy approach.
Fortunately, there are two competing teams at the Tevatron.
Sometimes they are friends who collaborate, e.g. when they need to exclude intervals of Higgs masses for which no team has a sufficient amount of data. But sometimes the redundancy and rivalry turns out to be helpful, and the alleged excess of dimuon events is an example. See
(The document above is top secret - as secret as a document freely available on the web and listed on the official conference website for Tuesday morning may be.)
If you care about the numbers, the tight/loose efficiency for one muon (as measured from the J/psi sample) is 0.92, the efficiency for two muons is therefore 0.92^2 = 0.84, and so is the ratio of the observed tight and loose events at D0, i.e. the ratio 149,161/177,535 (yes, this efficiency is much higher than for CDF, despite seemingly similar definitions of the events). The error is less than one percent, and so is the percentage of events with punch-throughs, cosmic rays, and decays-in-flight.
Hat tip: Tommaso Dorigo (who will need some time to be "forced" to understand that all the effects he considers important are actually negligible, the consistency check works almost exactly, and their paper was bunk)