So the big news of the week is that we've just had Moriond. CMS presented a reanalysis of their 13 TeV data on the diphoton excess; this included some additional data taken with zero magnetic field, equivalent to a 10% improvement. I heard that ATLAS pulled out of presenting a new analysis, though that's not obvious from their talk. CMS's new result, including 8 TeV data, has a 3.4σ local significance for the excess (up from about 2.4σ). The rumour I heard for ATLAS was 4.7σ local!! Which may be why the pulled out, that they got a significance too high for them to put faith in it.
The LHC is still in winter shut-down. We won't really know if this is something knew, or just a cruel quirk in statistics, till the Summer. On the other hand, the experimental collaborations have not been able to find anything wrong with what they've done so far. But on the original hand, at least half the reason people are so excited about this is because we don't have anything else to latch on to. This is the best year to make a discovery at the LHC. Given how long it will take for the next generation machine to turn on, this is the last chance for some people to see a discovery in their career.
Personally I remain somewhat sceptical. Partly that's my own personality, but even in my relatively short career I've seen a lot of potentially exciting things go away with more data. Also, this is not something anyone was predicting, which is both a good and a bad thing. The good is that it's a surprise, and that's how we learn. Plus it means I'm not horridly disadvantaged against the people who spent their lives studying three-loop SUSY or whatever. However, from a Bayesian perspective, a surprise corresponds to a lower prior: being unexpected makes it less likely.
That said, I've written one paper and am writing another on the excess. There's one thing I'm not sceptical about, which is that this is a citation magnet.
 |
| ATLAS (red) and CMS (blue) results, from Strumia's Moriond talk |
The LHC is still in winter shut-down. We won't really know if this is something knew, or just a cruel quirk in statistics, till the Summer. On the other hand, the experimental collaborations have not been able to find anything wrong with what they've done so far. But on the original hand, at least half the reason people are so excited about this is because we don't have anything else to latch on to. This is the best year to make a discovery at the LHC. Given how long it will take for the next generation machine to turn on, this is the last chance for some people to see a discovery in their career.
Personally I remain somewhat sceptical. Partly that's my own personality, but even in my relatively short career I've seen a lot of potentially exciting things go away with more data. Also, this is not something anyone was predicting, which is both a good and a bad thing. The good is that it's a surprise, and that's how we learn. Plus it means I'm not horridly disadvantaged against the people who spent their lives studying three-loop SUSY or whatever. However, from a Bayesian perspective, a surprise corresponds to a lower prior: being unexpected makes it less likely.
That said, I've written one paper and am writing another on the excess. There's one thing I'm not sceptical about, which is that this is a citation magnet.
No comments:
Post a Comment