The July 2011 Higgs Hints

[This report from the  July 2011 Grenoble EPS-HEP11 conference is out of date, but is maintained for historical interest.  The hints mentioned, which are discussed in detail here and were somewhat tenuous from the start, disappeared by the end of summer.  ]

Grenoble, EPS-HEP11 conference, 7/22/11

3:00 pm Kyle Cranmer of New York University presents the summary of the ATLAS search for the Standard Model Higgs thus far.  This involves combining many different search strategies.  Excludes 155-190 GeV at 95% confidence level,  beats the Tevatron experiments, which are now basically obsolete for this search.    First exclusion at high mass: 285-400 GeV now ruled out.   Interesting excesses possible 120-145 GeV at 2.8 standard deviations.  Particle physicists insist on 5 standard deviations for a convincing story…

[WARNING: see below for the meaning of this 2.8!!!]

3:20 pm Andrey Korytov presents the summary of the CMS search for the Standard Model Higgs thus far.  Again 6 different studies combined together.  95% exclusion ranges are 149-206, 300-440, and much of the region from 200-300.  90% exclusion 145-480.  Interesting excesses possible 120-145 but statistical significance hard to evaluate at this time.  (but somewhat smaller excess than ATLAS sees.)

Whew!  Exciting… but far too early to be sure this is anything interesting.  A very small number of events are driving these effects, [Looking again at the data, that's not really correct --- for ATLAS there are more like 20 events over a background of maybe 100 (will get these numbers more precisely ASAP), for CMS something similar] and this might just be tricks of statistics.   And measuring or calculating the background to the most important search, the one that looks for Higgs decaying to W’s, is very tough.  Mistakes could easily be made here.  But the fact that both experiments see the same thing is very provocative.  This is certainly something we’ll be watching over the next couple of months.

By the way, one other possible interpretation would be that a Higgs particle is in the excluded range of 150-180 GeV, but with a smaller production rate than predicted by the Standard Model.

UPDATE: Been talking to Kyle Cranmer in more detail.  Apparently the two LHC experiments did not share their info before today, so the similar hint of signal was revealed in today’s talks.  And the attempt to combine their results starts now, and probably will not be done in time for this conference’s conclusion.  Instead it may not be shown until late August at the next major conference… but this is speculation at this point.

We discussed what 2.8 standard deviations in ATLAS’s result means, precisely.  [caution: technical stuff ahead] This is before the look elsewhere effect.  It is the probability that given a strategy used for a particular mass (144 GeV I think) that there would have been this large a deviation just from background fluctuations.

A different question — and perhaps a more salient one — is what the probability would be that backgrounds in all the searches could have fluctuated in such a way that a deviation from expectation this large would have shown up.  The probability quoted is 8 percent.  That’s not so small.  So in this sense, the 2.8 standard deviations number from ATLAS is somewhat of an overstatement of the unlikelihood of this happening in the data by accident.

What gives the result more weight is that both ATLAS and CMS see something similar.

And it has also been pointed out that both DZero and CDF show minor excesses in the same Higgs mass region.

Stay tuned.  Within a few months we will know whether this is the real thing or not.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s