A longtime follower of this blog, Tony Smith, pointed out to me today this arxiv paper published three days ago. In it, CMS data from Run 1 of the LHC are used to speculate that there might be a second Higgs boson hiding in the data at a mass of about 145 GeV. Check out the two graphs that they produce.
The first one, shown below, is their own interpretation of the four-lepton invariant mass from CMS data and background in the H-->ZZ--> four lepton final state:

And the second which follows is the mass spectrum of photon pairs, again with their interpretation in terms of two bumps for the 125 GeV Higgs and 145 GeV "signal":

Now, rather than commenting on these graphs and their strange features (such as the variable binning shown by the first one, or the different bullet size of some of the points in the second), let us see what CMS officially produced - one wants to stick with the certified data first. The CMS four-lepton invariant mass distribution in the 125 GeV vicinity is shown below.

One notices that the background computed by CMS is significantly smoother than the one plotted in the first figure of the arxiv paper: the latter must have been created with some low-statistics Monte Carlo simulation. Anyway, the CMS data do show some 4+7+4=15 events in the 139-148 GeV region, where backgrounds are roughly predicted to yield 9 events. Is that an observation worth discussing ? 

The answer is of course no - the probability that one observes 15 or more events from a Poisson process producing on average 9 is, even assuming there is no uncertainty on that background expectation, of 4.1% - a 2-sigma effect. But that assumes no background uncertainty; if we take a 20% uncertainty on the background, the probability grows to 7.7%. Anyway, that is beside the point, as we had no right to focus on 145 GeV in the first place; rather, we have been looking at a very wide mass range for departures from the background, so we are bound to see a 2-sigma fluctuation somewhere. This is called "look-elsewhere effect", and I have discussed it ad nauseam in several other blog posts, hence please allow me to avoid re-doing that here.

Let us now look at the diphoton mass distribution: if there were a signal at 145 GeV there as well, one might still retain some hope that another Higgs boson is hiding there. The official mass distributuon published by CMS in 2013 from the diphoton decay mode is shown below.

I don't know about you, but I cannot see any enhancement at 145 GeV. If anything, there's some slight fluctuation at 135 GeV, but I do not think anybody would call that a signal of a new particle... 

So, what to make of the recent preprint ? I do not think it is a joke - in the sense that I believe the authors are serious. They took the time to write a text around the two graphs, and although the text contains some serious inaccuracies and the graphs themselves are of dubious origin, it should be taken as a honest piece of work until proven otherwise. 

However, if you were to ask my personal opinion, I would have to say it is garbage. While it is unfortunately very common nowadays, for phenomenologists who have no better idea to entertain themselves with, to take a 2-sigma fluke and building a paper around it, I have seen decent outcomes of that exercise. This article does not pass the mark, for a number of reasons. I prefer to not list them here, as I believe you should be allowed to form your own opinion on it. Hence click on the link and browse it: it's just 5 pages long, it does not contain formulas, and is only sparsely affected by occasional physics arguments. It should not take you more than 5 minutes...

As for me, I certainly do not have the time to dig deeper in the matter, considering the unsubstantiated nature of the claim. But maybe some of you will - for instance, one could go out and fetch the corresponding ATLAS graphs. Is there a 145 GeV signal there ? I leave it as an homework for those of you who have a half hour to spend. Good luck!