## Monday, August 08, 2016 ... /////

### ATLAS: a 3.3-sigma stop excess is the new leader

Wolfram Language and Mathematica 11 released today.
No discovery of new physics has been presented by ATLAS or CMS. There are inevitably some bumps and excesses most of which (or all of which) will go away. In a complex blog post, lots of new CMS excesses have been shown. At the end, I do believe that the highly localized 3.7-sigma excited quark bump is the most eye-catching gift from CMS at this point.

ATLAS has just released their big package of new papers. The independence of ATLAS and CMS is also demonstrated by the CMS' decision to write papers based purely on the 2016 data. On the other hand, ATLAS' new papers are mostly based on the whole dataset of 13.2 inverse femtobarns which were collected in the year 2015+2016=4031.

Instead of looking at the ATLAS' papers with modest excesses – there aren't many – let me pick only one paper that seems more intriguing than the rest.

The paper is
Search for top squarks in final states with one isolated lepton, jets, and missing transverse momentum in $\sqrt{s} = 13\TeV$ $pp$ collisions with the ATLAS detector,

ATLAS-CONF-2016-050
Go to Page 24 for the money graph. Before the paper was released, Kilian Rosbach gave his talk whose most important slide is on Page 9/36.

Because I am only looking at one Figure, it may make sense to repost the graph here.

The expected exclusion curves were given by the dashed black lines (plus minus the yellow 1-sigma experimental band). Instead, we got the thick red curves (plus minus the dashed red 1-sigma theoretical intervals). Can you see the difference? Well, you can. The difference is obvious.

Just to be sure, the compact interior of the curves is excluded. The infinite, mostly massive, ocean around these excluded islands are allowed.

You may see that the top squark (stop) excess emerges in two different channels: on the upper graph, the stops decay to a top quark and the neutralino,$pp\to \tilde t_1\tilde{\bar {t_1}}+X, \quad \tilde t_1 \to t+ \tilde \chi_1^0$ while on the the lower graph, the stops decay to a bottom quark plus a chargino,$\tilde t_1 \to b + \tilde \chi_1^\pm.$ The largest deviation in the signal regions is found in DM_low, 3.3 sigma. It's a region designed for the search of dark matter particles produced along with top quarks and this excess mostly causes the lower-than-expected red curve in the upper left part of the upper picture. (The upper picture shows many things in a mass-dependent way; DM_low is just a particular parameter-free set of events.)

Over there, e.g. a $650\GeV$ top squark (stop) along with a $300\GeV$ neutralino was expected to be excluded but it's alive and kicking. There is also a 2.2-sigma excess in the signal region 1 (SR1), probably the simplest one where they expected to see something if the stop were easy to be found.

This anomaly sounds sensible because the same paper with the 2015 data only – a paper I discussed at the end of June – was showing a 2.3-sigma stop excess. So the excess has grown as a well-behaved one should. (Not like a diphoton naughty excess that goes from 4 sigma to zero when the data quadruples.)

The 3.3-sigma excess sounds great if I tell you how many events were expected and how many were seen. The expected number of events in that set was $17\pm 2$. So you may imagine that you get fifteen or sixteen or seventeen or eighteen or nineteen etc. How many did they actually get? They got thirty-five events. (I wrote it in words so that the reader doesn't see the spoilers too quickly.) That's quite an excess. It's almost surprising that it's just 3.3 sigma but let's trust them that it is.

If this excess were due to new physics, the discovery could be announced using the data up to the end of this week, approximately. ;-)

Recall that CMS showed several modest related excesses involving bottom squarks (sbottoms). So it could be interesting or not. Obviously, the discovery-optimistic scenario has been that there would be 5-sigma excesses by now. There have been severl TRF blog posts reporting on possible or actual experimental bumps talking in favor of tops. The oldest one from early 2012 had to be unsubstantiated – otherwise we would already be using stops instead of light bulbs. ;-)

As Matt Buckley previously mentioned (without the right paper), ATLAS is also seeing approximately 3.0-sigma excesses in a multijet search for gluinos, in the channel 5jSR and especially 5jSRb1. Also, ATLAS is seeing a mild excess of stops in jets+MET see Page 26, Figure 11, the non-excluded island, or search for "mild excess".

Non-stop anomalies. ATLAS sees a 2.9 locally or 1.9 sigma globally evidence of a narrow $705\GeV$ scalar decaying to $ZZ^*$.

There's also a 3.5-sigma locally and 2.5-sigma locally excess in resonances decaying to $HZ$ or $HW$ for the mass close to $m_{W'}=3\TeV$. Another $WH$ excess has locally 2.6 sigma and is near $1.6\TeV$.