http://cms.cern.ch/cds/BPH-13-007


**Commenti dell'ARC chair da tenere in considerazione nei commenti che mandiamo**

Dear colleagues,

Many thanks for your participation to the institutional review of BPH-13-007.

You should be aware that the institutional comments in CDS and their
responses will be also posted in a twiki page shared with LHCb. This
twiki page is open to both CMS and LHCb general mailing lists:
https://twiki.cern.ch/twiki/bin/view/Sandbox/BsmumuCombComments

Joel Butler for the authors is taking care of filtering the comments
that are CMS-specific or may be be perceived as offending for LHCb.
There are very few such comments.

Please tell us if you wish that your comments -- or part of your
comments -- remain CMS-internal, or if you want to rephrase some of
them before sharing them with LHCb, or if you want your name to be
replaced by your institution, or if you have any other concern about
sharing your comments with LHCb.

Best regards,

Gautier


**Commenti dell'ARC chair da tenere in considerazione nei commenti che mandiamo**

Dear Institutional Representatives for the IRC of BPH-13-007,

thank you for accepting to serve as Institutional Review Committee on
the paper draft for BPH-13-007 entitled "Observation of the rare
B0s -> mu+mu- decay from the combined analysis of CMS and LHCb data" to
be submitted to Nature. For the preparation of your comments, I wanted
to point out a few conditions that went along with the approval of the
analysis yesterday as well as a few wording changes on the paper draft
that were the result of additional negotiations on the text with
LHCb. Please keep these issues in mind when composing your comments:

* Outcome of the approval meeting:

1) Use the "S/S+B weighted all categories" mass plot in the paper and the "six best categories" in the extended session (swap Figure 2 with Extended Data Figure 2) and modify the text accordingly
2) In all 2d plots (Figure 3 and Extended Data Figure 6) remove the 5 sigma and 6 sigma contours (or smooth them)
3) Include the individual BR results from CMS and LHCb in the extended data section with their uncertainties but without significances
4) In the summary part of the paper emphasize the “2 sigma compatibility” with SM (without reference to the 3.0 sigma “evidence”)
5) As a suggestion the major emphasis should be put on the Feldman-Cousins 3.0 sigma result instead of the 3.2 sigma figure based on the delta-log-likelihood

* Wording changes agreed upon with LHCb spokes:

@L23: Instead of
"It also produces the first 3 standard deviation evidence for
B0 -> mu+mu- decay."
new text:
"It also produces 3 standard deviation evidence for an excess
of events in the search for B0 -> mu+mu- decays."

@L204: Instead of
"It also produces the first $3\sigma$ evidence for B0 -> mu+mu decay."
new text:
"It also produces $3\sigma$ evidence for an excess of events in
the search for B0 -> mu+mu- decays."

Thank you for your attention to this matter. We are looking forward to
receiving your CWR comments by the deadline of Sept 8, 2014 (sharp). 

Best, Manfred (ARC chair).



**Commenti di Linda F**

13-15: If different rates... it would prove ==> *Any divergency from these
predictions* would prove...
16: LHC had the opportunity ==> LHC *has* the opportunnity
23: to date ==> credo sia implicito
23: it also produces ==> it also *yields* (metterei un sinonimo perché c'è
una ripetizione)
35-36: neglected but now had to be included ... results ==> neglected*, in
order to* match the more precise experimental results
44-45: Many theoretical ... short-coming ==> secondo me c'è qualcosa che
non funziona nella costruzione della frase
45: short-coming ==> shortcoming
46: SM predictions to guide ==> SM predictions*,* to guide
48-49: B0 mesons consisting ... d-quark into a dimuon ==> B0 mesons*,
*consisting
... d-quark*, *into a dimuon
53: internal loops as ==> internal loops*, *as
56-57: The SM predictions are ... SM predictions ==> The SM predictions
are: B()=... and B()=... *, very far, prior to the start of LHC, from *the
experimental upper limitis
65-66: uncertainties, would ==> uncertainties would
68: predicted in the SM ==> predicted* by *the SM
69: beyond the SM ==> beyond *this theory*
80-81: toglierei l'ultima frase
84: and and ==> and
85: strategies are different and complementary in the two experiments ==>
strategies are different and complementary. (implicito)
105: are expected for the two together ==> are expected. (di nuovo
implicito)
132: key ==> *the* key
174: S/(S+B) where S and B are the number of expected signal assuming SM
rate, and background events for a given category ==> S/(S+B)*, *where S and
B* are the number of expected signal and background events*.
176: Fig. 2 where ==> Fig 2*, *where
180: LHCb can be ==> LHCb*, *can be
183: deviations, σ, and ==> deviations* (σ) *and
186: Fig 3 where ==> Fig 3*, *where
221: in the data while ==> in the data*, *while
227: vertex; and variables ==> vertex; variables
234: published results each experiment ==> publish results*, *each
experiments

**Commenti di Gian Luca P**

Line 30: Could be useful specify that the observation is a consequence
of a 5 sigma statistical significance?  e.g.  ...decay could be
observed. -> ...decay could be observed reaching  5 sigma statistical
significance needed to declare the discovery.

Line 84: and and -> and

Line 90:  ...with high mass, 100 GeV to a few.. -> ..with high mass,
from 100 GeV to a few..

Line 92:   CMS is designed to be able to study B decays -> CMS is
automatically designed to be able to study B decays .

Line 93:  they instrument complementary geometrical volume -> they
work on complementary geometrical volume.  I didn't find the use of
instrument as a verb anywhere.

>From 151 to 155: I will reconstruct the sentence in order to be more
immediately understood.

 Both analyses include in the MVA the displacement between the
production and decay vertices of the B meson to discriminate against
combinatorial background. Therefore the efficiency as a function of
lifetime has a model dependent correction [22], which is estimated
through simulations assuming   SM dynamics. ->
Since both analyses include in the MVA the displacement between the
production and decay vertices of the B meson the eefficiency, as a
function of lifetime, has a model dependent correction [22], which is
estimated through simulations assuming SM dynamics.  (Or something
like that...)

Line 161: I guess the data-taking periods means to 2011 2012 and
consequently the energy in the centre-of-mass. I will replace
data-taking periods with the energy in the centre-of-mass.

Figure 3: I will add the SM prediction also to the two plots on the
left as in Figure 4.

Line 396: Same problem as line 93.

Extended data figure 6:  I will add the SM model prediction.


**Commenti di Stefano A.**

The Torino group wishes to congratulate the authors for this achievement and for the very well written paper. Below we have a few small comments.


Type B

1. There is no mention on how the backgrounds are treated in the extraction of the BR until L484. We suggest to move the explanation to the "analyses" section.

2. In the introduction, the shortcomings of the SM are described at length (L40-47). This puts the reader in expectation of a measurement in disagreement with it. Unfortunately this analysis does not show any evidence of deviation from the SM. We therefore suggest to give less emphasis to the shortcomings and perhaps higlight the successes of the SM !


Type A

L26-30 : consider removing from “The LHC experiments “ to the end of the paragraph. The point is already made in L209 and it does not say anything about the content of the present paper, so there is no need to make it in the summary.

L41: keeping in mind comment 1. above, we don’t understand the choice of references here. L41 says “recent years”, yet Refs [6,7] are from 1995. Perhaps in L41 [6,7] should be replaced by [8], and [6,7] should be moved to right after “(CP violation)”. 

L46: replace “mounted “  with prepared/designed/devised/envisaged

L50: a large space is present between the period and “In the SM"

L53: remove the word “complicated”

L53: "diagrams with" -> diagrams featuring

L57: “there was a large gap” -> The precision achievable by experiments was too low to enable a confirmation of SM predictions

L84: “deviations due to physics outside of it” -> deviations from it

L85-L92: wonder if this blabla on the strategies for the search for NP isn’t a bit outside the scope of the present work

L118: “multivariate analysis (MVA) discrimination variables that combine ..”  The variables are just variables, not MVA variables. Suggest : “m_mumu, and a set of kinematic and topological variables combined using MVA analysis techniques.” . Optionally add “These techniques allow superior S/B discriminating power and and efficiency “


L123-126: consider removing “The invariant mass … not detected”

L126: "There is also “ -> “Another source of "

L131: remove the word “dangerous”

L135: “direction of the muons relative to the beam direction” ->”relative to the beam axis" (avoid repetition)

L258: the link of Ref [6] is broken

L262: “A&A” -> replace with suitable abbreviation 

L288: link of Ref [19] does not work

L382: "to a few tens of mm” : most lifetime plots have ranges up to a few mm, not tens of mm. One can argue that a tiny fraction of B will fly for “few tens of mm” (= centimeters), yet we wonder if the words “tens of” are a typo.

L397: we believe the symbol for radian is “rad” rather than “r”

L408: “translates to” -> “translates into”

L492: mixing degrees and radians sounds awkward. Symbol for radians is “rad”

L433: “at the particle momenta in this analysis” -> for the momentum range of interest for this analysis

L441: “look for heavy objects decaying into two muons” -> remove the word “two”

L450: reference -> references 

L521: References -> Refs.

L528: “this channel’s efficiency” -> the efficiency of this channel

**Fine commenti Stefano A. **