Dear Bfys colleagues,
Please give me some better input for a general comment. I am also not certain that what I used from Patrick as a general comment is clear enough.
Tjeerd PS I will upload Wednesday evening.
Comments on LHCb-PAPER-2012-033 by Nikhef (imput form from Patrick, Rob and myself)
General comment =============== In the introduction the "weak phase gamma" is used. Gamma is an angle of the UT triangle, not a CKM phase. If you accept that CKM is the whole picture, then gamma is very well measured via global CKM fits.
Comments by line: ================= Title: Add "relative" to "measurements of the ... branching fractions". Leave only "decays" plural, the other words "singular".
Abstract: Replace "normalized with respect to Bbar0s ..." by "relative to the decays Bbar0s ...".
Line 11: Replace "which is measured to a precision of ~ 10^o - 12^o [3,4]" by "which is fitted to world data with \gamma = (68.5 \pm 3.1) degrees [3].
No Capri value for gamma can be found on the website of [4]. On the Utfit website, I found a different result for the uncertainty. If you want to keep the larger uncertainties, mention that those refer to direct measurements. It seems more logic to give both the value and the uncertainty than only a range of uncertainties.
Line 12: "Which are inherently free from physics beyond the SM" is a bit strong. I would say "for which the SM contribution is largely dominating".
Line 14-19: - The bar on B and Bs are different. Please, fix your templates. - Replace "These decays may be extended to multibody-modes" by "These studies may be extended to multibody decay modes". - I would put the sentence in lines 18-19 before line 14 where the different decay modes are introduced.
Lines 22-24: This sentence may be reformulated. For instance: "Another contribution from mixing, followed by a b->u W-exchange transition, is expected to be negligible compared to the b->c transition, because it is suppressed both by the Cabibbo angle and by helicity and color conservation."
Line 28: - Add "in 2011" after "... collected by the LHCb experiment." Otherwise the reader may think of 2012 and we now have more data. - Replace "The same data sample is used to observe the decay ... for the first time" by for instance "In the reference data sample of Bbar0s->D+s pi- pi- pi+ we observe the decay ... for the first time and we measure its relative branching fraction".
Line 40: Replace "resolution of 20 mum for tracks with high transverse momentum" by "resolution smaller than 20 mum for tracks with transverse momenta, pT, larger than 1.0(??) GeV/c."
Give a correct value for "high". Introduce the variable pT here.
Line 49: Replace "transverse momentum, pT," by "transverse momenta, pT," or "transverse momenta", or just "pT" when pT is already defined in line 40. Give a value in stead of "high". Use plural in the sum.
Line 51: Replace "the primary interaction" by "the primary interaction vertex".
Line 61: Replace "B candidates are formed by pairing" by "B0 and B0s decay candidates are selected by pairing".
B is undefined. Is that the same as the B_(s) below?
Line 63-67: There's an ordering problem here. It's not clear if D(s) daughters count as B daughters or not.
Line 70: I presume that with "m_K*0" is intended the world average mass. Same for "m_D+" in line 76. This should be mentioned as in line 67.
Line 80: Replace by "Candidate Xd and Xs final states are selected in a similar way, while only allowing invariant masses smaller than 3 GeV/c2".
Line 86: Remove "Lastly". The list of selections is not finished yet.
Line 92: What do you mean with vertex constraint? PV? You should cite DecayTreeFitter here, if that's what you use.
Lines 93-95: "To suppress ... is required to be nine" seems too condensed. An alternative may be: "To suppress charmless background a D+s decay vertex is required separated from the B vertex by selecting \chi^2(VS) >9 in the vertex separation fit. This is primarily important for the decay Bbar0->D+s K- pi+ pi-."
The "flight distance significance" seems to be also useful here.
Line 100-105: Do not use bullets. Use a table with the variables instead. Do not use bold "min" and "max". Do we need to know the long descriptions and that we take a logarithm?
Line 109: This FoM is only defined if you assume a BF for the decays you want to observe. What did you choose and why?
Line 117: This reads as if CB is a radiative tail only. What is the Gaussian used for?
Line 118-120: This sentence is not well understood. Did you mean: "In fits to data a common scale factor for the width of the CB and Gaussian is left free to account for a slightly larger resolution than in the MC simulations."
Line 151: Does the value of 86.8 use our measurement of this quantity? (If not it should).
Table 1: What are the sideband(s) regions? A figure with M(Ds->pi pi pi) and M(Ds->K pi pi) might help or otherwise give the value of the mass region limits.
Line 162: Give a reference to "sWeights". Where does "these decays" refer to? Do you mean here: "To study the B0(s)->D+s Xd(s) decays, the Xd(s) invariant mass ... are determined from the sWeights [..] in the following way" with some additional explanation?
Figure 3: The line for "Full PDF" does not have the same width in the label and for the curve.
Line 171: Please, be explicit on "the lower-lying strange mesons"?
Figures 4, 5, and 6: - Add "LHCb preliminary" in the plots. - Why is the binning so much larger than the mass resolution?
Line 191: Can we show the upper Bs mass sideband result? That it shows no significant structures by itself is not very reassuring with the present statistics.
Line 197: To evaluate the "significance" of the peak, please fix the mean of the Gaussian to the expected mass difference of 565.1 MeV. Otherwise you suffer from local look-elsewhere effects.
Line 199: I do not understand your comment: "variations in the background shapes do not change this conclusion". With such a small number of events of both signal and background, I would expect a reasonable dependence on the chosen shape. Flat, linear, threshold function at zero, polynomial, etc.
Line 200: Remove "therefore".
Figure 7: This is definitely not convincing. Re-binning into a smaller number of bins is advised, such that every bin has at least two events.
Line 203-204: - Why do you introduce e^q_rel and define them in the text? Just substitute the definitions in the formula. - The equation should be numbered.
Table 2: - Replace (10^-2 %) by (10^-4). In the equation epsilon should otherwise be multiplied by 100. - M >3 Corr has something to do with the M(X_s,d)<3 GeV/c^2 veto. Please, be explicit, use the same notation, and put units where appropriate.
Line 232: Is "two standard deviations or less" relevant? You give no information to the reader to decide if it's likely to be a statistical fluctuation or something to worry about.
Line 237: Replace "Systematic error" by "The systematic uncertainty".
Line 262: Missing first "yield", and ",". Our suggestion is: "The larger sensitivity of the Bbar0 yield than the Bbar0s yield arises, because these ..."
Table 3: - Replace "Summary of systematic uncertainties (in %) on the branching fraction ratio measurements" by "Systematic uncertainty contributions in percentage of the measured values of the relative branching fractions". Choose between "relative branching fractions" and "branching fraction ratios" for the whole paper. It is an exceptional choice to give relative uncertainties for fractions instead of absolute uncertainties. So write it clearly in the text. - Spell out "Efficiency". - "fs/fd for Bbar0 decays": replace "0.0" by "N/A" or " ".
Lines 266-268. - Move "relative" from "efficiency" to "systematic uncertainty". - Add this information as a 4th column to Table 3.
Line 273: Remove "and Summary". The results are derived here, not summarized.
Line 274: - The two equations below line 202 are not numbered (see Lines 202-203). - We use an equation to calculate a result, not to measure one.
Line 275: Replace "systematical" by "systematic".
Line 280: "an additional 35% more signal events relative to the signal yield" is not understood.
Lines 293-294: "... yield used here is only the signal within .. of the signal peak, which is 5505 events (97% of the total yield)" This is not understood.
Equation below line 295 (and in the abstract): Why in units of 10^-3, while the other relative branching fractions are given in units of 10^0 ?
Lines 311 and 315: Refs [3] and [4] have too large spacing after the title.
========================================================