[Clascomment] Comment on 'Comment on the narrow baryon peak reported by Amaryan et al.' by Volker Burkert et al.

[Hicks, Kenneth]

Dear Mikhail,

Our paper does NOT claim that our background is the only possible choice.  Instead, we simply point out that there are other choices of cuts and backgrounds which are also reasonable, but give a lower statistical significance.

Please do not overstate the conclusions of our paper.  The quotes below are your words, not ours.

Yes, perhaps it is better to continue this discussion when you are at JLab next week.

Ken

-----Original Message-----
From: osipenko at ge.infn.it [mailto:Michail.Osipenko at ge.infn.it] 
Sent: Monday, November 14, 2011 9:33 AM
To: Hicks, Kenneth
Cc: clascomment at jlab.org; burkert at jlab.org
Subject: Re: Comment on 'Comment on the narrow baryon peak reported by Amaryan et al.' by Volker Burkert et al.

Dear Ken,

while I agree with you about the main message of the paper, I disagree with the method, let me put it this way:
1) the signal significance depends on the choice of background shape,
2) Moskov's analysis overestimates the significance because of a particular (not motivated) choice of the background, Now in the paper under review you add one more point: "we choose our background and make a strong claim that Moskov's numbers are wrong, but ours are the only correct ones." I disagree with this. Why do we have to repeat Mosckov's mistake?

I will be at JLab next week for shifts we can discuss it in person then.

Best Regards,
                           Mikhail.

Let me reply to the other answers as well:


> Dear Mikhail,
>
> You have discovered for yourself the main point of our paper.  Since 
> the background shape is not well known, there are various ways to draw 
> the background.  Since the statistical significance depends on the 
> background shape, and also on the cuts applied, then I believe the 
> conclusions of our paper are well stated.
>
> The goal of the paper is simply to explain why the CLAS Collaboration 
> did not agree to publish the Amaryan et al. paper as a CLAS 
> publication. I think the paper accomplishes this goal.
>
> Please see additional replies embedded after your comments below.
>
> Best regards,
> Ken
>
> -----Original Message-----
> From: Mikhail Osipenko [mailto:osipenko at ge.infn.it]
> Sent: Sunday, November 13, 2011 6:34 AM
> To: clascomment at jlab.org
> Cc: Hicks, Kenneth; burkert at jlab.org; osipenko at ge.infn.it
> Subject: Comment on 'Comment on the narrow baryon peak reported by Amaryan et al.' by Volker Burkert et al.
>
> Dear Authors of the 'Comment',
>
> I read through the note and I had a general impression that it is not 
> very well written. Moreover, some observations are weak by themselves 
> (just as the Moskov's signal) and may trigger in a broader public 
> opposite effect to that expected. Let me point out these issues and then some text improvements below.
>
> Best Regards,
>                           Mikhail.
> ================================
> General:
>
> 1. Entire discussion of 'Statistical Anomalies' is very weakly 
> justified and hardly an external reader can agree with it. For example 
> I don't have the data points from Moscov's plots and I have not been 
> involved in the reviews, so I have tried to verify the statements made 
> by looking on the contested Moscov's Figures 8(a) and 9. In Fig.8(a) 
> indeed I read out 1331 counts in three incriminated bins, but already 
> here if I draw by-eye 'a smooth background' and estimate number of counts above it I get 101 counts (not 71 claimed).
> Furthermore, expecting in the same fashion Fig.9 I obtain 106 counts 
> in the peak above 'black dot-dashed line' MC background provided by Moskov et al.
> I am probably totally wrong in my by-eye estimates, but most of reader 
> have exactly the same means I have and will come to exactly same 
> conclusion: where is the discrepancy?
> --> I think your "by-eye" estimates are not very good.  If you like, I 
> --> can
> send you the calculations showing that the numbers in our paper are correct.
>   
Yes, my numbers might be not good, but then one has probably to add more details about how the right numbers can be obtained: the choice of background, number of events in each bin separated on peak and background and may be some plot?

> Could you please remind me if this 'Anomaly' was found by one of the 
> Working Group Review Committees?
> --> Please read the review committee's final report.  You will find 
> --> extensive
> discussion if you just read this reference.
>   
Thank you for pointing out this document. I just looked in it, but could not find any hint of this anomaly. Thus could you please indicate the exact location page and paragraph where it is written?
> 2. The message that few different Review Committees were appointed is 
> missing. I guess it is important to notice that there were 
> different/independent reviewers looking to this analysis, not just the same 'bad bureaucrats' waisting years.
> --> I don't know how you missed this discussion in the paper.  It is 
> --> clearly
> written in our paper that we had review committees and parallel analyses.
>   
It does not state that the Committees were made of different people. 
I.e. I may appoint a hundred committees, but always with the same members, do you think it will convince anybody?

> 3. In conclusion a statement that CLAS Collaboration has rejected this 
> analysis because of serious concerns is necessary.
> --> This is stated clearly in the abstract and also in the paper.
>   
I just suggested to repeat it once more in the conclusion. Some people read conclusion only.
> ================================
> Details:
>
> --> I will communicate with you individually about these detailed suggestions.
>
> page 1, p.1 - 'in the mass of the pK_L system, M(pK_L), at about 1.54 
> GeV' (no mass repeated twice)
>
> page 1 - 'missing mass ofF' is used, is it mistype or it is right way to write it?
>
> page 1 - you say pi+ and pi- are detected, not clear if proton is also 
> detected or it is detected in some cases?
>
> page 1, p.2 - 'reconstructed via momentum and energy conservation, 
> M_X(K_L).' I may try to hint what does it mean, but probably better to change wording like:
> 'assuming that the only unmeasured/missed particle in the event is K_L'.
>
> page 2, p.1 - 'due to other kinematic restrictions' - you could 
> explain better, probably it is acceptance, right?
>
> page 2, p.2 - definition of t_\Theta in terms of p_\gamma and p_K_S is 
> unnecessary, also because it would require to define these 4-momenta as well.
> The text 'the momentum transfer SQUARED' is sufficient. If one 
> interested will find it in Ref.1.
>
> page 2, p.4 - second part of the paragraph beginning with 'In the case 
> of Ref.1, the technical review...' is out of place or not necessary. 
> This section is about review procedure, not on the physics motivations 
> which come in following sections. At most has to be moved in paragraph 
> 5 after: '...many years of effort were spent examining this data 
> analysis.' Also substitution of reviewers of the Review Committees has to be mentioned here.
>
> page 4, p.3 - need to stress that according to the theoretical 
> motivations given in Ref.1 the lower t-cut the better is the channel 
> selection. One can argue that there are many other factors, but 
> practically this is the argument used for the actual t-cut.
>
> page 4, p5 - 'demonstrated in Refs 5 and 6, where the latter HAS remeasured...'
>
> page 4., p.6 - mention of 'ODU group' here and on the beginning of the 
> next page seems to be unnecessary, 'authors of Ref.1' is more than enough.
>
> page 5 - Conclusions - as a first sentence I would suggest something 
> like: 'The analysis of Ref.1 was rejected by the CLAS Collaboration 
> internal review because of serious concerns on the validity of its results.'
>
> ===================================================
>