[Clas12_rgb] concerning figure 5 of nDVCS paper draft, VGG predictions compared with our data

[Francois-Xavier Girod]

Dear all

I assume this is intended to go in the collaboration paper
These 2D plots are very hard to read. It's good for intermediary steps in
the internal process, when people look very carefully at the plots and
think about them for a long time. But for publication in my opinion it
would be better to show a pair of 1D plots, one with fixed Ju showing chi2
vs Jd, and one with fixed Jd showing chi2 vs Ju

There are some lattice calculations for these quantities, we should be
careful if we are going to publish results suggesting the collaboration
results disagree with lattice estimations

Best regards
FX

On Wed, Jun 5, 2024 at 9:57 AM ajhobart via Clas12_rgb <clas12_rgb at jlab.org>
wrote:

> the number of fits having the particular chi2, sorry I have omitted it
> from the previous screenshot, here it is
>
> Thanks
>
> Adam
>
> Le 05/06/2024 à 15:53, Sebastian Kuhn a écrit :
>
> Thanks for the explanation. Can you tell me what information is coded in
> the color of each block?
>
> On Jun 5, 2024, at 9:39 AM, ajhobart <ajhobart at jlab.org>
> <ajhobart at jlab.org> wrote:
>
> Hi Sebastian,
>
> For figure 5, the limit on the reduced chi2 that I put affects the set of
> accepted sets of Ju and Jd. The Limit is now 3.5, if I move it ot 5 I have
> a set of accepted sets of Ju and Jd that contains positive values of Ju
> (50% of selected fits). So maybe it is our choice that gave the impression
> that only negative Ju values are reasonable!
>
> My reasoning is that there are preferred values for Jd (positive around
> 0.15) but no preferred values for Ju, I attach here again the same plot
> showing Ju vs Jd versus chi2 from a different angle. It shows that on the
> Ju axes, the distribution for chi2 is almost Flat.<gsudaQzbHjfz8PDK.png>
>
> However, my take on this is that rather none of the combinations of Ju and
> Jd reproduces the data correctly. Even for the ones we chose to put in the
> figure 5, it is quite clear that the trend seen in the data is not
> consistant with the curves. Moreover, VGG predictions remain model
> dependant predictions and we do not intend to consider them as a strong
> physics conclusions on the values of Ju and Jd.
>
> thanks
>
> Best regards
>
> Adam
> Le 05/06/2024 à 14:38, Sebastian Kuhn a écrit :
>
> Just a quick question: Why are all the preferred values for J_u negative and for J_d positive? This seems very counter-intuitive, given that the quark helicity contributions to J have the exact opposite sign. This would require enormous (overcompensating) orbital angular momenta to explain.
>
> - Sebastian
>
>
> On Jun 5, 2024, at 4:07 AM, ajhobart via Clas12_rgb <clas12_rgb at jlab.org> <clas12_rgb at jlab.org> wrote:
>
> Dear Valery, RGB,
>
> following our discussion during the last RGB meeting concerning figure 5 of the nDVCS paper draft, we have performed a scan using VGG of different values of Ju and Jd and evaluating for each set a chi2 with the 9 data points that we extract with our analysis. Indeed, this scan has shown that there are other values of Ju and Jd that are more "preferred" by our extracted harmonics than the ones we had initially put in the paper. We have updated figure 5 accordingly (I attach it here also).
>
> color code for figure 5
>
>     //   ju       jd      reduced chi2
>     //-0.45    0.2  3.03423                blue
>     //-0.25    0.15 3.19268               black
>     //-0.2      0.15 3.1004                 Red
>
> Attached also you will find the plot with Ju, Jd and chi2 value for each set as you requested. It seems that there is a particular preferred range for Jd (around 0.15) but none for Ju.
>
>
> thanks
>
> Best regards
>
> Adam
> <bsa_q2xbjt_coloronline_newVGG.pdf><jujdchi2.png>_______________________________________________
> Clas12_rgb mailing listClas12_rgb at jlab.orghttps://mailman.jlab.org/mailman/listinfo/clas12_rgb
>
>
> _______________________________________________
> Clas12_rgb mailing list
> Clas12_rgb at jlab.org
> https://mailman.jlab.org/mailman/listinfo/clas12_rgb
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/clas12_rgb/attachments/20240605/fd2e8b4d/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: EWuebJ315AH4fMEj.png
Type: image/png
Size: 326619 bytes
Desc: not available
URL: <https://mailman.jlab.org/pipermail/clas12_rgb/attachments/20240605/fd2e8b4d/attachment-0001.png>