1) J/psi-h correlation in run12 pp 200 GeV - Bingchu

The current systematic uncertainty is large and maybe overestimated. It is suggested to reconsider how to get a more realistic systematics estimation.

Relatedly, it can be considered how to treat the overall systematics which are the same for different dPhi bins. An extreme treatment is adding a constant scale when fitting the B->J/psi ratio.

2) Run 12 p+p 200 GeV, NPE-h deltaPhi Correlations - Zach

Long’s analysis show some discrepancy of D-h correlation between data and pythia. We need to understand whether pythia is trustable to generate NPE-h correlation.

It is suggested to try different generators and different pythia versions to see how confident we are for the generated charm and bottom NPE-h correlation.

The acceptance correction could be different for NPE-h and h-h, especially when NPE has HT trigger. So the mix event correction should be done for NPE-h and h-h separately. 

Is hadron efficiency vs. pt from embedding applied? Need to check and update.

Page 11, the data is lower than both b and c contribution from simulation. This is better to be understood. The p[1] parameter in the fit ideally is not needed.

Will finish some more studies, especially “fit old data with new templates”, before deciding whether to put into Xiaozhi’s talk for QM.

3) J/psi and Upsilon analysis using MTD data - Rongrong

For many studies, nSigma dE/dx efficiency is not from embedding. We can validate the embedding nSigma dE/dx by looking at pion nSigma dE/dx efficiency with the same cut, using both data pure pion sample and embedding.

check the invariant m range for the count method, how many sigmas away are the cuts from the mean. This can be done both with individual fitting results pt bin by pt bin, and with a pol1 fit of the overall mean / sigma vs. pt. 

Page 15, for the trigger scale, can we do it run by run?

page 17, the yield is not 0 at 0 pt. We can use the TBW fit scaled by 2pi*pt...

Page 18, the MTD matching efficiency turn on rapidly between 1 and 1.5 GeV/c. It is suggested to try cutting on daughter pT 1.5 and have more controllable systematics from this part for QM.

Page 21, use mean pT as x position or do some correction for the decreasing spectrum in a wide pT bin.

Page 22, the MTD data points are systematically lower, better to understand.

There are usually at least 2 parameters to describe pt resolution vs. pt. The current study do a constant additional smear based on resolution from embedding. The resulting pt dependence of pt resolution may not be right. If possible with the significance of the J/psi sample, it’s better to have more degree of freedom. The fit on Page 28 can be done simultaneously for multiple pT bins. The daughter pT ranges for Jpsi and Upsilon are different, if the pt dependence of resolution is not right, the Upsilon line shape could still be off even the J/psi is tuned right.

Because the difference shown in Page 31 right, LS same event is better in terms of systematics, although the statistic error is larger.

4) J/psi->mumu v2 - Takahito