STAR task force for evaluating tracking efficiency uncertainty

[Petr Chaloupka <petrchal AT rcf.rhic.bnl.gov>]

Dear Yiding,

thank you for the question. Let me explain.
The discrepancy (large increase) from nhits>15 to nits>20 comes from the fact that we use charged tracks from decay K→3pi. This means that what is shown is a comparison of tracking efficiency for a tracks belonging to the primary kaon. However the primary kaon can never reach full way to the outer edge of the TPC (as most of primary tracks do) otherwise we would not be able to reconstruct its decay. In other words such track cannot have the maximum of 45 hits. In run18 (which data do you use?) we get to maximum of 30 possible hits, the average is about 25. So when you look at the results for nhits>20 it is quite an extreme cut for such tracks and hence there is a significant difference between data and simulation. One should be carefull to extrapolate to full lenght tracks (nhitPossible~45). I can argue that for a full lenght track the systematic error will not be larger that what is in the table so for example for the cut of dca<2cm and nhits>15 the 5% is reasonable upper estimate. The situation will be different for run19 with iTPC where we will have acces to tracks with more points. Then the estimates for nhits>20 should be closer to full lenght tracks.

With best regards,
Petr

On 08. 06. 23 1:08, yh66 via Star-tf-trkeff-l wrote:

      Dear experts,
      
           This is Yiding from Rice. Recently I am working on the
      systematic uncertainty of di-electron analysis. I plan to add
      embedding/data mismatching uncertainty specifically referenced in
      your presentation slide available at:
      https://drupal.star.bnl.gov/STAR/system/files/TrkEff.pdf. And I
      have one question about it.
      
           In Slide 15 of your presentation, you demonstrate the effect
      of different cuts on mismatching. I observed two intriguing
      aspects in this analysis. Firstly, the mismatching appears to be
      sensitive to the DCA cut, which aligns with similar observations
      made in various studies. However, what caught my attention is the
      substantial difference in mismatching between nHitsFit>15 and
      nHitsFit>20. To the best of my knowledge, previous di-electron
      analyses utilizing Run18 Au-Au 27GeV data did not report such a
      significant mismatching discrepancy between data and embedded
      electrons. Could you kindly shed some light on the underlying
      reasons behind this discrepancy?
      
           I appreciate your great study on the track efficiency
      uncertainty and would be grateful for any insights you can provide
      to help me better understand the question.
      
      
      BW,
      
      Yiding Han
      
      
      _______________________________________________
      
      Star-tf-trkeff-l mailing list
      
      Star-tf-trkeff-l AT lists.bnl.gov
      
      https://lists.bnl.gov/mailman/listinfo/star-tf-trkeff-l