Tau gen matching

[pieterdavid]

From @lwezenbe : the current tau gen matching settings in https://github.com/cp3-llbb/Framework/blob/CMSSW_8_0_6p/src/TausProducer.cc allow for very large distances between the reconstructed and gen-level taus, see the discussion in https://hypernews.cern.ch/HyperNews/CMS/get/tauid/897.html?inline=-1 for more information

[clacaputo]

Hi @pieterdavid, the code used for matching follows the guidelines by Htautau group.
In the hyper-news thread you point they said

More generally, we have used and recommended the matching suggested by the H->tau tau group for a while, see here https://twiki.cern.ch/twiki/bin/viewauth/CMS/HiggsToTauTauWorking2017#MC_Matching, which is to very good approximation also implemented in the current NanoAOD.

It's strange that there are very large distances, also considering the cut value here:
https://github.com/cp3-llbb/Framework/blob/CMSSW_8_0_6p/interface/TausProducer.h#L45

I will further investigate.

[lwezenbe]

Hello, What I was referring to when I was talking to Pieter was line 19: https://github.com/cp3-llbb/Framework/blob/CMSSW_8_0_6p/src/TausProducer.cc#L19. At first sight, it seems like the kinematic variables of the match are filled with a match found through the .genParticle() ref matching. This matching sometimes ends up matching a lepton fake with a gen tau that has nothing to do with it (the tau POG recommends using the .genJet() matching). The matching done afterwards that you are referring to seemed correct however. On 29/03/2019 14:54, Claudio Caputo wrote: Hi @pieterdavid<https://github.com/pieterdavid>, the code used for matching follows the guidelines by Htautau group. In the hyper-news thread you point they said More generally, we have used and recommended the matching suggested by the H->tau tau group for a while, see here https://twiki.cern.ch/twiki/bin/viewauth/CMS/HiggsToTauTauWorking2017#MC_Matching, which is to very good approximation also implemented in the current NanoAOD. It's strange that there are very large distances, also considering the cut value here: https://github.com/cp3-llbb/Framework/blob/CMSSW_8_0_6p/interface/TausProducer.h#L45 I will further investigate. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub<#290 (comment)>, or mute the thread<https://github.com/notifications/unsubscribe-auth/AqConRZqiGXOiWAYD9M3CfzSfL1VmACzks5vbhsTgaJpZM4cSSNa>.