Track selection for muons

[amartelli]

The plot below done on some MC events, BToKmumu shows the nValidHits for tracks in several cases:

• all reco tracks
• reco tracks matched to the genMuon2 (smallest dR provided < 0.02)
• reco tracks matched to the genMuon1 (smallest dR provided < 0.02) All tracks are essentially from PackedCandidates

I don't see correlations on DCASig or NormalizedChi2 (nor nLostHits etc...)

In addition to this, I attach a slide from an old study (different framework and samples) where it's visible the reco-acceptance for muons

From these, I suggest to consider as muon candidates, the packedCandidates tracks with

• barrel acceptance and pT < 3GeV (to be tuned)
• nValidHits > 10 (maybe smaller than something?) (to be tuned)

[amartelli]

Adding as suggested, nValidHits_vs_eta for all tracks, for pT > 3GeV and < 3GeV

The range 10-25 (nValidHits) is ~ correlated with being in the barrel