1) The tracking efficiency from only reading StEvent / StMcEvent (MC matching RC via StAssociationMaker with largest nCommonHits) are consistent with those from only reading StMuDst (MC matching RC via idTruth with largest qaTruth).

2) Considering more than one RC tracks can be obtained from one MC track: all RC tracks are saved when matching a MC track, the  obtained tracking efficiency is consistent with that in 1). 

Legend in file TpcEff_cut1.pdf:
Hybrid: track information from reading StEvent, while gRefMult from StMuDst
StEvent: only read StEvent / StMcEvent (MC matching RC via StAssociationMaker with largest nCommonHits)
StEvent + All matched RC: all matched RC tracks are considered.
StMuDst: MC matching RC via idTruth with largest qaTruth
StMuDst + All matched RC: all matched RC tracks are considered.
The nominator is filled with RC pt ==> large fluctuations in final tracking efficiency (peripheral).

file TpcEff_cut2.pdf: The nominator is filled with MC pt, track quality cut changed to be the same as D0 / Lc analysis

file TpcEff_cut3.pdf: The same as TpcEff_cut1.pdf  except nominator is filled with MC pt, and qaTruth>25 --> qaTruth>50