Below are the "efficiencies" for run 9 pp data (pi0 and gamma-rich triggers).  By "efficiency" I mean the ratio of of tracks after QA cuts to those before QA cuts, so it's really more of a "cut response" than an efficiency.  Either way, the "efficiency" (with quotes) is presented below as a function of track phi, eta, and pT.  For the calculation, I used primary tracks to which I apply the following QA cuts:

• Nfit > 15
• Nfit/Nposs > 0.52
• dca (global) < 1 cm
• |eta| < 1
• pT > 0.2

The pi0 and gamma-rich triggers are defined in the typical manner, and the hot tower list used in the calculation is attached.  To extract values, I fit the phi "efficiency" with a constant between -pi and pi, I fit the eta "efficiency" with a constant between -0.7 and 0.7, and the pT "efficiency" with eP*(1-exp(-sig*pT)) between 1 and 7 (eP and sig are free parameters).  Within statistical errors, the "efficiency" as a function of pT is identical between the two triggers.  The "efficiencies" as a function of phi and eta are very close, but not within statistical error between the two triggers.

Below are the track distributions before and after QA cuts: