As a fair comparison to the effect of the "new" cuts proposed on this blog, I'll make the same histograms using the same EEmcTrees, but with (as near as I can tell) the standard Weihong/IU cuts.  I'll also regenerate the "new" cuts plots using higher statistics.  51 runs are used for both the "new" and "old" cases.

Old Cuts

• pi0 p_T > 4 GeV (same in new cuts)
• M_gg < 0.6 GeV (same in new cuts)
• Each photon in the pi0 candidate must have at least 1.5 GeV
• Vertex Z position must be in +/- 150 cm.
• At least one tower in the event must have E_T > 1.5 GeV
• The E_T (using the origin rather than the vertex) of the hits must be > 1.5 GeV
• The central tower for each hit must have E_T > some Guassian function based on the phi and eta of the hit (parameters optimized using old Iron/Air MC)
• Total number of hits centered under the 6-18 towers used for computing the energy of the two hits must be no greater than 4.
• An "eta" computed by adding the radius of the vertex and the distance in the xy plane between the vertex and the position the pi0 would have intersected the Z=270 plane must be within 1.086 and 2.0.  (Note: if the vertex has very small transverse displacement, or if the phi of the vertex nearly matches the phi of the pi0, this "eta" is approximately the eta of the pi0 computed using the vertex.)

Note: IU traditionally had a cut on the photons being in the same sector.  This cut was forgotten in making the attached plots. 

New Cuts

(copied from earlier blog)

• photon energies > 3 GeV,
• pi0 p_T > 4 GeV
• M_gg < 0.6 GeV
• 4 cm < D_gg < 25 cm (D_gg is distance between the two gammas)
• M_gg > 0.005 [GeV] + 0.004 [GeV/cm] D_gg
• Preshower energy for the towers under the two photons each < 40 MeV
• Energy in SMD for each photon / photon energy > 0.012 for each photon

Note, there is no longer a cut on the photons being in the same sector.  [EDIT (06/26/2012): the M_gg vs D_gg cut is based on a bug.  Cuts now need to be re-evaluated.]

Results

The first page of the attached PDFs have the mass distribution for all pT bins, while the later pages show the results for each pT bin, just like those of the other blog.

Conclusions/Observations

Looking at the total number of statistics in the histograms, it seems the new set of cuts pass significantly fewer pi0 candidates--on the order of 1/2.  However, judging by eye, it appears the amount of "signal" is not much less with the new cuts.  The cut candidates seem to be mainly background. Furthermore, it seems (by eye) that the amount of signal in the TSP "new" cuts is about the same as in IU "old" cuts.