We have observed that outter most 2 eta rings-(both side) have significantly low E/P values than all the other rings.

[add mean E/P vs ring-ID plot here]

So from the recent track-cuts study I have done (click here) I have come up with 3 reasons for this issue.

1) Low statistics than inner rings ( So any fluctuations in the BGs will have larger impact than they would have in inner rings]

2) Because of low statistics N_signal / N_low-E/P-BG ( Black / purple  regions I have marked below) 

3) Relatively small N_high-E/P-BG ( yellow ) [ these tracks belongs to high energy deposition in the tower  around the trigger thershold : [Energy of the electrons +  extra ebrgy from pi0,ets]  So why statistics of these tracks are less because  thees particle  hadrons (pions) which gives extra enrgy could be do not like  (:)) to fly through forward eta in the calorimeter depositing some of their energy or since they and secondory electrons from pi0   have to travel longer distance than before ,  they likely to  loose their energy and deposit only less amount than they would have deposit in inner eta rings.]

[ two terms I used above.
low-E/P-BG === low energetic / low momentum electrons which gives small E because of poor energy resolution
high-E/P-BG === triggerd , matched electrons to the tower, which gives larger Energy than its momentum  because of extar energy from  hadron + pio decay. ]]]

Heres some plots of ring 1.

Megenta -- is the new cut,, or in the bottom right plot its everything above +1.0 in "nSigmaElectron" scale.

To compare with above here's a inner ring. --ring 1D -12

So If you compare the 2 images of two rings  you can see the following.

     Ring ID 1                                                         Ring ID 12

1)  N_all_1                                               <          N_all_19           

2)  N_signal_1 / N_(low-E/P-BG)_1           >          N_signal_19 / N_(low-E/P-BG)_19                 

3) N_(high-E/P-BG)_1                               <          N_(low-E/P-BG)_19

Since outter rings has small  high-E/P-BG and inner rings contain more of these inner rings raises the tower Energy  from some fator and not in outter rings.

So this also cases outter rings to has small E/P.

Now to solve this issue we can modify our "nSigmaElectron"  we should simply tighten it to select events above 1.0 .  Its also kind  of tightning the new cut I have introduced in my other page.