# The annoying gap problem

### In the dropbox link: https://www.dropbox.com/sh/75arjmb0wu07db6/AAAvUjPOWP9PL4g7c93JeAMpa?dl=0
+ Correct ptl weight is original one: (eff/MR/sigMix: nTrk6)
> having gap in onres thrust

| beam | thrust |
| --- | --- |
|  |  |

+ FINALCHECK_full (eff/MR_ntrk6 sigMix_ntrk12)
+ FINALCHECK_full (MR_ntrk6 eff/sigMix_ntrk12)
> having gap in long range onres beam

| beam | thrust |
| --- | --- |
|  |  |

(* ntrk12 onres are shown)
### The observation can be summarized in below
+ The main effect is on the efficiency correction actually. When using full ntrk12 to correct, beam's long range would start to have gap, but thrust would get closure.
+ NB: tgen in those plots are almost no change, this can be compared in 1st, 2nd folders. (sigMix is not the culprit)
+ Beam:
+ Thrust:

And why the bizarre thing happens:
I think it is because the efficiency correction factor is working good on the signal function, but for the background function, it will tend to **over correct(?)** in some region. And most annoying, the over correct effect happen in different region in beam & thrust ana.

Below I show the ratio of reco/tgen:

|| sig | bkg |
|---|---|---|
| beam | |  |
| thrust |  |  |

TFile *_file0 = TFile::Open("BELLE_reco_ntrk12_onres.root")
TFile *_file1 = TFile::Open("BELLE_tgen_ntrk12_onres.root")
_file0->cd()
TH2D* signal2PC_0_14_999_0_0_tgen = (TH2D*) _file1->Get("signal2PC_0_14_999_0_0")
TH2D* bkgrnd2PC_0_14_999_0_0_tgen = (TH2D*) _file1->Get("bkgrnd2PC_0_14_999_0_0")
signal2PC_0_14_999_0_0->Divide(signal2PC_0_14_999_0_0_tgen)
signal2PC_0_14_999_0_0->Draw()
signal2PC_0_14_999_0_0->Draw("LEGO")
bkgrnd2PC_0_14_999_0_0->Divide(bkgrnd2PC_0_14_999_0_0_tgen)
bkgrnd2PC_0_14_999_0_0->Draw("LEGO")