Hi all,

 as I mentioned in the friday meeting, the disagreement between data and
MC in the missing mass squared is only in the sample with one or more
kaons. I tried to switch on the kaon pid killing, since we were running
the MC using the montecarlo particle ID for kaons.
 Unfortunatly the results are not encouraging and the disagreement is
still there (actually it is a bit better but still there)

 in

http://www.slac.stanford.edu/~daniele/vub/rightkaon/allcomparisonmm2allcuts.ps
http://www.slac.stanford.edu/~daniele/vub/rightkaon/b0comparisonmm2allcuts.ps
http://www.slac.stanford.edu/~daniele/vub/rightkaon/bchacomparisonmm2allcuts.ps


 you can find the mm2 distribution after the cuts for all events, B0's and
charged B's. As you see the disagreement is just for Bcha (already known,
see my previous postings).



 Then, the remaning handle is a possible difference in the generetion.
Waiting for the reweighting from Alessio and Riccardo I tried to
understand which are the differences between neutral and charged B as far
as the missing mass squared distribution is concerned.

I must remind you that on data B0 and Bch distributions are similar while
large differences are present on MC.

In the following plots

  lepton cuts
  -----------

no normalization
http://www.slac.stanford.edu/~daniele/vub/mm2comp.ps

same area normalization
http://www.slac.stanford.edu/~daniele/vub/mm2comp_allcuts.ps


  ~all cuts
  ---------

no normalization
http://www.slac.stanford.edu/~daniele/vub/mm2compnorm.ps

same area normalization
http://www.slac.stanford.edu/~daniele/vub/mm2compnorm_allcuts.ps


you can see the mm2 distribution for 5 different categories on MC.

1: upper left:   B->Dlnu
2: upper right:  B->D*lnu
3: middle left:  B->D2*lnu
4: middle right: B->D1*lnu
5: bottom      : B->DXlnu


In red the charged B distribution is shown, in black the B0 one.
As you see the most evident difference is in B->Dlnu but also in B->D*lnu
is pretty different (actually all categories show differences).

For sure B->Dlnu has to be different since for B0 we have D+ and for B+
D0. These two D mesons decay in different charged multiplicities due to
the initial charge and then we should expect different resolutions.

For B->D*lnu we should have something different before the cuts. D*0's
decay in a soft photon or neutral pion while the D*+'s in charged soft
pion. Then the resolution should be better in the B0 case. We have a 30%
of D*+->D+pi0 but these decays can hardly explain such a difference. On
the other hand the situation should be worse
for B0 after the cuts, since the Qtot cut should worsen decays in a
charged soft pion.
Plots show that B0 are always worse.


I will try to repeate this study with similar multiplicity categories.


These plots show that the B->Dlnu reweighting can't fix the disagreement.
On the other hand the D decay reweighting could help.

 Daniele