Hi all,

 this is the new plot for the Mx cut optimization

 http://www.slac.stanford.edu/~daniele/vub/toterrmax.eps

 here I have:

 - recalculated the B-D systematics. As you see, the new systematics (for
Mx<1.55GeV) is around 6%. In July was ~8%. This is due to the better S/B
ratio.

 - recalculated the stat error as a function of the Mx cut.

 - the theoretical uncertainty is not redone and it is the same than July


 As a consequence of the better stat error and B-D systematics, the
best cut is higher now. The optimal value is 1.63GeV.

 My comments:

 - in order to fix a cut we should recalculate the theoretical
uncertainty. The selection is changed, the resolution is better. We should
redo it. As far as the method is concerned, let's discuss it tomorrow.

 - what about to add all the systematics correlated with the Mx cut and
the background subtraction in that study? We could redo the
main systematics (tracking, neutrals, K+, Kl) for four Mx cut
(1.4,1.5,1.6,1.7) and find the dependency.

 - very nice total error on Vub (it should be 10%)

 Daniele