Hi all, here I put the fit results for the new selection. As a comparison, in the second column I show the results with the old one (charged tracks + AS): http://www.slac.stanford.edu/~daniele/vub/fitresults.html Good news: - fit on the depleted sample for data is pretty perfect with the new selection - in general all fits on the depleted sample are fine - fits with multiplicity categories (not shown) do not show biases Bad news: - I see always a bias up with the new selection with respect to the CT+AS. - In particular the fit on the generic MC has is pretty different between the two selections I am studying in detail these effects. Daniele