We have seen a sizeable systematic problem in the distributions reconstructed with low candidates. This effect is illustrated in sc_mass:running_lsig; the measurement of the mass difference doesn't become stable until . For this reason we choose events with , a rather tight detachment cut.
Applying cuts of for the soft pion and gives a similar signal-to-background ratio for the as when we require . With this cut the systematic effect at low is not present. This leads us to conclude that the apparent systematic effect at low is not due to any physics, but instead may be caused by fitting distributions with large amounts of background.