I have been digging into the mysterious delta TOF01 feature; that is, when comparing the TOF for the extrapolated track from TKU between TOF1 and TOF0, I see a weird discrepancy between the measured TOF01 and extrapolated TOF01 for some tracks:

The periodic structure in the negative direction looks weird. So I changed my cut to select only those tracks in the weird periodic structure delta TOF01 < -1.5 ns. The events look very typical except for the position in TKU, where we get a nice doughnut shape:

Yum. So these are high radius tracks being mis-reconstructed as having momentum between 135 - 145 MeV/c. I plot measured TOF01 - extrapolated TOF01; so the extrapolated TOF01 is too large, i.e. they are reconstructed with too large a momentum. We have peaks in delta TOF01 like:

and so on up to about -20 ns. I think there is a reasonable case for peaks every 1.7 ns or so. The muon peak is at 3.2 ns relative to the electron peak, which is 28.6 ns absolute. So we can calculate the equivalent momenta between TOF01, which are listed below:

Not sure how auspicious or otherwise these momenta are. They don't look "special"...

Thinking more - I wonder if it is the track extrapolation at fault rather than tracker reconstruction? These tracks pass the TOF and tracker p cuts, so they are consistent from that POV... more digging methinks...