Properly combine beamspot and track d0 uncertainty for flavor tagging
This does not affect standard flavor tagging code: we only use this for "fast" b-tagging that we're working on implementing in the trigger. Also the difference is pretty negligible so far as I can tell.
Anyway, it turns out that I'm supposed to add the beamspot uncertainty to the track d0 uncertainty to get the total displacement uncertainty, as described in the tracking group's twiki. Thanks to @dabattul for pointing this out.
Edited by Dan Guest