Attached are two plots, one for pitch, one for yaw, and a spreadsheet with GPS times used to create the plots, showing the values set in ISC_LOCK state 110, DRMI_LOCKED_CHECK_ASC.  Yaw values are rather consistant, however if one looks only at the "locked without inital alignment" values, in yellow, the last bar in each group represents the last locking attempt that was successful at reaching NLN, and the value is lower than others in that group.  Pitch values are a bit more revealing, in that values over 1000 do not end with H1 in NLN.  The story seems to change around x = 20, which is likely an alignment change that was larger than the typical drifts.  After x = 20 the magents "no alignment, no lock, then initial alignment" shows a consistant reduction in the pitch offset after the initial alignment, blue, "initial alignment then lock."  At x=21 the pitch offset value is around 250 and locks, and at x = 33 the pitch offset value is around 200 and doesn't lock, so other factors are at play, and no absolute value can be confirmed with this sample.