Hi Matt. This looks like it's a problem with the mount, it's not being caused by PHD2. Here's a look at one of the worst-case guiding sessions (RA in red):
These occur roughly on 10-minute intervals which may correspond with the worm period. Another view shows how bad it is, this one with guiding corrections removed:
All of this occurred while the scope was on the west side of the pier. Once you did the meridian flip, the tracking performance improved significantly and these very large excursions disappeared until around 00:24. The meridian flip would presumably have rotated the worm gear to a different position so there may be a clue there as to where the problem lies. You should look to see if this side-of-pier dependency has also appeared in earlier sessions. Of course, having the RA drive system too tightly meshed is always a possible source of these kinds of problems.
Good luck,
Bruce