Whether that has anything to do with systemd, I don't know. It shouldn't, but as desktop environments do rely on udev to detect flash drives and discs, and udev may now have dependencies on systemd, maybe systemd is the root of those problems.
From the behavior, I doubt the problems you're exhibiting are due to systemd. Sounds like they're happening in user land which isn't really systemd's area (it's starting daemons & what not). That's my *guess* and nothing more. That deleting dot folders out of your home directory changes behavior strongly suggests that to me, however.
Since you mention udev, could it possibly have something to do with device ids changing, or even missing, in unexpected ways between boots? I have problems with that on my home Windows desktop. Well, maybe not exactly that, but after a soft reboot, I virtually never see my boot drive detected by my SATA RAID controller (boot drive is an SSD and I've a RAID5 for data & holding VMs). A hard reboot always fixes the problem. At first I thought it was a problem with the SSD, but the problem has happened with several different models & manufacturers, so I'm suspecting firmware on the RAID controller now (unfortunately no updates, and I seem to be the only one exhibiting the problem).
Don't take this as either an endorsement or condemnation of systemd. I really don't know enough about systemd versus init.d to contribute anything worthwhile to a pros/cons discussion. All I've observed is that it seems to be an old man vs. kids battle style battle. i.e. the kids are changing stuff and the old man proverbially yelling at the kids to get off his lawn (this analogy is not intended to make any assumptions or accusations regarding the age of the individuals on either side - for all I know systemd is being driven by the same people that came up with init.d and thought their first effort wasn't what they wanted or needed and tried a different approach).