Thanks for the quick response!

An intermittent wire would explain the unpredictability of the problem's appearance.

I'm not sure how to test the thermistor, though since the temperature readour continued to report values that seemed accurate, this seems less likely to be the source.

It was very weird watching the coordinates continue to change without the print head moving; it appeared the print data was still being sent and accepted as if the run was still in progress.

The combination of these two things particularly puzzles me.