Known issue with replay of crawls. Correspondence with Archive-It indicates that this is an issue with replay in WaybackMachine, not with the actual crawls.
“The issue in this case boils down to the way that the original site appends unique ID strings to each request for the course information shown. The ID generated by the archived site does not match the original from crawl time, so we end up with pieces missing…I think that it will be necessary to teach our Wayback replay software to ignore these unique IDs and just load the file with the closest matching URL. “