I think there needs to be clarity on this thread. As I am seeing a lot of ambiguity here. The issue of IP change varies.
The fundamental issue should be the unusual variation in the geo-location distance usually been used.
Obviously, the internet assists to connect people from vast number of locations, but if the change is within the parameters of the usual geo-location, there would be no issue [city/town/country], it depends.
A lot of users do not have static IP, so it is almost common for their ip to change every once in a while, to which the system via data analytics has a range to go by, but when a fresh login far out stretches the usual logged data, then a potential flag will be raised on the account. This is a similar system implementation used by email service providers. For those who travel a lot would have most experienced this when logging into said mail provider from a possible new location [thousands miles away || another country /continent]. It is almost imminent you would be asked to verify your identity that the new login is indeed the account holder and not another person who has gained access to said account.
Now back to the topic at hand, Yes! The underline term is ip change, but the issue resides on how much change from the recorded/ estimated log and a new login.
Looking from the very first post, I was a member on majority of those tracker listed when they were in operation and moved around geo-location wise and never had an issue and sharing with my brother. More so, what people fail to consider for the archilles heel for a ban after an IP change from a fresh login is what function(s) are performed. For instance, changing the user settings/details = is 100% a RED FLAG (absolute no brainer).