Genuine question: could this data be affected by the location of VPN exit points?
Absolutely a possibility. Depends on how often people are using VPNs, whether there’s a distinct difference between political ideology and VPN usage, and the average distance to a VPN exit point. I think it’s fairly safe to assume a similar distribution of VPN usage across political bounds and generally speaking geographically distributed fairly evenly, with perhaps major cities and higher educated areas using VPNs slightly more often but with the exit point also being closer to the true point of access.
If anything I would expect vpn exit points to be located in higher density areas where presumably internet connectivity is cheaper, and higher density areas tend to be more liberal
Not just VPN, but geolocation in general. I am in Ohio, but I am often geolocated as being in Chicago due to my ISP. Similar for mobile.
Keep in mind that many red states have attempted to essentially ban online porn with their overbearing age-verification laws. A savvy internet user would be less likely to pick an exit node in a red state, given the choice.