> Especially if it's a PHEV since WLTP expects a full battery all the time. Under 2L/100km for a 2 ton SUV is not realistic by any stretch of the imagination. The numbers on https://www.spritmonitor.de/ for example seem to allign more with EPA in my experience.

There's a question as to what you're trying to model; if you're looking at daily usage then relatively short distances from fully charged are probably relatively representative of a lot of usage (the mean and median distances driven on a daily basis are relatively short distances!).

https://www.spritmonitor.de/en/overview/0-All_manufactures/0... shows that data is somewhat all over the place v. manufacturer data, with some being close and others very far away from it.

In reality, I suspect what would be useful is to have a further cycle based on a 100km journey largely on a motorway, as a long-distance extra-urban cycle?