It’s great to see this kind of content.  But we need to note that the lack of proper controls during testing almost certainly changed the results. Impossible to say which way, or how much. For e.g. the lug pattern is different between the two tires used with the different wheelsets. The tire that tested slower appears to have squared off lugs that protrude past the edge of the tire. The disc rotors (and hubs, and potentially spokes) were also not the same between the setups. Things like tire inflation level might have also played a role (it wasn’t mentioned if they were the same PSI); this could have exacerbated the differences in the tire profile exposed to the air (a more teardrop shape is likely to be more aerodynamic than a square shape even when they are both ~42mm). It is impossible to say how much, these differences played a role, but we’re talking about 8W. There isn’t much in that. It’s likely that they did.

There is a reason we hold everything else equal when engaging in the scientific testing process. It’s to remove exactly these kinds of variables from the results. Maybe Ben just proved that one tire is faster than the other, or rotor selection matters. We can’t say. Unfortunately, all we can say with any certainty is that one wheelset/tire/rotor/hub combo had a lower CdA than the other in this test. That frustrates me, because instead of providing clear answers around a question that many of us wonder about, the situation isn’t any clearer.