Your perspective is mass, because you’re a weight weenie (and I mean this in the nicest possible way!), and for your wheels, you should be concerned about minimising mass at the outer edge of the wheel, to make acceleration easier, so putting heavy tyres at the extreme edge isn’t a good idea. If mass(tyre + rim) is the same in both cases, and both tyres have the same RR, then the heavier rim combo will be easier to accelerate and lower angular momentum.
However, I’m referring to rolling resistance, which can’t be accurately quantified in a lab (tyre running on a steel drum, which is what German mag Tour do in Conti’s lab.) as road imperfections have an effect. It comes down mostly to carcass suppleness, tread compound thickness/tackiness, tyre width & inflation pressure. The percentage differences in time are far far larger than the effects of bike mass & aerodynamics. It’s the single biggest bang for buck mod you can make.
I don’t subscribe to Tour, nor do I speak German, and their results are only applicable to riding on steel plates.
However, from some real world roll down tests, Pro3Races in a 25mm are quite good (better than 23 & 21, and if they made a 28 it’d be a winner), I’ll have to come back later with some other good un’s. I recall that Gatorskins are surprisingly good too.
You missed my ‘frictionless vacuum’ [1] assumptions, and it’s possible when you consider that a light tyre is ~200g and a heavy one 800+g, allowing a wide range of rim weights.
You’re reading too much into this H., it’s a theoretical argument, if you’re concerned about rotating mass then you’ll have light rims, tyres, veloplugs & alloy nipples.