Rolldown aero testing?

I am curious about undertaking aero testing by doing rolldowns of a hill and measuring speed. point is not to say it’s better than wind tunnel or chung testing…but if i wanted an attainable way to tease out qualitative (i.e., NOT quantitative) differences between variables, is this reasonable/worth my time? i searched for other threads couldn’t find any. sorry if i missed.

I have a no traffic, straight, smooth pavement hill near me. averages -6.1% but is steeper at bottom so speeds get higher. takes about 1 minute down, clocked it today from a standing still no pedaling rolldown, i get to 42 mph at the bottom. hill is short enough that i could do many repetitions during one session, and there’s a parking area right at the bottom so i can easily bring supplies/make setup tweaks right there. I’d use a wheel speed sensor, start exact same point from standing, no pedaling, and measure max speed at a known point at the bottom (there’s a driveway i can use as fixed speedcheck point). i’d do it the same way as wind tunnel testing, only change one variable between comparison runs/sets. compare max speed at that fixed point of all runs and see if can see any differences/trends. obv like chung testing would pick a day with no wind, etc.

at very least, feel like it could give me reasonable confidence with setup decisions and whether A or B is likely faster. the only downside I can see is that it would not give me any wattage or cda measurables as to how much faster/slower, and it’s just a plain yes or no answer, but 100% i can live with that given how easy and repeatable this testing could be for me.

thoughts? tips?

Sounds like fun, go for it. Jan Heine from bicycle quarterly and Rene Herse is one source for all sorts of roll-down protocols since. One of many references here: