Climbing a single big hill takes totally different requirements than climbing 100km of hills at varying grades. The first might not include as many descents. The second always has a descent after the hill climb. I’m about to do an experiment involving this difference. My thesis is that getting slightly heavier (and cheaper) wheels that are much more aero is better than ultralight wheels that are not as aero. The idea is that even through saving a few grams of weight will get you up the hill faster, the loss in aero advantage on the descent counters those gains later. In fact, the single best thing to improve speed over 20 kph is an aerodynamic position and aero adjustments in the bike of which an aero wheelset is huge compared to other bike aero improvements. Over 45 kph and 90% of all power output is to overcome wind resistence at that point, but that’s obvious, you can hear it. So let’s say you have a course with a lot of steep climbs that have 3% descents after the 9% climb. The clear winner is the person with the most aerodynamic savings (who can keep pressue on the pedals in the descents while staying aero).
But how much are aero wheels savings? Considering that the entire aero efficiency is only 30% bike (rest is position, which hands on hoods, forearms parallel to ground being most aero body position, not the drops). At most it will save about 4 minutes for a 40 km ride.
But there’s another reason to swap out my gravel wheels. Rolling resistence from having those large tires loses power over a 100km ride. In fact, 35mm vs 28mm is about 60 watts wasted. That’s a huge amount and does not even account for the aero advantages.