right and the most memorable long range attacking riding in years came on this stage
which everyone would totally have told you was a typical "grinder HC to MTF" "attack-in-the-final-1500m" parcours.
I think people are sometimes a bit too deterministic about routes and climbs - the Giro this year had Blockhaus which ordinarily is as selective as you can get, and yet it wasn't particularly exciting at all. Meanwhile somehow the Vuelta manages to luck its way into successive exciting stages at Formigal, which is about as unremarkable a cat-1 MTF as you can get.
This isn't to say that parcours don't matter but I do think it matters a bit less than some people might think. The TDF guys have been attempting to engineer exciting racing via the parcours for years now and about the only thing they've successfully done is make it so we don't have to sit through 5 sprint stages in the first week in a row, which is actually a good change.
Nah, Granon's too hard for it to be a final 1500m stage, especially after Galibier North, but I've said many times that I thought the Alpe d'Huez stage coming after it would mean that it would be a final climb shootout with Galibier used for attrition rather than what we got.
And of course the parcours matters - but it needs the riders to make it. The riders need the parcours on which to make the race, but the parcours needs the riders to make the race. When the riders don't have the parcours to make the race you get stuff like the 2012 Tour, where many of the best climbing specialists simply didn't bother turning up because the balance of TT vs. mountains was too much for them to believe they stood a chance (as I mention, you need enough mountain time to give the climbers enough to feel they can overcome their losses in the TTs, but ideally only if they pull out a climbing performance for the ages, à la Fuente '72 or Pantani '98); when the parcours doesn't have the riders to make the race you get stuff like the 2012 Giro, where some stages which were great on paper were dour because the riders simply didn't want to make the race.
However, the parcours is the part of the race that the organisers can control, and the parcours that they offer up often determines who shows up to the race (again, see why, say, Purito targeted Giro-Vuelta in 2012). At this stage in proceedings we don't know who will be starting in July, nor do we know who will or won't have form, get injured early in the season, get busted for doping, fall out with their teams, etc.. But we do know at least some of the details of the course, and as has been mentioned, there is a trend towards seeking lots of little bits of action, rather than a number of stages with major action. Sometimes major action turns up unexpectedly - and that's a bonus when it does - but the main focus, it seems, in recent years is not to make real queen stages or generate large gaps that require epic days out to overcome, but to try to minimise the days on which nothing at all happens, like, say, stage 5 in 2020 where there wasn't even a breakaway and it was diabolically bad.
This is also part of the result of the Tour's much larger appeal to casual fans than other races - the hardcore fans (that's us lot, we are on a cycling forum weeks after the season ends) may clamour for more TT mileage and real queen stages, but ASO don't need us. They need to ensure they don't run off the casual audience. In the 90s, real sprint trains were in their infancy, so although you'd get a lot of flat stages, late attacks and speculative attempts to foil the sprinters were much more common and much more likely to succeed; the sprint at the end was much less organised and slick, so there was more reason to tune in for those stages - plus they weren't broadcast for their entire running time, so you didn't have three hours of dead time between the break going and the break being reeled back, and the break would often get a much bigger lead in those days that made their chances of holding on much bigger anyway. Mountain stages, ASO can sell to a casual audience just by offering a few of the big names - Tourmalet! Alpe d'Huez! Galibier! Ventoux! It doesn't matter to the casual fan if all the stages are Unipuerto or suboptimal designs if they don't know any better, they just know these climbs are big and they're epic. But it does matter to the casual fan if they tune in and are bored, so ASO seem to be focusing more attention on making their flat stages less dull than they are on making their mountain stages the best they can be. After all, mountain stages draw the best audiences anyway.
And if we're honest, if that's their goal, then while it sucks for those of us that want to see epic stages and attacks from distance that can 90% of the time only be enticed with time gaps big enough to justify the risks... they're kinda succeeding.