I don't think it's as black and white (xP) as you think to label GSC the most skilled or skill-intense meta and eliminating RBY and XY right off the bat. From my experience every tier needs a slightly different set of skills and abilities for one to perform well in, as well as just general knowledge about the competitive intricacies of mons in general (prediction, damage and odds calculation, reading a gamestate, creating a gameplan, thinking ahead, etc.). What you have to keep in mind is that there is RNG in this game regardless of the generation. Even if it's the "best" player in the world going up against someone who is just decent at laddering there is a certain chance that the favorite in the match loses, due to the amount of randomness in every single turn.
I guess that is one of the common misconceptions about how RBY is only hax and therefor doesn't require that much skill. I blatantly think that's not true, I see it more as the playingfield that both competitors go into being a lot rougher and a lot less forgiving. But I don't particularly think rby games (or especially series) are decided or influenced by luck more often than any other gen. The better player still usually wins, just like in any other gen. There are enormous win-rates among certain RBY players, just look at the enormous separation in results between marcoasd and any other player really, not only having scored the most points due to the amount of seasons he has played in but also easily averaging the most points per season:
http://www.pokemonperfect.com/forums/index.php?threads/seasons-results-spreadsheet.3992/
I guess what also plays into RBY being a lot more skill-testing than people giving it credit for is how matchup is almost completely eliminated and every series is entirely decided by the decisionmaking of the competitors and the rng of the game. Other gens add a certain element of matchup into it. I am not saying any tier is entirely matchup based, but there are significant advantages and disadvantages you can get way before the game even started. The only question that stands is if you consider teambuilding a skill linked to playing or separate from it. I personally am more leaning towards the latter, considering there's players who build and play well and ones that only do either particularly well.
I think there is a certain problem with only looking at the spl results as data, because those stats have a couple of issues: There is an incredibly narrow sample size (9 games or less per season for most players), they don't factor in things like quality of competition (all respect to people like Chaser, but a 10-1 run in rby spl wouldn't be possible these days) and there are certain players who seem to appear in every spl and many that come and go. Just look at the BW results for example. Soulwind not only has the most wins, he also has by far the most consecutive starting seasons in bw spl. This is not to take away from his accomplishments, it should just show how impossible it is to reach his spl results for anyone else, assuming Soulwind keeps up results like he did.
Another question that arises is: does a wide separation between results automatically mean the gen is more skill-intensive? Doesn't a pool of players where nobody dominates mean that all the players are similarly skilled, without actually having too many implications on the tier played?
I guess you can't really link things like individual players' results to the quality of the metagame played, because by that metric blarajan's XY LC run of 11-0 and 7-4 would mean the tier he played would be the best, which I honestly haven't heard anyone claim ever.
