As I said, the metagame shifts as it will. Garchomp leaving the metagame is obviously a big change, just as Latias entering it or Scizor emerging as a top threat. With one threat no longer needing to be accounted for, some pokémon will obviously change their movesets to accomodate the new threats that emerge.
The shift from Ice Fang (and Earthquake!) to Stone Edge on Gyarados makes sense. Only Ice Fang worked on Garchomp, so you had to use it that to beat it. With Garchomp gone, Salamence is now the next biggest threat. Salamence is vulnerable to Stone Edge as well... and Stone Edge also allows you to catch other Gyarados, who still remains a top threat! Why not run that and kill two birds with one stone? It's a perfect example of a very simple metagame shift dictating a very simple moveset change. Nothing else about Gyarados changed -- in fact, its use as a defensive pokémon has gone up over time, opposite of expected with Garchomp leaving -- it just swapped one move out for another to deal with the threat that replaced Garchomp. Perhaps it's also because Water/Ice is a generally easy-to-wall type combination by most other Waters and it simply fell out of favor. *shrug*
Metagross never used Ice Punch much to begin with so, while its usage may have gone from "little" to "almost none," it doesn't really matter that much. Metagross never carried Shuca Berry back then either, so it would've had no business fighting Garchomp anyway. If you really care about such a pointless change in usage, though, the emergence of Metagross as a strong lead is a more likely cause of Ice Punch's drop. Metagross was a mere #19 in lead usage back in August 2008, where these days it swaps back and forth with Azelf as a top lead. Leadgross never carries Ice Punch.
Blissey usage these days is split almost 50/50 between Softboiled and Wish as a recovery move, with Protect often accompanying the latter. This takes up an extra moveslot which Blissey would otherwise be using for Ice Beam much of the time, even now, to break Gengar Subs and combat Salamence. Wish Blissey did not exist back then, so they were more able to make room for another attack.
Indeed, Bronzong has seen a large decrease in usage since last year and I even admitted that Garchomp's removal is probably a big cause of that. That said, there's more to the story. Bronzong actually stayed in the 10-20 usage range for a long time after Garchomp left, only in the past three months has it taken that huge plummet down to #30. Magnezone's steady increase in usage over time is also a possible cause for its drop, since Bronzong can ill afford to drop its Leftovers for a Shed Shell. Yet it's still very good against a number of other OUs, as you stated, so maybe it's partially one of those mysterious things that "just happened" like with Heracross falling.
I'll even give you a freebie. Jolteon's most popular item currently is Choice Specs. Obviously, Specs Jolteon was nowhere to be found back in Garchomp's day. Similar to Gyarados' shift toward Stone Edge, it's just fucking common sense you don't want to lock yourself into a Choiced Electric attack when the top threat in the game is a Ground pokémon. Its pseudo-replacement, Salamence, is neutral to Electric and so is the current #1, Scizor. Specs would probably lose favor on Jolteon if Garchomp made its return, or at least it'd be dry BPing more often just to be safe, and you might also see a slight decrease in Trickscarf Rotom-A and stuff too. It's not really a cause for alarm, it's a very basic adaptation to a changing environment.
But were there any drastic changes? I would say not. I suppose one could say otherwise, though, since what constitutes "drastic" comes down to opinion. Did the general usage pattern in OU suddenly change? No, the same pattern endures. Even the same pokémon are largely used at about the same levels, where the only two that Garchomp might have had a notable effect on moving are Bronzong down and Jirachi up. Even then, it's surely not the only reason those things happened. Did Ice attacks in general fall out of favor with people? Please, the coveted "Boltbeam" predates Garchomp by ten years and Ice Beam remains a top five most common attack.
For the most part, things stayed the same and only minor changes occured, which is expected with the change of metagame. You just removed a pokémon from the metagame, of course movesets with change to accomodate the fact! Random fire moves on pokémon that generally don't carry them are a lot more common these days too as a result of Scizor becoming a major force. Does that mean Scizor is overcentralizing and should be pushed to Uber? No, not really. That's as basic a response to the metagame shift as Gyarados increasing its usage in Stone Edge from back in the day.
I would say that 15% drop is a huge margin. It means that 15% less teams had a Gyarados, (and infact most of the pokemon on that list have dropped by that much or more) which is pretty significant when you consider that is Gyarados dropped 15% nowadays it would be nowhere on the usage charts.
If most every pokémon on that list have majorly dropped, then how does that say anything about Garchomp at all? I could see your point if those pokémon were largely considered good Garchomp checks (but Garchomp is actually considered the counter to like half of them) and suddenly they were replaced by new pokémon that Garchomp had previously been holding back. That's not the case at all. Yeah, so maybe the metagame was centralized around Garchomp quite a bit. But it was also greatly centralized by Gengar, whose heavy usage could hardly be attributed to Garchomp at all. Does it not revenge Salamence just as well as it could have revenged Garchomp? HP Ice is still 4x, Shadow Ball is still neutral, Gengar still outspeeds it without DD (or at +1 with Scarf). Does it not still Tbolt Gyarados into oblivion, who remained a same threat it always was? Please, tell me how
every pokémon enjoying higher usage proves that
only Garchomp was centralizing.
Gyarados is actually used more defensively these days than it was back then, being a vital cog in many a curreny-day stall team and just more often investing in
some physical bulk than it used to. If Garchomp was so centralizing, shouldn't that be the opposite? The better play would be to use Gyarados to check your opponents Garchomp, then set up your own and win. Garchomp is banned, well now you just kinda gotta deal with it and use Gyarados itself to sweep.
The exact opposite happened. Granted, this is somewhat off-set by the fact that Ice Fang usage decreased. Cool, generic metagame shifts. Ice Fang used less often due to Garchomp no longer being a threat, increase in defensive sets do to new pokémon becoming threats. Talk about centralization! :justin2:
Besides which, arguing that
every single one of those pokémon having use against Garchomp doesn't help your cause at all. Just a couple pages ago, I was told I needed half my team to be Skarmory / Scizor / Latias to check in on Suspect currently and now you're telling me I can just check it with all these other things as well? If that many pokémon have a potential utility to fighting Garchomp, how the hell can it be broken? A pokémon can't very well effortlessly sweep a team if everything under the sun can potentially check it.
I also truly believe it was Garchomps fault because it evened out once he left OU, more pokemon are now used less often, there is less CENTRALISATION.
Really? How much did it really even out?
Currently, Scizor is #1 in usage by a significant margin over #2, where usage evened out. The first pokémon with less than 10% of Scizor's usage is... Crobat, at #50. Hey, that's OU in a nutshell!
Garchomp was #1 in usage by a close margin over #2, Gengar, after which usage evened out. The first pokémon under 10% of Garchomp's usage was Umbreon, at #52. What do you know? Old-day OU was about the same size! (Actually, it was even larger since that was about the time we started dealing with tiers the way we currently do.) The only difference is that usage percentages ranged from 50->5% instead of 30->3%. It's a different scale but they're proportional and functionally identical. What does it matter either way?
Anyway, I'm going to bed now. Stop spouting wrongness long enough for me to get a good night's rest.