As someone who possesses extensive experience in both Übers and Anything Goes, I would like to bring up some stuff in this thread to aid decision-making, mostly by clearing up uncertainties regarding what would happen to the Übers metagame if certain clauses, such as the Species Clause and the Mega Rayquaza Clause, were to be lifted:
Regarding Mega Rayquaza:
1. In the opening post of this thread, the following is written regarding what the Übers metagame (the ORAS iteration of such, to be precise) would look like if Mega Rayquaza was allowed:
Mega Rayquaza was a terrific Pokémon due to its amazing offensive stats. In terms of damage output it was the undisputed best. Its qualities were undoubtedly best utilized on offensive teams. Using an offensively based team with Mega Rayquaza was considered superior to not using it. Hence, the only rational choice for players was to adopt it, making every viable team in the tier some sort of Mega Rayquaza offense. While both players could use it a make gradual optimizations in this pseudo mirror-match up, Ubers was deemed unenjoyable to play. Using the specified criteria for banning Pokémon from Ubers, it is clear that Mega Rayquaza was broken, thus it was ban-worthy.
I would like to point out that not only is Mega Rayquaza hardly a necessity in Anything Goes (in ORAS, SM, as well as USUM), but even among teams that contain Mega Rayquaza, not all of them fall under the playstyle of offense. In fact, in pre-Marshadow SM Anything Goes,
Thimo and I even made a quite decent
stall team with Mega Rayquaza on it. With or without Mega Rayquaza, every single playstyle has always been viable in Anything Goes.
2. However, it should be noted that I am unable to perfectly quantify exactly how much the managability of Mega Rayquaza in Anything Goes is correlated with the absence of the Species Clause in that metagame, most notably in the way that it enables teams to contain two or more Arceus. In other words, I am unsure of exactly how much less manageable Mega Rayquaza would be if a Species Clause was suddenly implemented in Anything Goes, or if Mega Rayquaza was unbanned in Übers without the Species Clause being lifted at the same time. However, it should be noted that, in both ORAS as well as pre-Marshadow SM, I have, on at least one instance, created an extremely solid stall team that was actually Übers-legal (in other words, it broke neither the Species Clause nor the Mega Rayquaza Clause).
3. At the end of the day, there is, however, no denying that Mega Rayquaza is insanely powerful, and there is no doubt that existing Übers teams would need to be reconfigured if this Pokémon were to be unbanned, regardless of whether or not the Species Clause is lifted.
For example, in a metagame in which Mega Rayquaza exists, Lugia would be forced to run 252 HP and 252 Defense EVs, along with a Defense-boosting Nature. This is because a Lugia that runs 252 HP and 160 Defense EVs, which is standard in Übers at the moment, would not be able to switch into an Adamant Life Orb Mega Rayquaza, even with Stealth Rock off the field. This is because a Lugia with such EVs would be taking 27.6 - 32.4% damage from a Multiscale-weakened Dragon Ascent on the switch, before it is outsped and pummeled by a second Dragon Ascent, which deals 55.2 - 64.9% damage. Over time, the Dragon Ascents would be outdamaging Lugia's Roost + Leftovers recovery, resulting in it eventually defeated. However, choosing to run no Speed on Lugia would make it weaker to other Pokémon, such as certain Ho-Oh, Primal Kyogre and Primal Groudon sets.
It should also be noted that in a metagame with Mega Rayquaza, defensive teams that run Lugia would need to have a specific switch-in to Choice Band Dragon Ascent, since even a Lugia with 252 HP and 252 Defense EVs cannot switch into that move, even with Stealth Rock off the field.
4. However, it should also be noted that if Mega Rayquaza was usable, then Mega Lucario - currently a prominent threat in the Übers metagame - would see almost no usage, especially if the Species Clause also happens to be lifted (just imagine Mega Lucario, a sweeper slower than Arceus, in a metagame in which teams can have multiple Arceus, most or all of which would most likely be EV'd to outun Mega Rayquaza). So the introduction of Mega Rayquaza (and if not that, then the lifting of the Species Clause) would result in teams not needing to be prepared for Mega Lucario, creating less restraint for teambuilding in a way.
Regarding the Species Clause:
In Anything Goes, the only common way by which the Species Clause is broken is when teams contain more than one Arceus. Sometimes, people use teams with six Arceus, but more often than not, teams do not include more than three, with most teams using two, from my experience. Teams that break the Species Clause in other ways, such as containing six Xerneas or Primal Groudon (as I have seen some people on Discord or Pokémon Showdown! talking about) are suboptimal. In any case, I personally see absolutely no reason to
not lift the Species Clause, and I have yet to see anyone coming up with any objective argument for keeping it. This also leads to my next point...
Regarding Primal Groudon:
Primal Groudon is a Pokémon whose usage and dominance would decrease dramatically if the Species Clause were to be lifted. This is because Arceus Formes can do many of the things Primal Groudon can do, such as setting and keeping Stealth Rock up, better, while at the same time being faster and having access to instant recovery. Perhaps the biggest reason why Primal Groudon is by far the most commonly-used Stealth Rock user in Übers is not because it is particularly good at this role, but simply because of the fact that it is by far the most splashable, since teams can only have one Arceus, which is usually used for another role, such as Defog, Swords Dance or Calm Mind. However, if teams could have more than one Arceus, then this completely changes. Primal Groudon would still have uses though, such as for checking Primal Kyogre (something it can do better than any Arceus Forme, especially if it knows Rest) and checking Xerneas (which only Arceus-Poison can really do, among Arceus Formes).
Regarding "uncompetitive" or "luck-based" elements of the game:
Many things are banned from Übers on the grounds of being "uncompetitive" or "luck-based." These include things that introduce a large amount of RNG-based luck into the game, such as one-hit KO moves, Moody, Hypnosis Mega Gengar, and evasion-boosting techniques. But these also include things that make the game more team matchups-based, and therefore luck-based, by creating common situations in which teams either win or lose at Team Preview. The latter category includes Baton Pass-based teams, while many people also seem to be wanting Gothitelle to be banned for very similar reasons.
Firstly, before I get to the main point, I would like to point out that absolutely
none of the things that are banned from Übers for being "uncompetitive" or "luck-based" see common usage in Anything Goes. The amount of usage Mega Gengar gets in Anything Goes already pales in comparison to the amount of usage Mega Rayquaza gets, but even when looking at the Anything Goes teams that do run Mega Gengar over Mega Rayquaza, the Mega Gengar rarely run Hypnosis. One-hit KO moves see almost no usage, and most of the usage that it gets comes from people tacking Sheer Cold onto Choice Scarf Kyogre (which itself is already incredibly rare) on the off-chance that they need the move when facing a bad team matchup. Moody and evasion-boosting moves also see almost no usage outside of Baton Pass teams. And regarding Baton Pass in Anything Goes, there is a total of one prominent player who uses Eeveepass, as well as a total of one prominent player who uses full Baton Pass teams.
Why do such things have such low usage? The simple answer is because they are inconsistent. What good would something that allows you to defeat even the best player easily, given the right RNG and/or team matchup, if you cannot win consistently with it, and can easily lose to anyone with it, which leads to you attaining an unsatisfactory win-loss record in the long run? Maybe for the purpose of cheesing a much more skilled trainer in a single battle in tournament play? Well, that leads to my main point, which is also by far the most interesting point I intend to bring up in this post:
Is it
actually logical to ban things for the luck and randomness that they introduce into the game? Think about this:
It is true that, either with or without one-hit KO moves, Moody, Baton Pass, or any of that luck-based stuff... Pokémon is a game in which anyone holds the possibility of defeating anyone else in a single battle, or a best-of-three set, as even without the aforementioned factors, this game still has a massive amount of random factors in it, such as RNG (freezes, critical hits, fully paralyses, etc) and team matchups (say a stall team VS. a Gothitelle, or a Whirlpool + Perish Song Arceus-Dark). But at the same time, in a scenario in which skill is attempted to be measured not in a single battle or best-of-three set, but rather, based on each trainer's consistency, embodied by their win-loss record over, say, a thousand battles against many different trainers and teams in the whole metagame (which I honestly believe to be the only real way to measure skill in Pokémon, but that's another story), then even
with one-hit KO moves, Moody, Baton Pass, and all that stuff allowed... the better players would
still almost always end up with better win-loss records than the worse players, since all forms of luck (RNG + team matchups + possibly more) more or less even out over such an incredibly large sample size, due to
the law of large numbers.
So since the better players are going to get better win-loss records than the worse players over incredibly large sample sizes, such as a thousand battles, regardless of how much luck and randomness exists in the game... then why would there be any need to ban anything on the grounds that it makes the game more luck-based and random?
I can only think of one reason, and that is to cater to tournament play, in which skill is attempted to be measured not through an absolutely gigantic sample size, but rather, in best-of-three or five sets, in which luck plays a far larger factor in deciding who wins or loses, and overall consistency in the long run can easily be rendered irrelevant. However, even without one-hit KO moves, Baton Pass and other similar stuff... tournament results, as they are, are
already extremely easily influenced by luck, in terms of both RNG and team matchups, and this especially held true in post-Marshadow SM Ubers, in which Gothitelle was everywhere, and made team matchups an extremely important deciding factor in battles - probably even moreso than hax/RNG.
So... if the logical conclusion is that Pokémon is a game in which skill can
only be determined over a large sample size, and yet a large sample size would result in the better players attaining better win-loss records anyway even in spite of the existence of luck-based elements... then what exactly would be the point in banning luck-based elements in the first place?
A general note (which does not reflect my own point of view):
My general observation of the Anything Goes community tells me that Baton Pass (especially full Baton Pass teams) and Gothitelle are really the only things people commonly complain about in that metagame. I have never heard any complaints about Mega Rayquaza, multiple Arceus, or even one-hit KO moves, Moody, and other luck-based stuff in that metagame.