Alright, here goes part#
2:
A new category I thought I'd add here in my second half of my response is U-turners/Volt Switchers, Several of which are incredibly prevalent in the metagame.
Volt Turners:
Genesect. Genesect is an extremely fast pokemon that generally carries a scarf and always carries u-turn. Genesect is a threat to a lot of sweeps, between being scarfed, having Download, and having incredible coverage. It can both revenge kill and pivot really well, and can cause problems for gengar, who is unable to switch in on any of its moves and can not kill it in a 1v1 scenario because genesect can u-turn before gengar even attacks (and the u-turn itself is not negligible with gengars defenses). Genesect may or may not be broken, but he is currently in the metagame and is therefore fair game to discuss as a problem, even if not technically a counter. Genesect is incredibly popular for entirely other reasons than gengar existing, and so doesn't contribute to overcentralization.
Rotom. Rotom-W specifically. Rotoms of every nature and every moveset run rampant in the metagame currently. It is a prime volt switcher, has a hard hitting hydro pump, and has access to trickscarf shenanigans, allowing it to both outspeed and neutar gengar, while its natural bulk (often invested) allows it to take at least one if not several hits from gengar, depending on the exact set of the rotom itself. Without scarf it volt switches slower than gengar, allowing a free switch in (with a bit of damage) to whatever you need. Being a slow turner as opposed to a fast turner like genesect, instead of saving itself, it allows a teammate to come in completely unscathed against gengar, which bypasses shadow tag almost entirely. Volt switch being on every rotom set in the metagame, this does not contribute to overcentralization, either.
Scizor. While scizor can not take two hits from gengar if scizor isnt specially defensive, scizor is another slow turner that has the added benefits of pursuit and bullet punch, two nightmares of gengar already stated in the first half of my response. Again, scizor is incredibly popular with or without mega gengar, and pursuit, bullet punch, and u-turn are all ou staples and nothing out of the ordinary.
Thundurus. Thundurus currently not being uber is a viable and powerful volt switcher that handles mega gengar incredibly well between priority taunt and volt switch, along with thunder wave at priority allowing it to cripple mega gengar for the entirety of the match setting it to 1/4 the speed it should have, allowing basically anything to outspeed it, shutting down its destiny bond and killing potential. These moves are nothing out of the ordinary on Thundurus, and he is currently ou, like genesect, and is so available for use and a valid and non-gengar dependant option.
Jolteon. Not the sturdiest of users, but with the same base speed as mega gengar and a strong, strong, attack able to threaten gengar before it mega's and speed tying mega gengar for the volt switch or even the attack (specs thunderbolt with a solid chance to ohko after rocks, volt switch doing a solid 61-72%). Gengar can noteably can not ohko jolteon, requiring two shots from either of his stabs, allowing jolteon to win any hard 1v1s against gengar not already mega'd and being a coiflip for an already mega'd gengar. Jolteon is OU, and these moves are standard, as is specs/life orb, life orb losing out on the ohko thunderbolt. Again, nothing out of the ordinary, standard set.
Crowbat is another noteable u-turner, but is oneshot by thunderbolt (56% chance, guarunteed after rocks). Crowbat does have base 130speed, tying mega gengar, and has both taunt and u-turn along with a choice band set that DOES ohko gengar if it wins the speed tie (using brave bird, 100% guarunteed ohko). This is another coinflip scenario (against the choice band set) but the odds are stacked in crowbats favor, with 50% chance for megar to go first and 56% chance to get the ohko, crowbat has an astounding 74% chance to win that matchup should gengar decide to stay. Again, these are standard moves, albeit in gen5 uu, but defog access has boosted his use in gen6.
Gliscor has u-turn, and is a slow turner (against gengar), but cant switch into gengar, as it does get 2hko'd, and generally does not run u-turn. However as a popular ou mon that doesnt get ohko'd it deserves a mention.
Landorus-T has a powerful u-turn and does require 2 hits, also having a commonplace scarf set can threaten to kill a mega'd gengar with eqarthquake or be a fast-turn pivot, along with stone edge should gengar not mega to avoid the eq. Common ou sets, common moves, threat to gengar and a solid u-turn user.
Defintion of a support/comparisions:
While gengar does fit the uber support role better than any other mon in ou so far, it is a fresh metagame and so may develop more, but as of yet gengar is the best candidate. However, a lot of more "traditional" supports have other options to contribute to the team, such as hazard removal, wish support, boost passing and status spreading. Gengar can not do these things, but most supports can't do what he does, he's got his own niche currently in the ou metagame, and generally benefits from an additional support mon on his team to help clear hazards which make gengar far less useful.
Deterrents to mega gengar:
Stealth rocks. Stealth rocks are gengars biggest enemy so far, as he needs to switch out and in at least once generally to do his job. Stealth rock damage allows many mons to nail a guarunteed ohko where it wasnt guarunteed before, put him in priority range if it's his second or futher time switching in, and generally just causes him to suffer, as gengar wants that hp to take *a* hit so he can clear his target, or to get up a sub, potentially more than one if he is forced out (sub blocking a pursuit trapper has been brought up in this thread and so I'm making it relevant here).
Status. Being paralyzed is incredibly bad for gengar, and often stops him from doing his job properly. Users like Klefki and Tornadus can paralyze him before he strikes them, allowing them to use futher moves (setting up screens or spikes in the case of klefki, or even swagger foul play shenanigans, and taunting or volt switching to something else in the case of thundurus) before leaving or going down (or even forcing gengar out). a paralyzed gengar can not even destiny bond sac itself and so stops perfoming its job as anything except maybe a perish trapper vs walls that don't hit back (blissey, infestation goodra, etc).
These two deterrents lead to another point, gengar is a support mon who can be reliant on another support mon. This either limits gengars performance or the way you build your team. While not necessarily strictly a con, this does restrict the way he is used in a lot of cases.
Prankster shenanigans. Gengar being a special attacker is quite vulnerable to the likes of t-wave swagger, as he can not really function, and getting lucky through confusion doesn't achieve all that much against most prankster users. Also having a weakness to foul play compounds the strength of users like klefki and Liepard against gengar. While neither of these really wants to switch in, they can both cripple mega gengar severely.
Choice users. Banded priority and scarfers are an issue for mega gengar, as his support role is generally to remove obstacles to a sweep, these revenge killer type pokemon can both handle the mon they would come in to stop and mega gengar. Things like scarfchomp, genesect, lando-T, banded scizor, banded azumaril, banded metagross do significant amounts of damage to gengar, meaning it cant switch in without being 2hkod before making a move, and it cant 1v1 the majority of these revenge killers. That being said, Gengar's job isn't really to take these mons on. Gengar can't really deal with them, though, and so it is a deterrent to using it as your answer to sweeper stoppers, requiring additional team mates to perform that job instead. I make this point to dispute all the "Gengar kills your wall and then sweeper x, y or z kill you." That kind of argument has been used a lot and ignores the commonality of revenge killers.
Over centralization:
While this may or may not be an issue, we don't have recent usage statistics to make solid arguments to this either way. We can, however, safely say that the presence of gengar must be considered with every team. That said, the same is true of Scizor, Genesect, Garchomp, Mega Kangkaskhan, Talonflame, Baton Pass Leads, etc, etc, etc. Any strong pokemon or pokemon type should be considered while building a team. You can't just arbitrarily jam pokemon together and expect the team to do well. Good teams have answers to everything, or at least answers to everything they need to answer- that's what makes them good.
Playstyle:
Again, with the freshness of the metagame and lack of recent usage statistics, we are limited to personal experiences on the ladder, which don't really hold much water as even someone who has played 100+ games has only seen a tiny fraction of the metagame. I can not comment on this with any amount of authority until smogon puts up more usage stats.
Sorry for the extreme length of these two posts (link to the first half:
http://www.smogon.com/forums/thread...on-read-post-383.3493175/page-35#post-5012992) but I wanted to give detailed and accurate responses so that the discussion remains coherent and reasonable, and so that my arguments have weight and will (hopefully) not get ignored/randomly naysay'd.
