Your not always going to have the meter to OS out of pressure so no one is going to armor out of your pressure every time you try to do it. but with the way some offense is I think its a good way to keep your opponent in check even if you only do it once or twice it makes them think about it at least.
Well, that would be one reason not to try it, for that case design of the "OS" doesn't matter.
For the rest cases, I think it's more intuitive and less taxing to input a reversal where you think a hole will be than to input it everywhere and effecively cancelling it everywhere except when you think your opponent is going to bait it. For the record, baiting it can still be done in a game with no reversal OS, but when you have OS, another layer of "yomi" is lost to it. This design removes some depth from game of reads when you have meter IMO. That, and it makes default level of inputs required a lot more of a hurdle.
I mean, that is an option after all, and one can't say that it is not beneficial for defender, but then you get situations where gap between characters with strong reversals and others increases, and then a case can be made that based on practical results (that will inevitably include heavy OS usage) offence can be built around existing OS and thus will require it to reach sensible odds of defending yourself successfully - and having a character who can use this technique to the best effect.
I doubt that this will increase variety of competitive matches, while still hurting casual side of the game (for what this point is worth in such a discussion).
You guys can't see the actual problem.
The actual problem is that we have a billion variations, half of which are not balanced to the rest.
They will not be, because it's hundreds of pairs to match up to one another.
They did this as a lazy solution; they can remotely push and shove the so-called balance but high numbers are both a tool and an excuse.
Option selects do not matter. The fact is that the actual game balance hinges on numbers of variables in the system.
Ergo a handful of these characters will always be "broken" aka overperforming.
You have a point, but what's your suggestion when it comes to a single particular issue? Ignore it because the game will be not balanced regardless?
I think I should point out that we discuss a concept itself here as much as balance implications of it. As in, is this a good baseline for system and character designs? Will it affect enjoyability of the game regardless of balance? Will it serve towards decreasing spread of power / increasing spread of outcomes in particular situations? How will it affect the game?