I think it’s actually very difficult to compare FGs which arguably had their glory days of fame in Arcades to RTS/MOBA games which came to define the wave of online competitive gaming when it comes to balance principle.
I do agree that when you look at a game like League of Legends, where they have over 100 Heroes to choose for games that are played in 5v5 team formats and they aim to release 1-2 new heroes every month that often “Balance” changes are incredibly knee-jerk and in some cases, are implemented based on performance trending instead of actual objective analysis (because like many have said, a few weeks or a few months is not nearly enough time to understand all possible matchups, strengths and weaknesses). But in a similar vein, the game is much easier to learn, there isn’t much mechanical skill requirement and one doesn’t need to develop unique timings or muscle memory to play, the same skill set is applicable regardless of which hero you opt to play with and as long as you’re accepting that there will always be imperfect balance (some heroes will always be better than others) then you can adapt and grow with a game style that promotes a changing meta and tries to give every hero a chance at the spotlight.
When it comes to a FG though, you do require significant mechanical skill to be able to play a character at the highest level, everything from understanding frame data and execution of incredibly frame tight links and traps and small balance changes to things like range, combo damage output, comboability, frames etc can mean significant re-work for the players playing these characters at a high level.
What I do disagree with though, is the argument that just because we are still learning about characters and matchups and discovering new tech that we cannot make adjustments in the meantime based on positive objective analysis. In FGs particularly, there will always be someone who is top tier and someone who is bottom tier, it would be incredibly difficult to have everyone be 5:5 with everyone else without the majority of the stylistic elements being lost.
A more aggressive balancing schedule with smaller more iterative changes could actually be beneficial to players having to adjust less whenever a new balance patch drops every 1-2 years and overhauls the entire game. It also allows developers to revert changes that cause unexpected behaviours, for instance, if we see that Cammy is undeniably top tier due to her combo-ability into big damage and relatively safe options, we can make small changes to her combo damage and see if that affects her matchups without necessarily altering her play style, available tech, or existing knowledge requirements.
That’s not to say that all things can be balanced by damage tweaks or stun/vitality changes, and I absolutely agree that any significant alterations that force people to relearn a character needs to be well thought out over a longer period of time with more ample understanding of a character.
Some unfortunate circumstance is that some characters will never be immensely popular, and will never have the same high level representation for us and the developers to truly understand the characters and the matchups. It’s very easy to “Balance” Ryu because we see so many Ryus, we know exactly how he plays and what his tools are and what he can do in the matchups at the highest levels.
It’s considerably harder to balance a hero like Gen, who has tons of tools but sees very little play, or a hero like E.Ryu or Oni, who’ve been around for a much shorter period of time and don’t have many serious adopters yet (or might never).