First, adding options that are dominant options, is actually removing options--or at least removing gameplay. This is usually the case with most "ATs"--adding a new input that is strictly superior to previous options, at least in many situations. This is because by definition they are outside the normal balancing scope of the game.
Additionally, adding options is adding
breadth, which is the
exact inverse of depth.
To digress though, the community's definition of the word "tech" or "AT" has sort of diluted to the point of nothingness. Some people who call these explicit game inputs/features "ATs":
- Pivot Grabbing
- Short-Hopping
- Crawling
- Wall Jumping/Clinging
- Link's Hyrule Shield
- Zairs
- DI
- SDI
- Ukemi ("Teching")
Then you've got edge cases of normal inputs:
- Dashing U-smash
- U-Smash OoS
- Up-b OoS
- B-reversal
- B-turnabout
- Various pivots
Many people called those "ATs" as well, even though they are just normal inputs for normal actions. This is separate from input lenience exploits, which are legitimate ATs by a more formal definition:
- Shutter-Step F-smash
- Salmon Smash
- DACUS
- Boost Grabs
- Roll-cancelled Grabs
These things don't add anything to the game and are bad design, or at least unfortunate oversights--classic Cocaine Logic things. We could achieve the same results in a much cleaner and accessible way by making f-smashes/dash grabs have more range
all the time, and not only when you pass the secret execution test. (Or in the case of DACUS, by making dashing u-smashes better--both faster and with more control.)
Wavebouncing (B-reversal-turnabout) is a tough call that could be classified either way. You can make a fair argument that it is legitimately an ordinary input behaving exactly as everyone should expect. You can also argue that it is not very intuitive and is prohibitively difficult to execute for manly players (especially on neutral-b), and that maybe it's not an "AT" but the game would be ultimately better without it. I could buy either argument. (Wavebouncing doesn't really compete with other options, but it's frequently optimal over normal special usage at most spacings...)
You could pose a similar debate over wave-dashing, but I'm not sure I'd buy it.
You could also pose this debate on moves that have IASA before the landing-lag window ends, except I think we'd all agree that this just shouldn't happen regardless of semantics, making it a moot point.
Then you've got bonafide "tech" that everyone agrees on: janky, character-specific glitches the bestow some unnatural advantage:
- Palutena's Lightweight Cooldown Bypass
- Rush (Hitstun) Cancelling
- Item-Toss Hitstun/Landing Lag Cancelling
- Various (Arbitrary) Move-Specific State Cancels (Diddy neutral-b, Greninja side-b, Lucario up-b)
- Peach Floating-Bypasses-Stale-Moves
- Wario Wacky DI
You'll note that the only difference between "tech" and "bugs" is whether they help the character win or not.
L-cancelling is a unusual case, because it was put and left in intentionally, albeit with minimal documentation and no player feedback. Because Intent is not actually the true criteria in play, no one would debate that l-cancelling is an "AT" by almost any definition.
The bottom line, past all the semantics and dubious classification, is this:
What percentage of the game's outcome should be based on who can push buttons better?
Despite having pushed a lot of buttons in my time, I
empathetically believe the only valid answer is
0%.