KowtowRobinson
[10] Knight
Force your stance mixup, and save a fast 3b for when she tries to 2a you out of it then, would be my best guess.
Hate Speech: Theory Fighter University: Remedial Math
- Thread starter Dr. Hates
- Start date
Force your stance mixup, and save a fast 3b for when she tries to 2a you out of it then, would be my best guess.
KowtowRobinson
[10] Knight
I see what you did there, lol.
sillycatsteven
[09] Warrior
<_<
>_>
*_*
@_@
>_>
*_*
@_@
Heaton
嶋野の狂犬
Still gotta disagree on that.
For SRSH B/K mixups, both options give you similar damage, around 60/55 respectively. However, there's a few problems.
- Even though the 6B+K effectively becomes an i10 transition, SRSH B and SRSH K are i18 and i19, respectively. Since a good player KNOWS that those are fastest options out of the stance, she can just mash on BB and beat both options. However, the stance still takes 20 frames to get into, regardless of her advantage - they still roughly 40 frames before impact to recognize that I'm in SRSH and mash out BB to beat me. There's no need to try to beat the entrance with 2A.
- For argument's sake, let's say they are slow and let me get the SRSH mix-up. The second problem is that both moves are insanely punishable - Sophitia gets a TAS B after blocking either options.
And now every Sophitia player going to NEC is practicing how to TAS B a blocked 1B...
damn-I-Suck
[12] Conqueror
Good speech hates, as always.
However, I still hold to the idea that crunching numbers, while it may give you a nerd-gasm, can't replace experience. Some punishers I only discovered by doing them in-battle, mostly by randomization when I try to get a feel for some characters. Not saying I alone discovered them (someone else probably figured that out long ago), but figured it out on my own.
Muscle memory/eye hand coordination will carry you much much further in SC than crunching numbers would. Most of the moves, there was no need to look at frame data because I pretty much already "knew" how fast/slow it was. There was no need to know "exactly" how bad a match-up between Sieg and Sophie was, as experience already told me how bad it was.
But back to the point, I don't think it's reasonable to take out the human equation. For example, by numbers alone, Sieg looks like garbage and it would be perfectly logical to assume that. Sieg can be played very unpredictably (even if unsafe), and his mix-ups are better than average. Assuming the opponent is NOT going to guess right or react right 100% of the time, eventually that means Sieg will get his hit, which for him usually means high damage.
Thankfully, your opponent (sophi) is not gonna attack constantly (even if she did, you can just predict and GI), so even if 100% of her moves are faster, there will be a moments when Sieg's slower moves can be sneaked in. In this match-up I turtle like CRAZY and typically only attack with 1 hit at a time (unless I get counter or juggle or ground N pound). I also abuse the hell out of 1k, don't know the frames to prove it, but it "feels" like one of his fastest moves.
Don't take this as me criticizing you, quite the contrary. I love your speeches. I just don't think it's really all that necessary to crunch all those numbers. Experience should tell anyone if the match-up is favorable or not. Seems pretty obvious to me. If your mix-up is guessed right and punished a lot, do it less often. Pretty simple.
The reason I don't think it's right to take out the human equation, is because that's exactly the situation were talking about here. Nobody is crunching numbers just to see the AI battle each other.
The human error factor is ALWAYS a factor. It's safe to assume your opponent will make mistake, even if rarely. Forcing your opponent into a big gamble of guessing games eventually makes them trip up, and get scared. Once the opponent is scared, that's when you got em' good. Putting fear into them, makes them make mistakes more.
One final point: frame data was discovered only through experience.
However, I still hold to the idea that crunching numbers, while it may give you a nerd-gasm, can't replace experience. Some punishers I only discovered by doing them in-battle, mostly by randomization when I try to get a feel for some characters. Not saying I alone discovered them (someone else probably figured that out long ago), but figured it out on my own.
Muscle memory/eye hand coordination will carry you much much further in SC than crunching numbers would. Most of the moves, there was no need to look at frame data because I pretty much already "knew" how fast/slow it was. There was no need to know "exactly" how bad a match-up between Sieg and Sophie was, as experience already told me how bad it was.
But back to the point, I don't think it's reasonable to take out the human equation. For example, by numbers alone, Sieg looks like garbage and it would be perfectly logical to assume that. Sieg can be played very unpredictably (even if unsafe), and his mix-ups are better than average. Assuming the opponent is NOT going to guess right or react right 100% of the time, eventually that means Sieg will get his hit, which for him usually means high damage.
Thankfully, your opponent (sophi) is not gonna attack constantly (even if she did, you can just predict and GI), so even if 100% of her moves are faster, there will be a moments when Sieg's slower moves can be sneaked in. In this match-up I turtle like CRAZY and typically only attack with 1 hit at a time (unless I get counter or juggle or ground N pound). I also abuse the hell out of 1k, don't know the frames to prove it, but it "feels" like one of his fastest moves.
Don't take this as me criticizing you, quite the contrary. I love your speeches. I just don't think it's really all that necessary to crunch all those numbers. Experience should tell anyone if the match-up is favorable or not. Seems pretty obvious to me. If your mix-up is guessed right and punished a lot, do it less often. Pretty simple.
The reason I don't think it's right to take out the human equation, is because that's exactly the situation were talking about here. Nobody is crunching numbers just to see the AI battle each other.
The human error factor is ALWAYS a factor. It's safe to assume your opponent will make mistake, even if rarely. Forcing your opponent into a big gamble of guessing games eventually makes them trip up, and get scared. Once the opponent is scared, that's when you got em' good. Putting fear into them, makes them make mistakes more.
One final point: frame data was discovered only through experience.