If we use the followup to CPFB (which can be done in ~10 STM)
SBsq (~7)
pair (...4?)
CLL + 4flip (10, going to go with shadowslice's stat here)
LSE (11, seems like a reasonable enough estimate as long as the 4flip you're forcing is either arrow or M2arrow)
= 42 STM
Could be pretty good. I'm a little bit worried about CLL+4flip recognition but maybe that will sort of be like OLL recog with time since you are just looking at U/D stickers. The movecount is nice and lookahead decent given that it's basically advanced Roux.
EDIT: Oh right, two other things: first, inspection is obviously hard but the CP is planned entirely in there, so there is no pause to do that stuff. Second, in order to get 10 STM CPFBs you will have to be fully CN. I think this is kind of neat, others may hate it, idk. Figured that's noteworthy.
EDIT: Also the pair making could take more than just 4 moves if you get a really ugly case. Hence why I wonder if it's better to go with: CPFB (10), SBsq (7), CLL + last corner (10), L7E. Don't know how much L7E can be optimized but my intuitive sense is that this might be more efficient, depending
SBsq (~7)
pair (...4?)
CLL + 4flip (10, going to go with shadowslice's stat here)
LSE (11, seems like a reasonable enough estimate as long as the 4flip you're forcing is either arrow or M2arrow)
= 42 STM
Could be pretty good. I'm a little bit worried about CLL+4flip recognition but maybe that will sort of be like OLL recog with time since you are just looking at U/D stickers. The movecount is nice and lookahead decent given that it's basically advanced Roux.
EDIT: Oh right, two other things: first, inspection is obviously hard but the CP is planned entirely in there, so there is no pause to do that stuff. Second, in order to get 10 STM CPFBs you will have to be fully CN. I think this is kind of neat, others may hate it, idk. Figured that's noteworthy.
EDIT: Also the pair making could take more than just 4 moves if you get a really ugly case. Hence why I wonder if it's better to go with: CPFB (10), SBsq (7), CLL + last corner (10), L7E. Don't know how much L7E can be optimized but my intuitive sense is that this might be more efficient, depending
Last edited: