B2 D B' L', F' R2 U2 D' * F // 2c3 (9)
(R2 D' B2 U L2 D R2 L2) // 2e2e (17)
* U2 F2 D2 B2 D2 F2 (+6-3=20)

beats mallard haha dumb computer

other:
(F' U D' R', F R2 B' U2 F) // 2c4 (9) but i did not want to sift through all the variations for a 21
B' D2 B2 L' (F' R2 F2 U B') // 2c4 (9)

(L' D B' U2 R) F' B' R2 U L2 U // 2c3 (11)
L2 B' D2 B D2 B' // htr (17)
L2 U2 F2 L2 // slice (21)
+1 -> 22

R (F U2 L) B2 D2 R2 U D' B' U B' (F2 U) // htr (14)

(F) F B L2 F', R' L' D' R2 U' R // 4a1 (11)
D' (U' R2 L2 D)
(B2 U2 R2 F2 L2) // slice (21)
+1 -> 22

rzp in 7 but not really, i noticed R L from the 4-mover gave nice U/D but bad E

other:
U' B' R2 L' U', F (F L2 B' R2 B) // 4a2 (11)
(F' U' D L' F) F2 D' B2 U L // 4c4 (10)
F' U R2 L B', R' B2 U L' U L // 4c3 (11)
(F) F2 U2 B', R' B2 U L2 D' L // 2c4 (10)
U B (L' F', L' B2 D L2 D L) // 2c4 (10)

(D' L D2 R, U') L2 U2 L2 F2 U' // 2c3 (10)
R2 F D2 B' D2 B L2 D2 F2 // slice (19)
+1 -> 20

other:
R2 B' L D2 R, F2 U2 F2 U' F // 2c3 (10)
R2 B' R F2 L, F2 U2 F2 U' F // 2c3 (10)
(D' L D2 R, U) D' L2 D2 F2 D // 2c4 (10)
(D) B U, F' R' F2 L2 B D2 L // 4c3 (10)
(D) D2 B U, F' L' U2 L U2 L' B // 4b2 (11)

(B2 D L', U B, U2 F2 D' L2 F) // 4c3 (10)
(U L2 B2 U B2 D2 L2 D)
(L2 B2 U2 B2) // slice (22)
+0 -> 22

I knew the 2c4s in 8 would be pain and I tried them anyway :(

D2 B U2 R' F' (R' B2 L D') // 2c4 (9)
(F2 R B2 R' U2 F2 R F2 L2 F2 R U2 D2) // slice (22)
+0 -> 22

also had D2 L' (F2 R' B2 R B2 R' B2 U2 F2 D2) // slice (21) but no +0

cool dr:
U (R' D2 R2 D', R' L' B D2 R F) // 4c4 (11)

(D' R F R L', B) D F2 U B // 4c4 (10)
(D' R F R L', B) F2 U F2 D B // extend, 4b2 (11)
(L2 B2 D R2 F2 U') // htr (17)
(R2 B2 U2 L2 F2) // slice (22)
+2 urgh

L D2 L2 F', U' L' D2 L2 D' R D' // 2c4 (11)
L' D2 L B2 D2 L D2 L // htr (19)
B2 D2 L2 F2 // slice (23)
+0 -> 23

sad

F D2 R F (R2 U L U' R) // 2c4 (9)
F D2 R F (R2 U' R U' L) // 4b2 (9), 21 from (L2 B2 U) switch which i checked but didn't find
D B2 L' F, D (B2 U B2 U' B2 U) // 4b2 (11)
D L' F' (U L) R' F2 U2 R D2 R' // 4c3 (11)
D L' F (R U', D2 B2 L' R D) // miscounted parity and didn't check, it's 4c3 with easy 19

L R2 U' R, B' U B' U B2 D' F
(D R' U F) L (D2 R D B2 U L)
(D R' U F) D L (D2 R D B2 U)
(F' D L' U', L B R' F' L B R)

L (D' B2 R) // eo (4)
D' B2 U R2 U F // 4a1 (10)
(U') R2 U2 L2 U^ R2 L2 D^ // htr (18)
F2 D2 B2 R2 // slice (22)
^ w

all my drs had absolute shit htr -> finish cases lmao

(F) F' R' D B' // eo (5)
U F2 B2 L' B2 L2 [1] D // 4b2 (12)
(U2 L D2 L F2 U2 B2) // 2e2e (
[1] L2 U2 L2 D2 R2 D2 (

(U' R' B' R' D') // eo (5)
B' D2 R' B' L' // 2c4 (10)
L F' R B2 R' F L' // htr (17-2=15)
U2 F' B D2 [1] L2 // 2e2e (20)
[1] D2 B2 D2 B2 D2 B2 (+6-3=23)

(U' R F U') // eo (4)
(L2 F2 B L' B') // 4b2 (9)
R' F2 L F2 R B2 R' // htr (16)
B2 U2 D2 B F U2 F' B' // solved (24-2=22)

(D R2 B)
R' U' L R2 D' L2 F2 D
(R' U2 R D2 L U2 L')
(F2 L2 U2 F2 U2)
+1

(R' D' F) // eo (3)
L (U L2 U B2 L) // 4b2 (9)
U' F2 D // htr (12)
i suck at finding finishes so

(R' D' F) L (U L2 U) [U' F2 D] // flipped 4a1 (10)
[] = D U2 R2 U // +1=11
{B2 L} // htr (13)
F2 B2 L2 F2 // M slice (17)
{} = R' L U2 R // +2=19

i had this interesting dr
(F') R2 F U F', R2 F2 D2 L U' R D // 4c3 (12)
(D2 R) // fake htr (14)
optimal for the eo goes through this dr which looks like some weird kind of flip
(F') R2 F U F' (D2 R, F2 D R' U L')

R (R' L D' L) // eo, dr-5c4e (5)
F R2 D' U2 B' // 4c3 (10)
(F2 D') U^ B2 U2 F2 U^ // htr (17)
R2 F2 D^2 F2 // slice (21)
^ w

yeah this was cool as fuck ngl

other:
L2 D F2 L, U (D2 B2 L2 B U) // 4c4 (10), bad
U' L' F' L' D, F2 R2 F L' // 2c4 (9), 21 opt
(F2 R' D B) R U', R U2 R' U2 R // 2c3 (11), 21 opt off some variation
(B D L2 B', R) D F2 D' F2 D R // 2c3 (11), 23 opt

(D') F2 B2 U // eo (4)
R' B' (L2 F U2 D2 B') // 4b2 (11)
(F2 R^ F2 R^) // htr (15)
(B2 D2 B2 D2 R2^) // slice (20)
^ w

other:

(D' F B) U (U F2 U2 R U) // 2c3 (9), 21 opt
(D L2 F B') U (L' F2 L2 U) // 4b2 (9), 21 opt
B (F D2 B, L R U' L2 U R) // 2c4 (10), 23 opt
somehow missed the dr on the other axis lol
B' (F D2 B, L' B2 U L2 D) // 2c4 (9), 21 opt

U R (U' B' R) // eo (5)
L2 F' U2 R2 D // 2c3 (10)
B^ R2 U2 B D2 B^ // htr (16)
D2 L2 B^2 U2 // slice (20)
^ w

other:
L2 B' R D L, B L2 F' D2 F2 B D' // 4b2
R2 U2 F' U2 B' U2 D2 R2 D2 R2 // slice in 22 but it's doomed

(R L2 D F, D) L2 U2 R' F2 R2 D // 2c4
L' F2 R' U2 R U2 R // slice+2e in 18, 2e is preserved in htr before R U2 R so we can do:
L' F2 R (R' U2 R) // htr
(B2 U2 R2 U2 B2) // slice in 22, doesn't +0

L (B R) // eo (3)
(L2 F D) // jzp (6)
R2 [D2 F2 D] // 2c3 (10)
(B' D2 B' D2 R2 D2 R2 B) // htr (18)
[] = F2 U2 B2 D' R2 // extend, S+E (20)
+1/+2

should have found 22 off the 15 to htrs but i did not want to do 7 HT finish
dr is 20 optimal but honestly unfindable
(R2 F U2 D2 R2 B' R2 D2 F)
can't really switch after (R2 F) because you have to blocky fish the other side to find the correct variation
21 seems somewhat findable but you have to do (U2 F D2 B) and switch

(L) F R B' L, D2 B' R2 B' D // 2c3 (10), annoying not to find the 22 from this

(B) F D L B, R D' R U2 L // 2c4 using cool 5c6e case (10)

i found optimal after the first 7 moves (23) but none of the optimals use that dr
(B) F D L B, R D' R' U L2 D2 L // 2c3 with extension trick (12)
(U2 F2 D^' R2 D^' L2 D' F2) // M+E (20)
^ w // M in 20, but it's +3 -> 23