#1
|
|||
|
|||
Is this right?
Fiddling with twodimes
http://twodimes.net/h/?z=2989462 pokenum -l27 7h 5s 3s 2d jc - 5c 8d 4s 3c - 5h 4c 2h 6d 5-card Draw 2-7 Lowball: 1482 enumerated outcomes cards win %win lose %lose tie %tie EV 5s 3s Jc 2d 7h 489 33.00 993 67.00 0 0.00 0.330 4s 5c 3c 8d 481 32.46 1001 67.54 0 0.00 0.325 4c 6d 5h 2h 512 34.55 970 65.45 0 0.00 0.345 http://twodimes.net/h/?z=2989456 pokenum -l27 7h 5s 3s 2d tc - 5c 8d 4s 3c - 5h 4c 2h 6d 5-card Draw 2-7 Lowball: 1482 enumerated outcomes cards win %win lose %lose tie %tie EV 5s 3s Tc 2d 7h 681 45.95 801 54.05 0 0.00 0.460 4s 5c 3c 8d 381 25.71 1101 74.29 0 0.00 0.257 4c 6d 5h 2h 420 28.34 1062 71.66 0 0.00 0.283 http://twodimes.net/h/?z=2989482 pokenum -l27 7h 5s 3s 2d - 5c 8d 4s 3c - 5h 4c 2h 6d 5-card Draw 2-7 Lowball: 59280 enumerated outcomes cards win %win lose %lose tie %tie EV 5s 3s 2d 7h 21872 36.90 37408 63.10 0 0.00 0.369 4s 5c 3c 8d 18354 30.96 40926 69.04 0 0.00 0.310 4c 6d 5h 2h 19054 32.14 40226 67.86 0 0.00 0.321 Keeping a J against 2 one card draws results in a 3 way coinflip? I was under the impression that keeping the J was a major error but seems to be only a small one? I originally ran this simulation with the 10 to see what equity the guy had when he chose to pat behind with his 10. I thought he probably made an error not drawing to the wheel and figured patting a rough 10 would be correct. Obviously I was wrong. 46% equity? Is it really that high? |
#2
|
|||
|
|||
Re: Is this right?
Even three-way, it's pretty unlikely 10% of the pot is worth as much as the river action he gets when he draws and hits.
|
#3
|
|||
|
|||
Re: Is this right?
Yup, it's correct.
T7532 vs. 8543 vs 6542: We can estimate that other players are drawing to 12 and 13 outs so it should be something like 27/39*25/38 = 45.5% chance of them both bricking. (Not quite accurate.) To get the right answer, we can count the two-card cases where the T loses by hand. There are 39 cards left unspecified so there are 1482 possibilities (like pokenum says) 9x: 4*38 7x: 3*38 6x: 3*38 2x: 2*38 A[T987]: 4*13 K[T987]: 4*13 Q[T987]: 4*13 J[T987]: 4*13 T[T987]: 3*12 8[T987]: 3*12 5[T987]: 1*13 4[T987]: 2*13 3[T987]: 2*13 That gives 801 ways to lose, which is just what pokenum says, 45.95% chance of winning. Last-draw equities against two players drawing one tend to flatten out because there is a substantial chance for each player of bricking quite badly. For example, 532 vs 8543 vs 6542: 21% to win 742 vs 8543 vs 6542: 20% to win T532 vs 8543 vs 6542: 25% to win T9532 vs 8543 vs 6542: 41% to win Unless an opponent is pat, a live draw will have substantial equity. Drawing four to a deuce is still 10% to win against these two draws. |
|
|