Gå till innehåll

Engelskt ord för oslagbar strategi/system


Recommended Posts

Postad
sage

 

I believe their method of creating a list and then using that to create the optimal strategy is flawed.

1) There is no clear method to rank hands that will lead to an optimal order. Weird things happen in the equilibrium strategy like at 7bb deep 43s is played while 73s is not.

2) Another equilibrium strategy (http://www.holdemresources.net/hr/sngs/hune.html) which is as far as I can tell identical to mine, has discontinuities for when certain hands should be played. For example 63s should be played 7.1-5.1 bb deep and 2.3 bb deep.

3) The small blind edge listed on the SAGE website is .002 bb/hand (for 7bb deep), my calculation for that is .015bb/hand. That is the edge listed for the equilibrium strategy behind sage. I cannot generate the actual range of that equilibrium strategy, but because the small blind edge is different it makes me believe that the equilibrium strategy is flawed.

 

Below are the win rates of the equilibrium strategy, and a strategy which exploits SAGE. Note: these are all in big blinds per hundred hands.

 

Equilibrium versus Sage

2 bb .0969 bb/100

3 bb .4454 bb/100

4 bb .4602 bb/100

5 bb .4488 bb/100

6 bb .5221 bb/100

7 bb .8024 bb/100

8 bb .7874 bb/100

 

Exploitive versus Sage

2bb .0969 bb/100

3bb .4602 bb/100

4bb .4910 bb/100

5bb .4895 bb/100

6bb .6033 bb/100

7bb .8462 bb/100

8bb .8960 bb/100

 

This isn't a huge win rate, especially because with <10bb you won't be playing many hands before the tournament is over, but for a HU SNG grinder it could make a difference.

 

*** Math Content ***

All of these calculations were done with PokerStove and a calculator so that others can check them if they wish. Checking whether my code is correct or not is not easy. These numbers are slightly different that what my program calculates because the percent of hands which call is not calculated correctly with PokerStove. Lets say the open range is AA, and the calling range is AA, poker stove lists both ranges as 0.5% of hands, but when the sb opens AA the probability that the bb has AA is much lower because there is only one combination of AA left, my program takes this into account while PokerStove was not designed to work this way.

 

Equilibrium as sb versus Sage as bb (7 bb deep)

My open range: 22+,A2s+,K2s+,Q2s+,J2s+,T3s+,95s+,84s+,74s+,63s+,5 3s+,43s+,A2o+,K2o+,Q2o+,J7o+,T7o+,97o+,86o+,76o+

Sage's calling range: 33+,A2s+,K2s+,Q3s+,J5s+,T7s+,A2o+,K3o+,Q5o+,J7o+,T 9o+

sb equity when called .46490

sb open percent 66.5

bb call percent 48.6

ev for sb = (1-.665)*(-0.5)+(.665)*((1-.486)*(1)+(.486)*(.46490*7-(1-.46490)*(7))

= 0.015494434

 

Sage as sb versus Equilibrium Strategy as bb

Sage's open range: 22+,A2s+,K2s+,Q2s+,J2s+,T3s+,95s+,87s+,A2o+,K2o+,Q 2o+,J3o+,T5o+,97o+

My call range: 22+,A2s+,K2s+,Q2s+,J6s+,T7s+,97s+,A2o+,K2o+,Q6o+,J 8o+,T9o+

sb equity when called .46287

sb open percent .656

bb call percent .487

ev for sb = (1-.656)*(-0.5)+(.656)*((1-.487)*(1)+(.487)*(.46287*7-(1-.46287)*(7))

= -0.00153993504

 

Total EV = 0.00851718452 bb/hand ((0.015494434-(-0.00153993504))/2)

= .85 bb/100 (big blind not big bet)

*** End Math Content ***

Postad
I believe their method of creating a list and then using that to create the optimal strategy is flawed.

1) There is no clear method to rank hands that will lead to an optimal order. Weird things happen in the equilibrium strategy like at 7bb deep 43s is played while 73s is not.

2) Another equilibrium strategy (http://www.holdemresources.net/hr/sngs/hune.html) which is as far as I can tell identical to mine, has discontinuities for when certain hands should be played. For example 63s should be played 7.1-5.1 bb deep and 2.3 bb deep.

3) The small blind edge listed on the SAGE website is .002 bb/hand (for 7bb deep), my calculation for that is .015bb/hand. That is the edge listed for the equilibrium strategy behind sage. I cannot generate the actual range of that equilibrium strategy, but because the small blind edge is different it makes me believe that the equilibrium strategy is flawed.

 

Below are the win rates of the equilibrium strategy, and a strategy which exploits SAGE. Note: these are all in big blinds per hundred hands.

 

Equilibrium versus Sage

2 bb .0969 bb/100

3 bb .4454 bb/100

4 bb .4602 bb/100

5 bb .4488 bb/100

6 bb .5221 bb/100

7 bb .8024 bb/100

8 bb .7874 bb/100

 

Exploitive versus Sage

2bb .0969 bb/100

3bb .4602 bb/100

4bb .4910 bb/100

5bb .4895 bb/100

6bb .6033 bb/100

7bb .8462 bb/100

8bb .8960 bb/100

 

This isn't a huge win rate, especially because with <10bb you won't be playing many hands before the tournament is over, but for a HU SNG grinder it could make a difference.

 

*** Math Content ***

All of these calculations were done with PokerStove and a calculator so that others can check them if they wish. Checking whether my code is correct or not is not easy. These numbers are slightly different that what my program calculates because the percent of hands which call is not calculated correctly with PokerStove. Lets say the open range is AA, and the calling range is AA, poker stove lists both ranges as 0.5% of hands, but when the sb opens AA the probability that the bb has AA is much lower because there is only one combination of AA left, my program takes this into account while PokerStove was not designed to work this way.

 

Equilibrium as sb versus Sage as bb (7 bb deep)

My open range: 22+,A2s+,K2s+,Q2s+,J2s+,T3s+,95s+,84s+,74s+,63s+,5 3s+,43s+,A2o+,K2o+,Q2o+,J7o+,T7o+,97o+,86o+,76o+

Sage's calling range: 33+,A2s+,K2s+,Q3s+,J5s+,T7s+,A2o+,K3o+,Q5o+,J7o+,T 9o+

sb equity when called .46490

sb open percent 66.5

bb call percent 48.6

ev for sb = (1-.665)*(-0.5)+(.665)*((1-.486)*(1)+(.486)*(.46490*7-(1-.46490)*(7))

= 0.015494434

 

Sage as sb versus Equilibrium Strategy as bb

Sage's open range: 22+,A2s+,K2s+,Q2s+,J2s+,T3s+,95s+,87s+,A2o+,K2o+,Q 2o+,J3o+,T5o+,97o+

My call range: 22+,A2s+,K2s+,Q2s+,J6s+,T7s+,97s+,A2o+,K2o+,Q6o+,J 8o+,T9o+

sb equity when called .46287

sb open percent .656

bb call percent .487

ev for sb = (1-.656)*(-0.5)+(.656)*((1-.487)*(1)+(.487)*(.46287*7-(1-.46287)*(7))

= -0.00153993504

 

Total EV = 0.00851718452 bb/hand ((0.015494434-(-0.00153993504))/2)

= .85 bb/100 (big blind not big bet)

*** End Math Content ***

 

 

tack 8-)

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Gäst
Svara i detta ämne...

×   Du har klistrat in innehåll med formatering.   Ta bort formatering

  Endast 75 max uttryckssymboler är tillåtna.

×   Din länk har automatiskt bäddats in.   Visa som länk istället

×   Ditt tidigare innehåll har återställts.   Rensa redigerare

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Skapa nytt...