DRich
DRich
Joined: Jul 6, 2012
  • Threads: 72
  • Posts: 6266
September 16th, 2020 at 8:10:20 AM permalink
Thanks guys. These are the types of discussions we need more of on this site.
Living longer does not always infer +EV
Runlikegod777
Runlikegod777
Joined: Oct 15, 2017
  • Threads: 18
  • Posts: 71
September 16th, 2020 at 11:04:24 AM permalink
I asked this same question to wizard or another top guy on here few years ago,they showed the math as 1.8% house edge. California 3 card poker right?
charliepatrick
charliepatrick
Joined: Jun 17, 2011
  • Threads: 32
  • Posts: 2111
September 16th, 2020 at 1:40:44 PM permalink
Quote: ssho88

I did this three years ago, Combination Analysis results for 3CP, delear show 1 card, WITHOUT ante bonus, raise if equal or better than X92( X is dealer's upcard, Q, K or ACE).

I proposed the X92 strategy as well but it may NOT the optimum strategy. Anyone can verify it ? gordonm888 ?

See image : https://ibb.co/52DjWwy

Yes I can confirm the strategy and house edge of 1.802206%.

The worst hands you can have versus a low card is one with no matching suits nor any straight penalty cards. The ranks of your hand (assuming you don't have A,K or Q) are irrelevant as it's only how many outs the dealer has that matters.
Perms: 1128 Win: 0 Lose: 726 Tie: 0 DNQ: 402 Ch1: 0 Ch4: 0 Ch5: 0 TEV: -0.9308510638297872 Cards: 2s 3h 5d/9c

As an example here are the close hands versus a Queen...
Perms: 1128 Win: 291 Lose: 829 Tie: 8 DNQ: 0 Ch1: 0 Ch4: 0 Ch5: 0 TEV: -0.9539007092198581 Cards: 2s 9h Qd/Qc
Perms: 1128 Win: 270 Lose: 849 Tie: 9 DNQ: 0 Ch1: 0 Ch4: 0 Ch5: 0 TEV: -1.0265957446808511 Cards: 7s 8s Qh/Qs
logicGuy
logicGuy
Joined: Oct 24, 2017
  • Threads: 1
  • Posts: 8
September 16th, 2020 at 2:12:51 PM permalink
Quote: Runlikegod777

I asked this same question to wizard or another top guy on here few years ago,they showed the math as 1.8% house edge. California 3 card poker right?


I was under the impression that CA 3-card poker has a commission and this game did not. Otherwise I believe it is essentially CA 3-card poker.
gordonm888
gordonm888
Joined: Feb 18, 2015
  • Threads: 38
  • Posts: 2561
September 16th, 2020 at 8:18:02 PM permalink
Quote: charliepatrick

Yes I can confirm the strategy and house edge of 1.802206%.

The worst hands you can have versus a low card is one with no matching suits nor any straight penalty cards. The ranks of your hand (assuming you don't have A,K or Q) are irrelevant as it's only how many outs the dealer has that matters.
Perms: 1128 Win: 0 Lose: 726 Tie: 0 DNQ: 402 Ch1: 0 Ch4: 0 Ch5: 0 TEV: -0.9308510638297872 Cards: 2s 3h 5d/9c

As an example here are the close hands versus a Queen...
Perms: 1128 Win: 291 Lose: 829 Tie: 8 DNQ: 0 Ch1: 0 Ch4: 0 Ch5: 0 TEV: -0.9539007092198581 Cards: 2s 9h Qd/Qc
Perms: 1128 Win: 270 Lose: 849 Tie: 9 DNQ: 0 Ch1: 0 Ch4: 0 Ch5: 0 TEV: -1.0265957446808511 Cards: 7s 8s Qh/Qs



Ssho88, this looks right to me.

The decision to not fold your hands that are Jack-hi and lower vs a dealer 2-J is not close. And the dealer's upcard will be 2-J with a frequency of 10/13 and you will have a J-high or lower about 30% of the time -so this is a big deal.

On a hand like 2s 9h Qd vs Qc you will lose whenever the dealer has at least one T, J,Q,K or A as one of his two other cards. Notice that simplifies the analysis, because you don't need to worry about the dealer making a straight or pairing the Q -those would involve cards that are in the range T-A, which I have just said is an automatic loser. You also lose when the dealer has 9 and a third card higher than a 2. There are basically 27 cards left in the deck from 2-8 and you win only when the dealer has 2 of those 27 cards without those two being a pair, and without those two cards being the same suit as the queen.

If you peek at a neighbor's hand you can sometimes make different decisions vs a Q,K or Ace depending upon how many of those 27 low cards you see -but, as I remember it that is worth less than 0.1% in house edge. My PC's hard disk is still in the the clean room at the data recovery shop so I can't quote anything specific.
So many better men, a few of them friends, were dead. And a thousand thousand slimy things lived on, and so did I.
charliepatrick
charliepatrick
Joined: Jun 17, 2011
  • Threads: 32
  • Posts: 2111
September 17th, 2020 at 2:05:09 AM permalink
Quote: gordonm888

...If you peek at a neighbor's hand....

The only info I have is for regular 3CP.

I ran a quick simulation where you're able to see all cards in the other five players and then make the decision whether to play or fold. As the numbers were small it's only a rough estimate. The basic House Edge (5 4 1) for 3CP is 3.373% (EoR 2.015%) and the House Edge went to 2.621% (EoR 1.523%).

The sim always played K-high and gradually folded various Q-highs. Note it often (about 18%) played zip hands.
The fold rates were about
10% at QT2
25% at Q87
50% at Q64
67% at Q32
82% at zip
gordonm888
gordonm888
Joined: Feb 18, 2015
  • Threads: 38
  • Posts: 2561
September 17th, 2020 at 4:40:30 PM permalink
Quote: charliepatrick

The only info I have is for regular 3CP.

I ran a quick simulation where you're able to see all cards in the other five players and then make the decision whether to play or fold. As the numbers were small it's only a rough estimate. The basic House Edge (5 4 1) for 3CP is 3.373% (EoR 2.015%) and the House Edge went to 2.621% (EoR 1.523%).

The sim always played K-high and gradually folded various Q-highs. Note it often (about 18%) played zip hands.
The fold rates were about
10% at QT2
25% at Q87
50% at Q64
67% at Q32
82% at zip



The reason that Q32 is folded so much less (67%) than "zip" (*2%) is because the player has a Q in his hand, depriving teh dealer of that card.

As I remember (I've done this same calc, lol) most of the advantage comes from times when you bet with zip -because you have zip so frequently.
So many better men, a few of them friends, were dead. And a thousand thousand slimy things lived on, and so did I.
ssho88
ssho88
Joined: Oct 16, 2011
  • Threads: 41
  • Posts: 451
September 17th, 2020 at 6:47:38 PM permalink
Quote: gordonm888

The reason that Q32 is folded so much less (67%) than "zip" (*2%) is because the player has a Q in his hand, depriving teh dealer of that card.

As I remember (I've done this same calc, lol) most of the advantage comes from times when you bet with zip -because you have zip so frequently.



I have a regular 3CP EV calculator for any specific 3CP hand, you can input your cards and up to other six players cards(rank and suit), you can even input dealer's hole card if you can see it(LOL!).There are three types of EVs, RAISE ALL THE TIME, RAISE ACCORDING TO BASIC STRATEGY(Q64) and RAISE ACCORDING TO HOLE CARD STRATEGY.

I can help to calculate the ev if you wish.
Last edited by: ssho88 on Sep 17, 2020
gordonm888
gordonm888
Joined: Feb 18, 2015
  • Threads: 38
  • Posts: 2561
September 17th, 2020 at 10:18:47 PM permalink
Since my 3CP calculator is on my hard disk that failed catastrophically (and is still at a clean room facility) I will need to rely on the grace of others. And of course, I trust any results you produce, ssho88.

3CP is actually one of the easiest casino games to model, don't we all agree?
So many better men, a few of them friends, were dead. And a thousand thousand slimy things lived on, and so did I.

  • Jump to: