ssho88
Joined: Oct 16, 2011
• Posts: 655
September 25th, 2019 at 4:53:49 AM permalink
Here are the BJ game rules :-

1) SINGLE deck, Peek, H17, No Surrender.
2) DAS, D10(hard 10 and 11).
3) Split to maximum 2 hands for ALL pair.
4) 3 hands per round and shuffle after each round.
5) No hit to split Ace and Insurance pay 2 to 1.

House edge = -0.373%( favorable to casino).

Bet 3 hands in a round, and use Hi-Opt II counting system(with index play) + Ace side count to play the game.

I beleive that each hand will have different EV because each hand can see different number of cards before apply the correct strategy and hence with different playing efficiency. Therefore, EV(Hand3) > EV(Hand2)> EV(Hand1).

Any expert here can help me to simulate this game and calculate the EV of each hand ?

James
MichaelBluejay
Joined: Sep 17, 2010
• Posts: 1589
Thanks for this post from:
September 25th, 2019 at 1:11:01 PM permalink
It's a tall task and I won't have time to get to it any time soon, but I can tell you that your theory is generally correct, just not for 3 spots. In single-deck games with a low house edge, an old play was to play all 7 spots, betting progressively from small to large on each spot, so as the EV went up for each successive hand, so did the bet size. Certain it wouldn't work for 3 spots.
Last edited by: MichaelBluejay on Sep 26, 2019
Presidential Election polls and odds: https://2605.me/p
ssho88
Joined: Oct 16, 2011
• Posts: 655
September 25th, 2019 at 2:27:45 PM permalink
Quote: MichaelBluejay

It's a tall task and I won't have time to get to it any time soon, but I can tell you that your theory is correct. In fact, in single-deck games with a low house edge, an old play was to play every spot, betting progressively from small to large on each spot, so as the EV went up for each successive hand, so did the bet size.

Thanks for your reply. That's exactly what I am trying to do !

I think this can be simulated easily with softwarw CVDATA, anyone with CVDATA here can help to do the simulation ?

James
KevinAA
Joined: Jul 6, 2017
• Posts: 282
September 26th, 2019 at 7:48:19 PM permalink
Shuffle after each round? Is this video BJ?
ssho88
Joined: Oct 16, 2011
• Posts: 655
September 26th, 2019 at 8:45:33 PM permalink
Quote: KevinAA

Shuffle after each round? Is this video BJ?

See rule no 4 in my first post. It is NOT video game.

Beside that, can you simulate it for me ?
KevinAA
Joined: Jul 6, 2017
• Posts: 282
September 26th, 2019 at 9:24:55 PM permalink
Quote: ssho88

See rule no 4 in my first post. It is NOT video game.

Beside that, can you simulate it for me ?

3 hands per round means player #1 gets two cards, player #2 gets two cards, player #3 gets two cards, and the dealer gets two cards. Then each person stands or hits etc and the dealer does the same, and that's the end of the round. Then you re-shuffle.

That is exactly what video BJ does, though each person has their own shoe, which means it doesn't matter what cards they have, you only play basic strategy, as if you're there by yourself. Then the computer reshuffles.

I have never seen a live table game re-shuffle after each round. Usually they hand-shuffle, but even the places that use a ShuffleMaster will still deal through the deck the same way, about 75%.
ssho88
Joined: Oct 16, 2011
• Posts: 655
September 26th, 2019 at 10:39:54 PM permalink
Quote: KevinAA

3 hands per round means player #1 gets two cards, player #2 gets two cards, player #3 gets two cards, and the dealer gets two cards. Then each person stands or hits etc and the dealer does the same, and that's the end of the round. Then you re-shuffle.

That is exactly what video BJ does, though each person has their own shoe, which means it doesn't matter what cards they have, you only play basic strategy, as if you're there by yourself. Then the computer reshuffles.

I have never seen a live table game re-shuffle after each round. Usually they hand-shuffle, but even the places that use a ShuffleMaster will still deal through the deck the same way, about 75%.

Read my first post, it is a SINGLE deck game, reshuffle after each round. There are 5 box on the table, but a player only allow open max 3 box.

player#3 can see more cards before any action . . .
MichaelBluejay
Joined: Sep 17, 2010
• Posts: 1589
September 26th, 2019 at 11:08:09 PM permalink
I read too quickly and didn't see that you'd be betting only 3 spots. Doubtful you can get an edge that way, unless maybe the game is already something like -0.18% to begin with. I edited my first reply to reflect that only 3 spots is dicey.
Presidential Election polls and odds: https://2605.me/p
unJon
Joined: Jul 1, 2018
• Posts: 4345
September 27th, 2019 at 4:06:28 AM permalink
Quote: ssho88

Read my first post, it is a SINGLE deck game, reshuffle after each round. There are 5 box on the table, but a player only allow open max 3 box.

player#3 can see more cards before any action . . .

Your sort of asking if the three seat in the first round of a single deck game (so no bet variation) is +EV based on composition strategy. Question almost answers itself.
The race is not always to the swift, nor the battle to the strong; but that is the way to bet.
ssho88
Joined: Oct 16, 2011
• Posts: 655
September 27th, 2019 at 4:14:59 AM permalink
Quote: unJon

Your sort of asking if the three seat in the first round of a single deck game (so no bet variation) is +EV based on composition strategy. Question almost answers itself.

I beleive that EV of each hand will be better than -0.373%, how we know that especially the HAND3 EV is +ve ? Can you show me the calculations OR confirm it with simulation ?
gordonm888
Joined: Feb 18, 2015
• Posts: 4810
September 27th, 2019 at 10:48:18 AM permalink
Either in a book or on a website, I believe that many years ago I have seen a graph/plot of the average +EV vs number of cards exposed that can be gained by perfect composition -dependent play in a single deck BJ game. It may have included the Insurance decision as well as index plays and other close plays (and I don't know whether the insurance decision is relevant to ssho88's game.) Anyway,

(1)does anyone else remember such a plot?

(2) and possibly provide a link or reference?
So many better men, a few of them friends, are dead. And a thousand thousand slimy things live on, and so do I.
ssho88
Joined: Oct 16, 2011
• Posts: 655
September 27th, 2019 at 7:45:02 PM permalink
Quote: gordonm888

Either in a book or on a website, I believe that many years ago I have seen a graph/plot of the average +EV vs number of cards exposed that can be gained by perfect composition -dependent play in a single deck BJ game. It may have included the Insurance decision as well as index plays and other close plays (and I don't know whether the insurance decision is relevant to ssho88's game.) Anyway,

(1)does anyone else remember such a plot?

(2) and possibly provide a link or reference?

Gordon,

Instead of using perfect Composition strategy and make the task easier, we can use the BEST single deck counting system to do the simulation. Anyone here can use CVDATA to do it ? Is the CVDATA result comparable to perfect strategy result ? would appreciate hearing your thoughts.

James
ssho88
Joined: Oct 16, 2011
• Posts: 655
September 27th, 2019 at 7:57:42 PM permalink
Gordon,

Other AP already simulated it with Hi Opt 2 counting system, the EV OF HAND1, HAND2 and HAND3 is -0.351, -0.320 and -0.286 respectively.

How much further can we reduce the ev if we apply perfect composition strategy ? Would appreciate hearing your thoughts.

James
gordonm888
Joined: Feb 18, 2015
• Posts: 4810
September 28th, 2019 at 12:34:44 AM permalink
Quote: ssho88

Gordon,

Other AP already simulated it with Hi Opt 2 counting system, the EV OF HAND1, HAND2 and HAND3 is -0.351, -0.320 and -0.286 respectively.

How much further can we reduce the ev if we apply perfect composition strategy ? Would appreciate hearing your thoughts.

James

Do you which decisions contributed to the Hi Opt 2 improvements and by how much? I am surprised by how much the EV changed.

1. Insurance is the most important decision and Hi Opt 2 should do reasonably well for that, although a dedicated insurance count would pick up some opportunities that Hi-Opt2 will miss. For your 3-hand application, Insurance may not be important - and you could probably do the Insurance math in your head better than Hi Opt 2.

2. The 2nd most important decision is 16vT which occurs on about 3.37% of all hands given the rules you have posted. The major issue with Hi Opt 2 counting system is that for any 16 v T decision, a 5 is the card that most makes you want to Stand and a 6 is the card that most makes you want to Hit. Hi Opt 2, like every other counting system, treats the 5 and 6 as if they are both low cards and not as cards that have opposite effects on the decision. This is probably the single largest inefficiency of the Hi Opt II count for your application.

I don't know how much better you can do with 16vT by using perfect composition strategy. My wild-assed guess, based on some work I have been doing, would be an improvement of <0.001 (edit: <0.1%) in HAND2 EV. HAND3 EV would obviously be a bigger improvement.

3. Hi Opt 2 should be excellent on 15vT and fairly good on 16v9.

4. The Hit/Double decisions on 10vT and 10vA can be flipped by only a few low cards coming out. The decision to double 10vT depends critically on no Aces being out (because dealer's downcard is known not to be an Ace) - not sure whether the index criteria for Hi Opt2 factor in the Aces correctly.

5. 12v3 depends most importantly on whether 9's are out (and also 7's and 8's) -which is a blind spot for every counting system designed for betting purposes. Frequency of 12v4, including multicard 12's, is about 0.007, so I suspect you can't get much more than another 0.0001 in EV. I don't really know, though.

6. There are many other decisions that can flip due to card composition ( and not just the 18 index plays that the books talk about) but most of the effects would be small because the decisions occur less frequently or the decisions are not quite as close.

Sorry I don't have a better answer.
Last edited by: gordonm888 on Sep 28, 2019
So many better men, a few of them friends, are dead. And a thousand thousand slimy things live on, and so do I.
MichaelBluejay
Joined: Sep 17, 2010
• Posts: 1589
September 28th, 2019 at 1:50:30 AM permalink
ssho88, what strategy deviations/indices are you using? Also, what are the details of your ace side count?
Presidential Election polls and odds: https://2605.me/p
ssho88
Joined: Oct 16, 2011
• Posts: 655
September 28th, 2019 at 6:39:54 AM permalink
Quote: MichaelBluejay

ssho88, what strategy deviations/indices are you using? Also, what are the details of your ace side count?

Full Hi Opt 2 + Ace side count
MichaelBluejay
Joined: Sep 17, 2010
• Posts: 1589
September 28th, 2019 at 12:58:15 PM permalink
Yes, what are they?
Presidential Election polls and odds: https://2605.me/p
tomchina123
Joined: Aug 26, 2015
• Posts: 305
September 13th, 2023 at 7:04:31 AM permalink
it is a so old thread. but i want to say sth to show i am still reading this forum.

it is some years, since i came into this site.

SSHO88 is much better now, i guess, no matter income or ap skills.

for me, i made some good money on being an Ap. thanks to Wizard and Eliot.
for my skills, i have all needed to beat this game. i can write the CA , software to beat this game.
but is there any good site which accepts Chinese players and offer 5 hands or 3 hands in one round single deck bj online, and pay?
ChumpChange
Joined: Jun 15, 2018
• Posts: 4484
September 13th, 2023 at 7:47:54 AM permalink
I don't even like these rules, how is the HA so low?
Here are the BJ game rules :-
1) SINGLE deck, Peek, H17, No Surrender.
2) DAS, D10(hard 10 and 11).
3) Split to maximum 2 hands for ALL pair.
4) 3 hands per round and shuffle after each round.
5) No hit to split Ace and Insurance pay 2 to 1.
House edge = -0.373%( favorable to casino)

My preferred rules:
1) SINGLE deck, Peek, S17, Surrender.
2) DAS.
3) Split to maximum 4 hands for ALL pair.
4) 75% penetration
5) Hit to split Ace and Insurance pay 2 to 1.
What's the HA, and do these games even exist? Maybe I have to go to an 8 deck shoe on the \$100 minimum tables to get this.

I think the last \$5 table I saw pre-pandemic was
Rules:
1) 6 deck, Peek, H17, Surrender.
2) DAS.
3) Split to maximum 4 hands for ALL pair.
4) 75% penetration
5) No hit to split Ace and Insurance pay 2 to 1.