DrOakland: Card dependent strategy

[DrOakland]

Off the top in a single deck game, the composition of the cards you hold can change certain basic strategy decisions. For example, a 12 consisting of 10+2 should actually be hit against a 4, because one of the 16 bust cards is in your hand. Another: any hand of 16 containing a 5 should stand against a 10. There is website giving numerous "fine point" strategy changes for various numbers of cards held...

My question is, has anyone figured out the cumulative value of all these strategy changes? The "Wizard of Odds" gives the figure as 0.0018%, which seems awfully low to me.

DrO

[Don Schlesinger]

> My question is, has anyone figured out the
> cumulative value of all these strategy
> changes? The "Wizard of Odds"
> gives the figure as 0.0018%, which seems
> awfully low to me.
On p. 177, Griffin tells us that using all such c-d plays, the player's edge in SD is 0.04%. As the edge is usually thought to be no more than 0.01% for the player, without the plays, I'd say, at most, all the changes are worth 0.03, but if Michael says 0.02 (rounded), I wouldn't argue with him.

You have to understand that the quantity of plays doesn't matter very much if each is worth, literally, only a couple of thousandths difference.

Don

Don

[DrOakland]

> On p. 177, Griffin tells us that using all
> such c-d plays, the player's edge in SD is
> 0.04%. As the edge is usually thought to be
> no more than 0.01% for the player, without
> the plays, I'd say, at most, all the changes
> are worth 0.03, but if Michael says 0.02
> (rounded), I wouldn't argue with him.

> You have to understand that the quantity of
> plays doesn't matter very much if each is
> worth, literally, only a couple of
> thousandths difference.

> Don

Well Mike said 0.0018%, which rounds to 0.002, not 0.02. (Wow! I caught Don in a math error of an ORDER of MAGNITUDE do I get a prize?)

Anyway 0.03 sounds more interesting (and profitable) than Mike's figure. The applicaton here is the single deck online casinos with great rules, which already have a tiny player advantage.

DrO

[Don Schlesinger]

> Well Mike said 0.0018%, which rounds to
> 0.002, not 0.02. (Wow! I caught Don in a
> math error of an ORDER of MAGNITUDE do I
> get a prize?)

> Anyway 0.03 sounds more interesting (and
> profitable) than Mike's figure. The
> applicaton here is the single deck online
> casinos with great rules, which already have
> a tiny player advantage.

At the bottom of p. 17 of Griffin, Peter gives 0.04% as an upper limit. I think Mike has a typo in what he has written.

I sent him an e-mail and will let you know what he says. My eye went right over the extra decimal place, because I didn't think Mike could be off by that much!

Don

[Don Schlesinger]

> At the bottom of p. 17 of Griffin, Peter
> gives 0.04% as an upper limit. I think Mike
> has a typo in what he has written.

> I sent him an e-mail and will let you know
> what he says. My eye went right over the
> extra decimal place, because I didn't think
> Mike could be off by that much!

Mike has checked his work and has confirmed his error to me, by e-mail. If I get his permission, I will post his response later today.

Don

[Don Schlesinger]

Hi Don:

You're right. Thanks for the correction. I think I made a mistake with the decimal point and somewhere else as well. Going just 2 and 3 player cards deep I get a savings of 0.02%, 0.013% for 2 cards and 0.007% for 3 cards. Assuming it continues to diminish as the number of cards increases I feel 0.03% is going to be correct when rounding to 0.01%.

Regards,

Mike

Don speaking now: The above confirms what I wrote earlier: I believe perfect total-dependent BS in SD, s17 gives the player an even game or even a 0.01% edge. Using composition-dependent plays in SD raises that player edge to 0.04%, as now confirmed by both Shackleford and Griffin.

[DrOakland]

> Mike has checked his work and has confirmed
> his error to me, by e-mail. If I get his
> permission, I will post his response later
> today.

> Don

He confirmed it with me too. I got to correct both the Wizard and the Don on the same day! woo-woo

My interest was in adding to the tiny edge off the top on Unified Gaming online casino software. They have a single deck with s17, DAS, and LS. Mike gives the player edge as 0.16%. With the card dependent plays, brings it maybe to a full 0.2. Let's see, I play 200 bets of $100 each hour, 0.2% of that action makes... $40 an hour! I better start selling seminars

DrO

[Don Schlesinger]

> Let's see, I play
> 200 bets of $100 each hour, 0.2% of that
> action makes... $40 an hour! I better start
> selling seminars

Sounds much too easy. I'd be deathly afraid of being cheated.

Don

[ET Fan]

First, Unified Gaming has refused to pay several people because they believed the individuals were using "robots." In some highly publicised cases, the individuals eventually were payed. I don't think there's much doubt though, that some UNpublicised cases did NOT get paid off. Ie., the players who got paid were the ones who knew how to work the publicity game, posting complaints to "Gambling Grumbles," the OPA, etc.

If you play 200 hands per hour perfectly, they will suspect you of using a robot.

Secondly, realize that you need a bankroll ~$67,000 to justify making $100 bets with a 0.2% advantage. That's at full Kelly, which is a very rough ride. If you intend to continue making $100 bets through the downswings, yout BR should be (as a rough guess) three or four times that amount. So call it $200,000. This is money actually AT RISK.

Before I risk 200 grand on an internet casino, I'm going to want to know a lot about their organization. Cheating can and does occur at internet casinos. I haven't heard any reputable reports on UG cheating, but of course you hear grumbles about all the software comps.

I would want it in writing, signed in blood, that the computer generated cards are weighted similarly to real cards. All casinos go on about their great random number generators, but just TRY to pin them down on the weight issue.

Have you factored in the fact that the computer never makes mistakes, but you will? Also, suppose you get a good hand and the connection is lost (along with the hand). Hope you realize the operator on their end can monitor the games and can sever a connection at will. Of course, "these things happen" at random as well ... ;-?

I made a couple thousand counting UG's 4d shoes before they changed the rules and procedures. It's hard for me to believe, after all this time and experience, that UG is giving anything away.

IMO, the way to make a few bucks on the internet is to go after the bonuses. Boring, inelegant, but EV/var is hard to argue with.

If, after all this, you decide to go for the $40 per hour, I wish you luck.

ETF

> He confirmed it with me too. I got to
> correct both the Wizard and the Don on the
> same day! woo-woo

> My interest was in adding to the tiny edge
> off the top on Unified Gaming online casino
> software. They have a single deck with s17,
> DAS, and LS. Mike gives the player edge as
> 0.16%. With the card dependent plays, brings
> it maybe to a full 0.2. Let's see, I play
> 200 bets of $100 each hour, 0.2% of that
> action makes... $40 an hour! I better start
> selling seminars

> DrO

[DrOakland]

> IMO, the way to make a few bucks on the
> internet is to go after the bonuses. Boring,
> inelegant, but EV/var is hard to argue with.

> If, after all this, you decide to go for
> the $40 per hour, I wish you luck.

> ETF

Well my tongue was semi-firmly in my cheek concerning $40/hr. I do realize the bankroll issues, and appreciate your info. I'm a latecomer to Internet gambling. Yes, inelegant bonuses are definitely the thing. But when I just want to play (I know, horrific reflection on my character), it's nice to think I have an edge, however tiny.

DrO