Question about soft doubles.

[bjanalyst]

All playing strategy decisions depend on the Effects of Removal and Full Deck House Advantage for the particaulr playing startegy and the given number of decks as found in Don Schlesinger's Black Jack Attack 3rd edition and the tag values of the count or derived count which is a linear combination of the primary count and side count which is chosen to maximize the ABS(CC) between the EoR and the tag values of the derived count.

Double A2 v 3 if HL + 1.5*(Am8c) >- 7*dr with CC = 90% where dr = decks remaining and CC = Correlation Coefficient. A2 v 3.pdf
Double A5 v 3 if HL + 6*(AA78mTc) >= 12*dr with CC = 79% or just use double if AA78mTc >= 2*dr with CC = 78% A5 v 3.pdf

[Dbs6582]

The easiest answer is because that is what the sim says to do. You base your indices on sim data, not logic.

Hey Three, I understand where Lsubz274 is coming from. It helps to remember things if you understand the logic behind it. I thought his question was good and have asked the same thing in the past. I came to the same conclusion MidWest player came too. It’s because you can hit the A3 another time if you get a low card. I don’t know if that’s 100% correct but I’m sure that’s part of it.

It looks like Lsubz274 got his handle the same way I did, by randomly typing some letters and numbers on a key board. Lol. If I’m wrong, sorry about that Lsubz274. Looking back I wish I’d given my handle a little more thought and shortened it. But I’ve come to like Dbs. However, I wish I’d left off some of those numbers.

[bjanalyst]

lSims and/or using LSL technique with EoR are fine for determing indices and when to deviate but the results must make logical sent also.

So here:
Double A2 v 3 if HL + 1.5*(Am8c) >- 7*dr with CC = 90% where dr = decks remaining and CC = Correlation Coefficient.

Logic: If Am8c is a large positive number then there is a deficiency of Aces remaining in the shoe which is good for player doubling A2 v 3 since less chance player doubles with an Ace getting a doubled stiff and less change dealer has an Ace in the hole for an A3 which hand will probably not bust. Also the excess 8's means if the player hits his A2 with an 8 he doubles to a perfect 21. Unfortunately if dealer has an 8 in the hole he has a total of 11 which is not good but it is still better to have extra 8's which would give player who doubles his A2 and perfect doubled 21.

Double A5 v 3 if HL + 6*(AA78mTc) >= 12*dr with CC = 79% or just use double if AA78mTc >= 2*dr with CC = 78%

Logic: Double A5 v 3 if AA78mTc >= 2*dr. If AA78mTc is a large positive number then there is a deficiency of Aces, 7's and 8's left in the shoe . This means it is less likely dealer's hole cars is an Ace giving dealer and A3 and also less likely dealer's hole card is a 7 giving dealer a total of 10 and less likely dealer's hole card is an 8 giving dealer a total of 11. Also extra aces will make player A5 into a 17 which is the weak pat hand and extra 7's and 8's will make the A5 into a still So a deficiency of Aces, 7's and 8's helps the player doubling A5 v 3. And Aa78mTc large means extra Tens which means dealer will bust more often which is also good for the player doubling A5 v 3.

So LOGIC is important and the formulas found from SIMS or EoR analysis must make logical sense. This is a good check on the results.

[DSchles]

"Further equate the dealer higher bust rates of 23&4 versus bust rates of 5&6."

Is there an explanation for this egregiously incorrect statement other than you simply misspoke?

Don

[BoSox]

How about this. . To double soft 13 vs 4 has to be +3. To double soft 15 vs 4 has to be 0.

Sent from my SM-N950W using Tapatalk

On the soft 13 vs 4 requiring a +3 for doubling the dealer has a better chance of busting with that +3. Otherwise, by only hitting the hand you will want to be able to hit again for the second time after receiving an ace, 2,3 or 4 as the first hit card. You may also want to hit again for the third time on some hands.

With the soft 15 vs 4 in the times when you only hit the hand the chance where you want to re-hit again are basically only limited to receiving an ace or a 2 as a first hit card.

[Freightman]

"Further equate the dealer higher bust rates of 23&4 versus bust rates of 5&6."

Is there an explanation for this egregiously incorrect statement other than you simply misspoke?

Don

My bad. Mixed it up.

[Three]

It’s because you can hit the A3 another time if you get a low card. I don’t know if that’s 100% correct but I’m sure that’s part of it.

That's what I originallly pointed out in the link I posted. The hitting EV is what is different. Not the doubling EV. I like to let the curious see the data rather than try to think they would understand what data comes from logic. Often in BJ your logic gets you to the wrong place.

[Tarzan]

I'm not going too far into depth on soft doubles, since I've mentioned about this before at some point, but as was mentioned already in the thread, key cards can have quite an impact. You mentioned Hi-Lo +4 to double A,5vs3, but there is such a thing as being at Hi-Lo +10 with the optimal decision to hit. There's such a thing as Hi-Lo index -2 with the optimal decision to double A5vs3! The Hi-Lo index is sort of broad brush and if you have lots of (4,5) removed, lots of surplus (7,8) in the remainder, this weighs in quite heavily, pushing in the direction to hit rather than double. This can be calculated to the very exact, but is more like a fun-filled trivia fact unless you are able to do the required practical application at the tables using the composition dependent index on soft doubles.

[ZenMaster_Flash]

Post #21 is Tarzan's best post in years.
That is because it is concise and perfectly
understandable.

There are indeed counts like The Tarzan Count
and two (2) others that come to mind that
arm the user with P.E. between 0.8% and 0.9%

These are obviously not for casual recreational users.

[ZeeBabar]

So what should recreational card counters do with soft doubles? I don't know how much I am losing by doing soft doubles strictly with Basic Strategy unless TC is heavily (TC-4) negative and I just hit.

[Freightman]

So what should recreational card counters do with soft doubles? I don't know how much I am losing by doing soft doubles strictly with Basic Strategy unless TC is heavily (TC-4) negative and I just hit.

I ignore, depending on an increasingly crappy count, soft doubles in the following my hand order.
A2, a3, a4, a5, a6, a7. I choose not to double a8, unless I’m having a crappy game, but I must be in solid ground.

This is measured against dealer hand holding in the following order.
2, 3, 4, 5, 6.

So, for me, i will double almost (really keep and eye on dealer 2) everything in positive counts. Soft doubles have a lower success rate than hard doubles.

I really haven’t answered your question, since I would need to look it up, but the riskier doubles, those where you have a2 or a3, I would add a couple if points to the published index and make it more of a risk averse double.

[Stealth]

The easiest answer is because that is what the sim says to do. You base your indices on sim data, not logic.

+1

Best answer.

Best not fall into trap of what "seems" to be right and use what is right!

[Dbs6582]

+1

Best answer.

Best not fall into trap of what "seems" to be right and use what is right!

Of course this is true. I don’t think anybody is saying we should do something against the sim data.

Today I know all the correct plays without thinking, but when I was learning them I used logic to help me memorize the correct play. Maybe my logic was totally flawed, but it still helped me remember the right play. As with others, early on I wondered why one should hit some soft hands, mostly the lower total soft hands, and double on others. Even though it might be faulty logic, I rationalized it with the logic MidWest Player gave.