BS Deviations: Not Increasing Your Bet

acercher

acercher

Member
Hi Folks--I've been reviewing the archives for threads on situations that call for deviations from Basic Strategy and learning quite a bit. However, almost all such threads focus on times when you need to get more money on the table and therefore you're impelled to take actions that BS would deem to be sub-optimal if you were playing regular blackjack.

But I'm curious if the same logic applies in the reverse situation -- times when BS would dictate splitting or doubling down, but for whatever reason you may be reluctant to increase your bet. I can think of two scenarios (and there are probably more), but can't come up with a rule of thumb to deal with either.

First scenario: Last couple of hands, and you feel you've sized your bet correctly. But you're dealt a pair of aces or eights, or an 11 against the dealer's 6, or a similar situation where it's clearly beneficial per BS to double your bet. I can see not doubling the 11 to avoid putting new money at risk, but I have a very hard time not splitting those two pairs in order to improve my chances of winning even if it does muck up my carefully calculated bet. Is this a situation which only applies to spitting pairs, and if so, only those two "defensive" splits?

Second scenario: Middle of the final round, and you've started a progression taking into account the 500 max bet, such as betting 75, then 150 if the first bet is lost, then 300 if necessary. You have 150 or 300 on the table when you get pairs or a double down situation where BS calls for doubling your bet. Obviously, if you lose, you've screwed-up your progression, and may have fatally compromised your bankroll. On the other hand, it seems crazy to ignore a chance to increase the odds of winning money when that was the point of initiating a progression in the first place. Is this another situation where only a "defensive" splitting of aces and eights makes sense, or does another logic apply than was used in the first scenario?

Thanks for your thoughts!

Acercher
 
One distinction needs to be made between increasing your expected value (EV) or increasing the chance of winning the hand. These are often do not correlate. Basic strategy is based on increasing EV because it is simulated using million of hands to be similar to the long run. In tournaments, chance of winning a particular hand is more important than the long run. Your bankroll is compared to other bankrolls at the table, so there is no really long term accumulation of chips. Basic strategy recommendation for doubling or splitting increases EV mostly by increasing the amount of money being wagered on specific hands either to win more like 11 vs 6 or lose less like splitting 7s against 2. Often this is accompanied by either no significant change in the odds of winning the hand or even reducing the odds of winning the hand. The later can be caused by the limitation of getting one card once you double your bet. The second scenario you gave illustrates this. If you are doing a progression and lost 75 and then bet 150 and got 11 against dealer 6. This hand has probably the biggest EV if you double down. Because you will likely take one card anyway, the odds of winning the hand would practically not changed. However, if you lost the hand, you can bet 300 and be ahead. If you lost the double, then it is hopeless. The chance of losing a hand is 48% in general. So, the chance of losing 2 hands in row is 23%.
 
The_Professional has it right. Your goal is not to maximize EV of your hand. but to increase the odds of making something happen, whatever that "something" turns out to be. The examples related to getting more money on the table do not contradict this, because in those cases, that "something" is to win more than your original bet.

There are other goals which are common and which can involve changes to basic strategy, but are not (directly) related to getting more money on the table. Check out this thread in which some of us worked together to develop a strategy for when your goal is to maximize your chances of winning at least the value of your original bet (i.e. don't push and don't lose).

https://www.blackjacktournaments.com/threads/strategies-for-must-win-one-bet.8201/

In that thread I presented some work I did on generating some strategies for your initial hand. The discussion then went on to develop a more complete strategy. In a similar vein, here is my initial hand strategy for when you want to maximize the possibility of a win or a push (i.e. don't lose).

http://gronbog.org/results/blackjac...one/generated/complete/0.0/p1.X/strategy.html

and my initial hand strategy for when you need to win at least twice your original bet (i.e. when to double and when to split).

http://gronbog.org/results/blackjac...one/generated/complete/2.0/p1.X/strategy.html

When executing a progression, do not double or split. Your goal for these hands is not to lose, so use my "don't lose" strategy, unless it's the final hand, in which case use the "must win" strategy.
 
Another time we deviate from basic strategy without necessarily putting more money on the table is when trying to correlate the result of an opponent. For example, if you need to get the same result as another player. One obvious example is that if your opponent stands stiff, then you would also stand on any hand to guarantee the same result. There is a table in Wong's book showing the strategy for correlation.
 
Thanks for your reply, Professional. If I'm understanding it correctly, one key point you made is that doubling or splitting per Basic Strategy doesn't necessarily increase your percentage chance of winning that hand; it often just increases the money in play for that hand. Increasing the money at stake improves the overall Expected Value (which is why Basic Strategy calls for it), but in a tournament the immediate situation dictates whether you want to stick with your original bet or put more money at stake. Correct?

Question for you or anyone else: is there any link or reference you can point me to which would rank split pairs and double down situations where the chance of winning is actually and materially increased, in contrast to other opportunities where there is no or minimal improvement in the chance of winning?

Thanks--Acercher
 
Correct-
I think there several BJ books that show that information. Blackjack theory by Griffin is one. It is also possible to be in Modern BJ book by Norm, which is apparently free from his website.
 
Thank you for the links and information, Gronbog. A couple of questions:

1. Is the information in your link to "strategies for must one win one bet" intended to be the same as the strategy generally known as "a push is as bad as a loss"? It seems like the concept is the same, but the information differs. As I understand the "push is as bad as a loss,'' the difference from BS is that you stand on 12 against a dealer 2 or 3; stand on 16 against a dealer 8 or 9; and stand on 15 against a dealer 10. It seems like I'm either misreading your table, or misunderstanding the "push is as bad as a loss" strategy, or the two phrases actually refer to two different strategies. Your verdict?

2. Am I reading correctly that your table for "when you must win at least twice your original bet," says that with a pair of aces, it is preferable to double down rather than split them if the dealer has a 2-9? It seems extraordinary to me that it's better to take one card to a 12 total instead of taking one card each to two aces. What am I missing?

3. Finally, you wrote not to double or split when executing a progression, and to use your "don't lose" table. But that table contemplates splitting almost every pair except 5 and 10. So I assume that I would just treat a pair of 7s as a 14 total, and look to see how to play a hard 14 against the dealer's up-card. But is that also your recommendation regarding how to treat a pair of aces, i.e., as a hard 12?

Thanks for your advice--Acercher
 
acercher said:
Question for you or anyone else: is there any link or reference you can point me to which would rank split pairs and double down situations where the chance of winning is actually and materially increased, in contrast to other opportunities where there is no or minimal improvement in the chance of winning?
Click to expand...
No double will increase your chance of winning the hand overall. The reason is that you give up the ability to hit again, if necessary. In fact, you will see that there are no doubles at all in my "Must Win" and "Must Not Lose" tables. If you're doubling in a situation in which you would only take one additional card anyway (e.g. 11 vs 6), then your chance of winning the hand stays the same.

When it comes to splitting. There are non-basic strategy splits which increase your chance of winning the hand. All of the non-basic strategy splits on my "Must Win" table are examples, provided you go on to play the proper strategy after splitting (see S. Yama's strategy in the same thread).
 
acercher said:
1. Is the information in your link to "strategies for must one win one bet" intended to be the same as the strategy generally known as "a push is as bad as a loss"? It seems like the concept is the same, but the information differs. As I understand the "push is as bad as a loss,'' the difference from BS is that you stand on 12 against a dealer 2 or 3; stand on 16 against a dealer 8 or 9; and stand on 15 against a dealer 10. It seems like I'm either misreading your table, or misunderstanding the "push is as bad as a loss" strategy, or the two phrases actually refer to two different strategies. Your verdict?
Click to expand...
Yes, it's the same situation. I don't know where you got your BS deviations from. If it was from Wong, then realize that my work was done many years after Wong's with much faster computers. I believe that my simulations were more complete as a result (more iterations = more data). I trust my own work. Whether you do is up to you. The additional hard hand difference I have from your list is to stand on 16 vs T. Wong did not consider soft hands or splits in his work.

acercher said:
2. Am I reading correctly that your table for "when you must win at least twice your original bet," says that with a pair of aces, it is preferable to double down rather than split them if the dealer has a 2-9? It seems extraordinary to me that it's better to take one card to a 12 total instead of taking one card each to two aces. What am I missing?
Click to expand...
The problem with splitting is that it results in more overall pushes (one hand wins, the other loses). This is especially true for splitting aces, where you only get one card on each split hand. Pushes are no good in this situation. It turns out to be better to get the money down and hope to win the single hand.
acercher said:
3. Finally, you wrote not to double or split when executing a progression, and to use your "don't lose" table. But that table contemplates splitting almost every pair except 5 and 10. So I assume that I would just treat a pair of 7s as a 14 total, and look to see how to play a hard 14 against the dealer's up-card. But is that also your recommendation regarding how to treat a pair of aces, i.e., as a hard 12?
Click to expand...
The splits on the Don't Lose table are a good example of what I was talking about above. They generate extra overall pushes which are ok in that situation. As for playing a progression, yes, I should stipulated to ignore the splits in the Don't Lose table. I kind of did when saying that about the progression in the first place, but thanks for clarifying things.
 
I should add that, if you hover your mouse over the strategy for given cell in my tables, you will see the success rate for the various plays available. Some are extremely close. Splitting or doubling A,A vs T and A are examples of this when trying to win 2 bets.
 
You must log in or register to reply here.
Back
Top