Question about strategy table...

Discussion in 'Blackjack Tournament Strategy' started by rounder21, Oct 3, 2006.

  1. rounder21

    rounder21 New Member

    On p.131 of Wong's CTS, there is a table (BR1 stands on these totals if win both ways). I'm not understanding how this table is to be read. I know the columns down the left are opponent totals and across the top are dealer's up card. It says on p.130 that you should deviate from normal hit/stand strategies if you are BR1 and you win if both hands have the same outcome. But the table doesnt make sense. For example if opponent has 21 and dealer has 2, the table says 14. Does that mean you should stand on anything 14 or higher? or does it mean deviate from normal strategy when you have a 14, which wouldnt make sense because why would you hit a 14 and not a 13 in the same situation? Someone please clear this up for me.

    Thanks,
    Rounder21
     
  2. Hollywood

    Hollywood New Member

    Not sure if this is exactly what you're referring to, but it sounds a lot like Wong's 'push is as bad as a loss' theory, in which if you are in a MUST WIN situation (a la elimination hand/final hand in which you cannot LOSE or PUSH, but must absolutely win and win ONLY to advance), then correct strategy is to STAND on a 15/16 vs a dealer 10. i also believe numbers have been re-run in the recent past and that his suggestion to also stand on a total of 14 vs. 10 (your question) is now largely considered mathematically incorrect (hitting is the right play).

    -hd.
     
  3. London Colin

    London Colin Top Member

    Hmmm ....

    As I understand it, the table says when to stand as BR1 if you just need to match your opponent's result. It gives a figure for every situation, regardles of whether or not it is a deviation from basic strategy.

    But I'm struggling to understand the logic of that 14 too. The table shows 14 against a dealer 2 or 3 if BR2 has 21, and also 14 against a dealer 2 if BR2 has 20. The implication is that you should deviate from BS and hit a 13 in these circumstances.

    But why? You would think that since you now have to win the hand (at least for the BR2 21 case), you should just follow basic strategy. I'm confused.
     
  4. Monkeysystem

    Monkeysystem Top Member Staff Member

    Read it like a BS table

    The win both ways table is meant to be read the same way as Wong's other tables in his other books. In your example of opponent having 21 and dealer upcard 2, you stand on hard 14 and hit hard 13.

    You're trying to hit to a total of 21, or to beat the dealer's made hand by hitting on 13. If you hit 13 you bust only about 5/13 of the time - a little over a third. If you hit 14 you bust about 6/13 of the time - almost half. If you bust in this case you lose automatically. If you survive your hit and the dealer busts, you win.

    The logic here is if you stand on a stiff, you depend on the dealer to bust the deuce, only a 35% chance. If you hit the 13, your chance of preventing the swing as described above is greater than 35%. If you hit the 14, it's less than 35%.

    Hollywood has a valid point about the accuracy of the tables in this book. I've found a few discrepancies in them myself. The tables were developed using a simulation program that may not have used adequate sample sizes.

    Hope this helps.
     
  5. KenSmith

    KenSmith Administrator Staff Member

    That's not what rounder21 is asking about. Instead, he's referring to Wong's advice on how to play your hand to correlate with your opponent.

    rounder, In the particular example you site, BR2 has 21, the dealer has a 2 up, and the table for BR1 says '14'. Wong means that you should hit your hand until you have a total of 14 or higher. If you have 12, you hit. You also hit 13. You stand on 14, 15, 16, etc.

    I recently calculated my own table of this information, and my opinion differs from Wong's on many entries. I used infinite deck H17. For the example situation of dealer 2 and BR2 21, you should hit 12 but stand with 13.
     
  6. Monkeysystem

    Monkeysystem Top Member Staff Member

    Hair Splitting Percentages

    You don't necessarily need to win the hand. A pushed 21 also accomplishes your goal. True, the chances of that happening are a pretty low percentage. But the percentage differences in your expectations in hitting or standing on 13 and 14 are also pretty low, and the low percentage expectation of pushing 21 makes the difference.

    That is, if the table is correct.
     
  7. London Colin

    London Colin Top Member

    Still confused

    Ken, as I said above, I'm struggling to see how the correct play for this situation could possibly be anything other than basic strategy. With BR2 on 21, he is guaranteed a win or a push. Surely you can therefore disregard his hand altogether and play your own just as you normally would in order to maximise you chances of winning? (Which matches your calculated result.)

    I've read the Monkeysystem explanation since writing the above. Makes sense. Thanks Monkey!
     
    Last edited: Oct 3, 2006
  8. Monkeysystem

    Monkeysystem Top Member Staff Member

    Tournament Logic

    If Ken found that standing on 13 is the right play in this case, you can take that to the bank. Wong's tables have discrepancies in them.

    But here's the logic. In tournaments you often don't make playing decisions that maximize long term EV. You're looking to accomplish a specific goal in one hand. Let's use Ken's result as an example.

    Ken's result of hitting 12 and standing on 13 is driven by the goal of correlating your opponent. In this scenario you correlate your opponent by winning your hand or by getting a pushed 21.

    If you stand on a stiff, you're depending on the dealer to bust. If you hit the stiff, you're trading the risk of a double bust for the benefit of possibly beating the dealer's made hand or pushing 21. Ken's numbers must be the result of this tradeoff being favorable if you hit 12, but not if you hit 13.
     
  9. toolman1

    toolman1 Active Member

    More words

    Ken Smith pretty much answered this question but my answer is a little wordier and may help you understand better.

    The table is to be read: Stand on what is in the body of the table. If you hit when below what’s shown and as a result go over what the table shows, then stand.

    To clarify the purpose of the table:
    To quote from the page preceding the table: “If it is the last hand, you are BR1, and you win the table if BR2 and you have the same outcome, then you should deviate from normal hit/stand strategy in certain situations.” Now let’s think about the specific item that you have a problem with for a minute.

    Your opponent has 21
    You have 14
    Dealer shows 2

    This is a situation where you are hoping for a dealer bust as your most probable way of “having the same outcome”. You cannot beat your opponent (he has 21) so by not busting and therefore staying in the game, you have the best chance to advance in Wong's opinion. I can only assume that Wong concluded that hitting a 13 and trying for a total that would beat the dealer is better than standing on the 13 and hoping for a dealer bust. Whether or not the table is accurate or not is another question. Maybe it needs some tweaking.

    Also keep in mind that this table is a composite of some basic strategy plays and some deviations from basic strategy.

    Good luck if you try to memorize this table, I can’t.
     
  10. rounder21

    rounder21 New Member

    Thanks everyone...

    Now I am understanding what Wong meant. But how do I know whether to trust the results of wong or ken on the discrepancies? Ken I think you said you used infinate decks and H17 to simulate the hands. It is my understanding from what I've read in gamblers forum that the correct play actually varies slightly according to the number of decks AND penetration. Is this true? If it is, I suspect the differences are minor. In the tournaments I play they are 6D and penetration is down to about 1/4 of one deck. I'm not clear on the logic behind the reason(s) the correct play varies with penetration, but I remember the article I read (I think by Arnold Snyder) it said that the deviations from basic strategy in count systems will vary according to the penetration and that is why different simulations come up with different results. I'm going to hold off on memorizing this table until I am more clear. I would rather just play basic or vary with the count I have learned than to play off of a table I memorized that is incorrect. Its much harder to relearn something than to learn it right the first time. I suspect there is not much difference anyway in the gain one would obtain from knowing the correct plays in this situation, but every little bit helps. Memorizing the tables is the easy part for me. I'm a little cloudy on the logic but its more clear now the way a few people on this thread explained it as far as the reasons. Thanks again for your prompt responses.

    Rounder21
     
  11. RKuczek

    RKuczek Member

    pushes

    Rounder - remember that in Basic Strategy - pushes can be beneficial to the player in maxing ev - you may be more likely to push and less likely to win - but get a higher ev because you will lose fewer hands - both Wong and Ken are certainly taking this into account - your situation - a push is the same as a loss - so you don't want to max ev - you want to max the win probability - not the same thing -

    if Wong and Ken differ - trust Ken -
     
  12. rounder21

    rounder21 New Member

    Good point...

    Similar to what was stated earlier in this thread. The table I was referring to was the "win if both players have the same result" table. Different than the "push is the same as a loss" table. There are just minor differences from basic strategy in the "push is the same as a loss" table. And I guess the author (Wong) is referring to pushing with the dealer and not with the opponent in the "push is the same as a loss" table. For example you would stand on a 16 v 8, 9, 10. You would stand on a 15 v 10, you would stand on a 12 v 2 or 3 and you would hit an A7 v 8. I think thats right, I'll have to go back and check. But the idea is that if you are going to push with the dealer, you wont have enough to advance, so therefore youre striving to win and your logic makes sense when you explain it that way. B.S. is looking for positive EV. Your logic also applies to the "win if both players have the same result" table. If you push and your opponent wins, you do not win the table. I think. Its starting to become more and more clear.

    Rounder21
     
  13. BlueLight

    BlueLight Active Member

    Expansion of table 5

    My result when the dealler shows a 2 and BR2 has 21 is for BR1 to stand with a hard total of 13 or more for inf deck or 1 deck and for dealer hits or stands with soft 17.

    But what if BR1 has a soft total? Then BR1 should stand with a soft total of 19 or higher. An expanded table could have a double entry for this situation as follows.....H13 S19


    .............BlueLight
     
  14. London Colin

    London Colin Top Member

    I think all figures in Wong's book come from simulations, rather than being the results of a process of reasoning. So he doesn't really mean anything. It's instructive to try and think through the factors that must have given rise to the results though.

    Also, everything is predicated on the assumption that you are not card counting. Something as fundamental as the idea that you take the low against a single competitor is presumably only true because the house has the edge. If the count is such that the players now are more likely to win the hand than lose it, then I guess that instruction is no longer valid. (Not sure if a +1 true count has reached that tipping point though, since +ve EV may not mean > 0.5 chance of winning the hand - due to the influence of DDs BJs and splits.)
     
  15. rounder21

    rounder21 New Member

    Thanks for the responses...

    I think the count has to get EXTREMELY high for the player to actually be favored to win the hand, as you said, because of the influence of doubling and splits and the higher pay for blackjack that actually gives the player the money advantage in high counts. It is VERY rare for a player to be expected to win a hand even in high counts. Not sure exactly how high it has to be but I may look into that. I use counting for the first 20 hands just to give me an idea of when its a good idea to raise my bets or back down and wait for the count to rise again if it is low and for playing strategy variations in the first 20 (25 hand tournament). But I sometimes find that I should bet contrary to the count anyway if I am way behind or comfortably ahead. I've realized, as tourny experts have said, counting is unimportant. But when youre a counter, its hard to sit at a table and not count, and I figure it cant hurt. Also, heres whats kind of confusing for me. The tourny I play in pays 2-1 for a blackjack. Which in this case gives the player about 1.5% off the top I believe. The count has to drop to at least -2 TC for the house to have an edge. In the beginning I thought I would just bet high and gain a big lead and then stay ahead or drop my bets if the count dropped, but this didnt work. I was eliminated after the first 4 hands when I tried this and very frustrated. Still a risk of ruin even if you have an edge. hmmm. I made it to second round last week and almost third being pretty conservative until the end. I'm confident that I am about average as far as tourny skill where I play. So I am confident in betting lower until the end. But as a counter, I'd really rather play in a tourny where the casino has the edge off of the top (with play money of course), because competing against the other players, I would rather not have them have the edge when they are betting like wildcats. I would rather be the one to know when the count says its okay to bet like a wildcat even though most times its not. Thanks again...

    Rounder21
     
  16. London Colin

    London Colin Top Member

    Counting

    I'll defer to others here with much more expertise, but I don't think that 25 hands is long enough for a small player edge to make any real difference to the benefits of betting small.

    It's good to know that if you do need to make a big bet the odds may be slightly in your favour, but I think you still need to keep your powder dry and wait for the right moments to make a move.

    After all, a counter can go thousands of hands without seeing a return on their theoretical edge.
     
    Last edited: Oct 4, 2006
  17. KenSmith

    KenSmith Administrator Staff Member

    How'd you get a copy of that chapter of my book? :D
     
  18. BlueLight

    BlueLight Active Member

    BR1 Table

    Even more amazing is when the dealer has a 10 up and BR2 has 21 then BR1 should stand with hard 15 not 17 as Wong's table shows. This would be 1.08% better than hitting hard 15 to a total of 17 or higher.

    In Edward Thorp's book he gives tables for standing with hard totals and soft totals. So why not a table for soft standing numbers for tournament BlackJack?


    ................BlueLight
     
  19. London Colin

    London Colin Top Member

    While we've got our Wongs out ...

    I have a question about another table - Table 4, Probability of Win, Push, Lose.

    I assume these figures are the results of simulations, but I'm wondering if it is possible to come up with a formula from which such figures (or a reasonable approximation to them) can be derived.

    What I mean is something that takes the parameters -
    • Single player Win/Push/Lose probs
    • Number of players in each category
    • Factor(s) that represent the correlating influence of the dealer hand on the players
    - and churns out an answer.
     
  20. BlueLight

    BlueLight Active Member

    Table 4

    There might be a way to come up with more accurate figures for table 4 by some formula but there are too many traps to fall into and come up with erroneous results. The best way to get better results would be by computer simulation play millions of hands by computer program (10 million is good enough). Also a simulation that shows not just win, lose or push , but also win 2, win 1.5, win 1, push, lose 1, lose 2 to account for double downs, splits, and blackjacks.


    .....................BlueLight
     

Share This Page