Long term/short term
  • Everyone here keeps insisting that the only thing that matters is the long run and short term results mean nothing.
    That being the case,why do some some BS moves call for whats better in the short run,rather than in the long run?
    Thare are some card combos that,by doubling down,you reduce your long term winning rate,but by doing so you increase your short-term winnings.
    If eveything should be played for the long run,why isn't every BS move geared towards the play that has the highest winning percentage?
  • NYB - That doesn't sound correct (or I'm missing something). Can you give an example or two of the double downs that increase short term winnings but reduce long term win rate?

    Grifter

    p.s.....Not "everyone" has insisted that short term results mean nothing.
  • NYB said:
    Thare are some card combos that,by doubling down,you reduce your long term winning rate,but by doing so you increase your short-term winnings.
    If eveything should be played for the long run,why isn't every BS move geared towards the play that has the highest winning percentage?


    That's not right NYB. There are some double down moves that will cause you to win LESS HANDS but MORE MONEY. I don't think it's possible that something that would increase your short-term winnings would reduce your long-term win rate.
  • I'd like to see an example of a "BS strategy" that favors short-term over long-term. Every example on every BS card I own can be traced right back to "max EV" period, except for those "risk-averse" type BS and index plays where the goal is to reduce variance rather than maximize EV. But pure BS is simply based on "what produces the best return on a unit?" and this is _always_ calculated as "long-term" since short-term is meaningless. Ever seen a case of 12 vs 10 where you wished you had stood, because you drew a 10 and broke, while the dealer turned over a 6 and would have broken with your 10? That is short-term. And it is luck. I've personally had too many short-term wins and losses, but I really care about the "trend" which is a long-term concept.

    I don't see how anyone can possibly play the game and care about "short term results." Why not just flip a coin, or bet red or black on the roulette wheel and have a better chance of a good "short-term" result.

    Betting on the "short-term" is kind of like saying "my team won the last three quarters of the game 10-3, although they lost the first quarter 28-0. The long-term (game) result is really the one that matters. And don't go down the dark hole of saying "what about the O/T game? The short-term (OT) result was the one that mattered." No it wasn't. Without the previous 4 quarters, that result would not have happened...
  • LeonS - Yep, I think that is what NYB is looking at (less/more). That's why we need an example of a couple of the hands.
  • Leon/grifter;
    Yes,thats what I'm talking about.I don't have my charts in front of me,but will post some examples later. In a nutshell they are hands that by DD,you decrease your chances of winning that hand,but increase the money you can win on that hand.If you are truly only concerned about long-term results,it would seem that the hand with the largest winning percentage would be the way to go,not the hand with the largest payout.You can increase the payout of any hand by raising your bet.I'm not aware of anything that will
    allow you to increase the odds of you winning a particular hand(except for not following BS in these cases).
    By lowering the chances of your winning a hand in return for being able to double your winings on that hand,you have left the long-term zone and are now focused on a single hand-the very thing most of you say is irrelevant.
  • Perhaps this is an example:

    A/7 vs 3 (Double deck, Hit 17, no surrender)

    I play in tournaments with 14 hands per round so I'm definitely concerned with "short run".

    Sometimes I have a big bet out and get A/7 vs 3 (or 4, 5, 6 for that matter). The correct BS move is to double but I might not want to risk any more money (maybe it is the last hand and I already have the optimal bet) I'm concerned with my best chances of winning the hand.

    I think I'm more likely to win the hand by just standing than by doubling. Is this correct logic???
  • Skunk - Yes, if you are just concerned about just winning the hand, then standing would be the correct move. For A/7 vs 3, if you stand you will win 58% of the time, if you double you will win 54% of the time.

    Grifter
  • A7 vs 3 is a good double, but A7 vs 2 is not a good double. Why so? The
    difference is the amount of difference between doubling and standing win percentage.
    When you dd A7 vs 3 you win more money and effectively
    lower the HA in the process. When you stand on A7 vs 3 you win less money
    and effectively increase the HA. This applies to short-term, mid-term and any
    other term. The same may not apply to those special considerations described
    by Skunk.

    Playing your DD hands correctly reduces the HA by 1.50%
  • NYB said:
    Leon/grifter;
    Yes,thats what I'm talking about.I don't have my charts in front of me,but will post some examples later. In a nutshell they are hands that by DD,you decrease your chances of winning that hand,but increase the money you can win on that hand.If you are truly only concerned about long-term results,it would seem that the hand with the largest winning percentage would be the way to go,not the hand with the largest payout.


    You're still confusing the two. Take A7 v. 3 for instance. I don't know the math off hand but let's say you get this hand 100 times. If you stand, you may win it 75 times and if you double you may win it only 60 times. With $10 bets, those 75 wins would net you $500 after 100 hands. The 60 double wins would net you $800. Now I'm just making these numbers up but it's got nothing to do with short-term/long-term. The object is to make the most money, not to win the most hands, be it short or long term (which are really the same).

    Tony Gwynn had a batting average of around .330 for his career. If he were my batting coach I don't think I'd say, "I don't care about hitting .330 over thousands of at-bats, I want to get a hit NOW."

    Sorry for the above comparison. I suppose it only works if you're a baseball fan.
  • Leon- Your numbers are closer than you may think. The expected return
    on A7 vs 3 with B/S:

    Hit= Plus .0902
    DD= Plus .1776

    With my magic deck I'll deal out (10) A7 vs 3. Is there any player that would
    not DD on each and every hand?
  • A couple of incorrect definitions are being tossed about...

    1. If you are interested in winning a _hand_ and not in maximizing your EV, then there is _no_ double down that is better than hitting. Not a single one. Because for any double down you can identify, there is a card you can get that would leave you wanting another card. But that misses the point that on hands where you double down, you already have a positive EV, and anytime you have a positive EV, you'd naturally prefer to get as much money on thet table as possible. And since you have already placed your bet, doubling down is the only option to increase your bet.

    2. Double downs are _not_ about "short term". They are about maximizing EV on _that_ particular hand, not on the "long-term". It doesn't take a lot of math skill to understand the idea that anytime we get two cards vs the dealer's up card, where we stand to win more often than we lose, we want more money down. And the decision becomes do I (a) hit, which has the best chance of winning the hand since I can draw additional cards if desired; or (b) double down, which decreases my winning chances for the hand since I only get one card, but I get to double my bet.

    A simple example, A2 vs 6. Wizard says that the EV for hitting is +.167, while doubling is +.180. But since we doubled our bet, had we just hit it one time our EV would only be +.09, which is worse than hitting. But by getting 2x the money on the table, we double that hit one time EV. Clearly I'm going to lose A2 vs 6 more often when I take a single card, than if I hit so that I can hit again. But the EV is higher.

    Nothing to do with short-term or long-term. It is _the_ way to maximize profit on _that_ particular hand, nothing more, nothing less. That's all BS ever is. For the conservative player, you might choose to look at the EV for hit vs EV for double, and if they are within some small delta of each other, hitting rather than doubling decreases bankroll variance since you end up betting less. But there is a definite cost to EV for doing so. so-called "risk-averse" index strategy departures are based on this same idea. For those that would prefer to slightly reduce overall EV in return for reducing overall variance as well.

    I'm not quite sure where the "BS is about long term" comes from. It is about as "short-term" as short-term can be, namely to maximize the outcome for a single hand.
  • Ray:

    That is _exactly_ the point of course. So long as one understands that +.16 is > +.10, they would take the play that delivers +.16. Hopefully every time. Without any thought about "long-term or short-term" at all, because this is about "one hand".
  • SSR, that was my point exactly that this discussion has nothing to do with short vs. long term, nor is there any difference between the two. What's got in the short term is good in the long term.
  • Okay,I'm beginning to get the idea.Math is not my strongest point. I've pretty much just memorized the BS charts and the top few changes due to count,and a few composition changes.
    Is there a chart somewhere that gives the exact EV for every three card set?
    I generally just go by the-"The next card will be 1 of 13,if 7 or more cards help my hand or are neutral, thats the right move" school. But in several DD situations,thats not the case.
  • Ray said:
    Leon- Your numbers are closer than you may think. The expected return
    on A7 vs 3 with B/S:

    Hit= Plus .0902
    DD= Plus .1776

    With my magic deck I'll deal out (10) A7 vs 3. Is there any player that would
    not DD on each and every hand?

    Ray- Your numbers are right of course, but before someone asks, let's add this to clear it up. The logical choice here would not be between "Hit" and "DD", it is between "Stand" and "DD".....If you stand, the EV is about .16, so we are actually comparing .16 to .1776......but the answer is still the same; always double down.
  • Grifter- You are correct as usual...Thanks
  • NYB:

    go to www.wizardofodds.com.

    click

    go to the bottom of that page and click

    That appendix has charts for every possible 3-card scenario (two for you, one for the dealer.) There is a separate chart for hit, stand, double and split (where appropriate). You can pick any starting hand + dealer's up-card, and find the EV for any legal playing decision to see how close some really are.

    Back to the "risk-averse" idea. hit or stand is pretty irrelevant to that discussion, but splitting and doubling is right on target, since you have to double your current bet and increase the "risk". Each decision has one perfect EV-maximizing option, unless you find two decisions that have the same EV to 3 dec. places, the number of digits Mike's site contains). Some are so close as to be almost indistinguishable (remember that these numbers are decimal percentages, so an EV of .013 means 1.3% advantage, or $1.30 for every $100 bet, so very small numbers mean very small EV changes).

    Hope that helps. Note at the top of that appendix the assumptions he made. The most important is S17. The numbers change each time a rule changes of course...
  • stainless steel rat said:
    Some are so close as to be almost indistinguishable (remember that these numbers are decimal percentages, so an EV of .013 means 1.3% advantage, or $1.30 for every $100 bet, so very small numbers mean very small EV changes).


    SSR - Your math is wilder than your war stories Would you care to correct that for all the good people?......How about 1.3 cents instead of 1.3 dollars for every $100 bet?

    Grifter
  • So,for the $5-25 crowd,we'd need to play several brazillion hands before that particular decision
    costs us more than a single bet.Interesting.
  • In BJ, there is very little difference in being on the negative side (.5) or
    positive side (.75). To have the best chance, you can't afford mistakes
    and must play each hand with the best science available, B/S and/or CC.
    Every individual card that falls has some impact and small impacts are the
    name of the game because they accumulate. With the exception of blackjacks
    , we will see these small differences accumulate in hit/stand, splits and DD
    decisions. No mistakes allowed...........
  • Grifter:

    am I missing something?

    With an expected value of .013, which turns into a 1.3% gain, is not 1.3% of $100 going to be $1.30???

    For example, when I pay an 8% sales tax here, I end up paying $8.00 for every $100 I spent. :)

    Or are we talking about the EV for a single hand, and then saying a hand is 1/100th of an hours play, so changing the EV for a single hand is only 1/100th the actual value we expect to win/lose over the course of an hour?

    Maybe my explanation was not clear. But on a single hand, if I have an EV of .013, I expect to win $1.30 every time I get that hand and bet $100.

    Clearly, for any specific hand, you have to take the EV, times the average bet, to compute the average win for _that_ hand. Then if you want to ask "OK, how does this affect my overall EV, I'm not interested in the EV for this particular hand by itself" the correct answer becomes that EV times the percentage of played hands that specific hand should come up. So maybe that is where you were heading with this? This is probably more accurate than the 1/100th approach, since it is pretty easy to compute the probability of a specific set of 3 cards being dealt.

    Otherwise I am lost, since an EV of -.5 means I expect to lose exactly 1/2 of the bet made on that hand, or 50% = .50. 1% = .01. etc...

    Or maybe it is well beyond being too late here on New Year's eve?

    BTW, did visit the MS coast today. Highway 90 looks like a war zone. Getting around is extremely difficult, and old routes no longer work. We didn't go in via the interstate since we were in Laurel and drove down to Hattiesburg and then down highway 49. We decided to come back home as it was about as busy as one would imagine. We spent more time looking at old landmarks from when we used to live down that way, than we did trying to figure out how to get to one of the two casinos that appear to be functional according to the news.

Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!