Not splitting 8's if Surrender of offered...?
  • What is th corrent play? I hate splitting 8's against a 10...lol

    Enjoy th day, Jim
  • The math says:

    EV for splitting 8,8 vs dealer 10 is -.4807

    EV for surrendering any hand is -.5000

    Which would you rather lose? 48% or 50%? I'll take 48. :) Note this is for an infinite deck, which makes it most applicable to 6D/8D shoes. For SD/DD the numbers might change.

    basic strategy is always correct if you are not counting.

    It is correct maybe 80% of the time even if you are...
  • In a 6deck, Hit soft 17, DAS, DA2, Late surrender, I think the correct strategy is to surrender 8-8.

    The following website generates strategy tables for any set of rules. I hope that it is correct!

    http://www.blackjackinfo.com/bjbse.php
  • No Skunk, thats not correct. For 6D DOA DAS LS splitting 8's vs. 10 is -47%, and surrender is -50%. Its better to split them. HOWEVER... note this one important clue about variance...

    If you split 8's against a 10 fifty times, the 100 unit bank you have will contain 53 units when complete. With the same 100 unit bank, if you surrender 8's vs. 10 fifty times, the bank now has 75 units. To each his or her own.
  • I am fairly new to casino blackjack, but I have been reading the posts here for a while and have been learning some great techniques from all of you veterans, but I an a little confused about the following statement.

    If you split 8's against a 10 fifty times, the 100 unit bank you have will contain 53 units when complete. With the same 100 unit bank, if you surrender 8's vs. 10 fifty times, the bank now has 75 units.


    I understand that you would have 75 units if you surrender every time, but I thought that the -.48 was on the original unit bet and not the two units you risk after you split. I hope this would yield a final bank roll of 76 units at the end of 50 hands splitting every time.

    Could someone please clarify this? I have spent the last few months learning the Basic Strategy and went to AC last weekend and played the tables for the first time. I really enjoyed the trip and want to ensure I can play as long as possible with minimal loss on my next visit.
  • JessieRC – I think I can clarify it….You are correct that you will have 76 units after splitting fifty times. I think Nickels was trying to show that you are putting twice as much money at risk by splitting that led to your question.

    Let’s start over and use 100 times playing 8/8 vs 10:
    If you hit………………………..You will lose 54 units
    If you split…………..…..…...You will lose 48 units
    If you surrender…...........You will lose 50 units

    Therefore, the correct basic strategy answer is obviously “split”……but let me bore everyone with a common sense illustration of what the above numbers really mean to the average member of this forum.
    - Let’s assume 99% of the members play a $10.00 unit.
    - This hand will occur about 1.45 times per 1,000 hands.
    - Therefore you will have to play 68,965 hands to get the 100 occurrences shown above……….or 690 hours.
    - It will cost you $20.00 (two units) total to surrender instead of split.
    - So…..It will cost you 2.9 cents per hour to surrender instead of split.
    Bottom Line: If you are “uncomfortable” splitting and putting up the extra money, then go ahead and surrender. It is not a big deal in real play with this hand.

    Am I saying to deviate from basic strategy?....Absolutely not!.....All I am trying to show is that many of these “close” plays are incorrect mathematically, but really don’t mean much to the recreational/occasional player when you sit down at the table….and that’s where it really counts, not at a computer.

    Regards….Grifter
  • let me add another thought to Grifters accurate response.

    Many worry about "variance" which is an issue. N&B was illustrating the point that if you surrender the 8's vs a 10, you lose 1/2 unit every time. If you split, on occasion you will use 2 units, and we all know that "streaks" happen, meaning that you could lose 2 units repeatedly for a period of time that could be minutes or months.

    Other hands fit this. A4 vs 4 for example... .09 to hit, .12 to double. Difference is .03 better EV to double. But when you double, you plunk down twice as much money for the house to take as when you just hit. So for that slightly larger EV, you just racked up a larger potential variance (swing in your bankroll). You will either win twice as much or lose twice as much. Or, effectively, for this hand your betting unit is twice as large, while your bankroll remains constant. ROR calculations will clearly show that doubling your unit size increases your ROR.

    Many use "risk averse" playing indices when counting which simply say "when the EVs are close, very close, use the play with the smallest variance or risk, which means double/split less frequently than BS or normal counting BS departure indices suggest. Which reduces your ROR (and profit of course).

    If you play enough, you'll understand the "variance" issue. In my previous playing post, I am in the middle of "experiment number 5" trying to double a trip bankroll of $1000 before losing it. 3 of the previous 4 doublings were pretty quick, one was significantly slower. I am now almost 5000 hands into the 5th test, and am at $1645. I have been all the way down to $35 remaining where my last bet on a big + count was $35 and I won, and climbed back to well over the starting $1000 level, then back to $500 or so to where I am now. That is "variance". It can be sickening to have a TC of +10, and get three consecutive 15-16 hands to the dealer's 10 and have to surrender. Or get 12-13 five times in a row to the dealer's 10 up, and lose them every one, each with a max bet out. You can claw and fight for hours to get up a couple of hundred bucks, then lose it all in 3-4-5 consecutive hands where you have a significant advantage, but luck lies on the casino's side of the table for this shuffle...

    The issue of variance is an important one to come to grips with. Because it will come to grips with you whether you like it or not. :)
  • oops...

    I was thinking of 8-8 vs Ace. This hand you do surrender in 6 deck H17.

    Also, thanks for the illuminating variance discussion!
  • You are correct. I personally try to avoid H17 games, since most of the casinos I play seem to offer both in the shoe games, and in the better SD/DD games, S17 is pretty normal. Also A7 vs 2 is different in H17, among other oddities...

    I think all the numbers I quoted were for S17 games if I didn't lose my mind...
  • [Many use "risk averse" playing indices when counting which simply say "when the EVs are close, very close, use the play with the smallest variance or risk, which means double/split less frequently than BS or normal counting BS departure indices suggest. Which reduces your ROR and profit of course). ][/quote]
    stainless steel rat per your comment above where could one find the "risk averse" indices?? What is the cut off point for a close EV??
    When using a progression (which I do) when I have a big bet out, to protect the BR, the risk adverse may be the best play
  • That's a good question. I've never been a "risk averse" user myself, but I know people that do, and to be honest, I don't know where they got their indices. I believe a couple of well-known BJ players have posted some risk-averse indices from time to time.

    you can use CVData to compute your own indices, and you can choose the play that has the minimum variance when the EVs are "close" (You have to define "close" of course, as if the EVs are too different, you can affect your bottom line more...)
  • For 6D or 8D DOA DAS LS games The worst losers that do not surrender are 15 vs. 9 or Ace, the afore discussed 88vs.X, 77vs.X, and Hard 17 vs.Ace. Personally I only toss in the 88vs.X due to the variance when wagering 2 or more units (NOTE: this could be at RC=0 or +1 or even +2). At A5 RC of +2 or any 4 unit bet I will also toss in the 15 vs. 9&Ace.

    As always, Your Methods May Vary. The five hands listed here I call the "Gang of Five" on my website.

    Also check carefully your H17 charts for 4,6&8 decks with Surrender. The hand 77vs.X & Ace is close as is 88vs.X and 15vs.9. The balance between 10's and A234 affects the Hand Soft 17 vs. 2. An excess of the latter may indicate a DD of the hand. Drawing a 10 is a loser, the A234 draws are clear winners... statistically, that is ;o)
  • Here's my surrender plans, based on HiLo:

    16 vs 9-10-A, always surrender
    15 vs 10, surrender at TC >= 0
    15 vs 9 or A, surrender at TC >= +2
    14 vs 10, surrender at TC >= 3

    The 88 vs 10 I vary. At negative/zero counts (which means min bet out) I split or not randomly, to confuse anyone watching. At TC > 0, where I have a bigger bet out, I surrender to reduce variance. At higher counts, splitting seems to produce more variance (and often more losses, such as when I split, get a 10 on each, but the dealer has a 10 down which was predictable with the high count). 88 vs 10 is almost a coin toss on surrender, split or hit. Surrender loses 1/2 bet, splitting can win or lose 2 bets, hit can win or lose 1. I don't particularly like risking 2x the money for a very small potential return. It is nice to win 'em, but when you lose 'em, you just tossed out an hours worth of work or more... :)

    I tend to follow BS and I18 departures for most other hands, although there are a few that probably need some analysis based on risk. I've been working on a general rule for some things like splitting at way negative counts when I have to "play all". Splitting 4's vs a 5 is good at reasonable counts, but at negative counts, it loses more than it wins when I do it. But I don't want a bunch of indices to deal with so I'm working on a few very general rules to keep it simple rather than a specific index for each possible situation which is messy. One case in point is 9 vs 3. At - counts, this loses more than it wins, while at + and zero counts, it wins more than it loses. There ought to be a couple of good numbers to remember such as "at TC < X, forget doubling 9's, if that is possible...

    Obviously the things at + counts are most important, because that is where the most money is on the table... But since 1/3 of the hands are at - counts, some plans there are useful...
  • Found the following on the wiz of odds web site:
    These plays are close, thus they may be candidates for not putting more money out :?:

    My wife knows blackjack basic strategy but insists on being a 'hunch' player. What are the hands that are closest to 50-50 so that my wife can limit her hunch plays to those?
    I'm pretty sure this has already been answered but it is a good question so I'll take it again. Here are some borderline plays, as evidenced by having a card counting index equal to zero. This list is for multiple decks where the dealer stands on soft 17 and double after a split is allowed. The best play is listed first and an alternative second.
    · 9 against 2 (hit/double)
    · 12 against 4 (stand/hit)
    · 12 against 6 (stand/hit)
    · 16 against 10 (hit/stand)
    · Soft 13 against 5 (double/hit)
    · Soft 15 against 4 (double/hit)
    · 3,3 against 2 (split/hit)
  • I like the example of Soft 15 vs. a 4... you have a winner, and make it a better winner when you DD. Even if you hit, its still a winner. 8-8 vs. a 10 is SUCH a loser to begin... and you're asked to split. I opt for a "financial surrender" as opposed to the "strategy surrender" when I have a 4 unit bet ALWAYS, and when struggling or losing with a 2 unit wager. IMHO I'd rather save the 1/2 bet left-over from surrendering 88 vs. 10 for a better hand. For 1 unit , yes I ALWAYS split.

    The basic book says split'em and I agree. It gets outta my comfort zone at times.
  • I agree with Nickels thinking. "It gets out of my comfort zone at times."
    My play would depend on the count...hit if low...stand if high... I never split against 8,10,A...especially if I have two green in the circle.

Howdy, Stranger!

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