3:1 on suited blackjacks?
- Thread starter EasyRhino
- Start date
Even money on a 2:1 blackjack comes out differently than on a 3:2 blackjack.jimbiggs said:
On a 3:2, you bet 2 units. You can insure up to 1 unit. Let's say that's what happens. If dealer has BJ, you push on your original bet and win 2 units from the insurance. If dealer does not have BJ, you lose 1 unit (insurance) and win 3 (BJ) for a net of 2. Hence the even money (which I'm sure you knew, but it's for comparisson's sake).
On a 2:1, say you bet the same 2 units and insure for 1. If dealer has BJ, you push the main bet and win 2 units on insurance (like even money). If dealer does not have BJ, you win 4 units on the main bet and lose 1 unit on insurance for a net of 3 units. So you can't just "take even money." You still have to see what the dealer holds. To take even money you'd have to bet 1.333 units for insurance, which isn't allowed because you only made a 2 unit original bet.
So, if the casino offers you some bogus "even money" on the BJ, don't take it because you'll be giving up one bet unit every time the dealer doesn't have BJ. Just take the normal "insurance" if the count justifies it. And, as Sonny said, it's still totally independent from the main bet, so the count index stays the same.
ihate17
Well-Known Member
This should be interesting
We know that in the regular blackjack game, even money means insuring a blackjack and really nothing more.
So question #1, is the 3-1 payout on a suited blackjack, for all suited blackjacks or does it have to be a winning blackjack. If the promotion pays it on all, then taking even money would not enter the equation.
If it has to be a winning blackjack to get 3-1, my indice would have to be very high. Since on a 3-2 the hi low indice is +3, does anyone know how to figure what it should be when it is 3-1 or 2-1?
ihate17
We know that in the regular blackjack game, even money means insuring a blackjack and really nothing more.
So question #1, is the 3-1 payout on a suited blackjack, for all suited blackjacks or does it have to be a winning blackjack. If the promotion pays it on all, then taking even money would not enter the equation.
If it has to be a winning blackjack to get 3-1, my indice would have to be very high. Since on a 3-2 the hi low indice is +3, does anyone know how to figure what it should be when it is 3-1 or 2-1?
ihate17
ihate17
Well-Known Member
Not understanding this
Should not this difference change the index for insurance?
If BJ is 2-1, and even money is bogus. If you win the insurance bet, you still get even money, since insurance pays 2-1 but you can only bet 1/2 your bet max. So, $100 bet, you get $100 on even money or insurance. If you lose your insurance bet you get paid $200, minus the insurance loss of $50, or $150. In the regular game if you win or lose your insurance bet you would be getting just even money.WumpieJr said:
Should not this difference change the index for insurance?
You know I was driving around the other day and was thinking just that! I think I used -0.58%. What would be the base HA for your game?MGP said:
I think I got the 3 to 1 was worth 1.7%.
We're probably farther apart now than we were before lol. (Due to my math which is really what I am wondering about.)
I think I agree with what Sonny said (as usualihate17 said:
. Insurance is a side bet that pays 2-1 and has nothing to do with anything except whether the dealer has a 10. So I don't think it would change the index number.
They just could no longer say "even money" - something I suspect most people that take it don't even know why they're getting it.
Hi,Kasi said:
For 8D H17 DOA DAS LS SPL3 I get
TD EV: -0.553895516371016%
So the difference I get for suited BJ paying 3:1 is:
Suited Bonus - No Bonus EV: +1.861370835809486%
A little different than what you got. I know for sure that my calculations are exact without the bonus. I've never had anyone double check my values for the suited bonuses, but I believe they are exact based on calculations I did by hand...
A lot of my values for suited 678 and 777 differ from the Wizard of Odds' values in his appendix btw but I still believe I'm doing mine correctly.
Thanks MGP - I appreciate you taking the time to do all this me being simless and all.MGP said:
Differing from the Wiz - that takes confidence lol! Maybe you're using a different number of decks or something.
So you are saying your hand calculations agree with your 1.86% value for the bonus? I get a suited BJ will occur 1.18628% of the time in an 8D game.
It's hard to show the whole calculation without a spreadsheet. Yes I do get the same overall prob of suited BJ and please don't be silly and suggest I can't tell how many decks my CA is usingKasi said:
! There are actually several places I've disagreed with the Wizard and he's aware of them. To be fair though I've only focused on BJ and it's incredible how many different games he's covered in the amount of time he has - I have no idea how he goes through a game so fast.The key part though is how you're probably dealing with suited BJ's against Dealer A/10. That part is very tricky and it took me awhile to figure it out. I found a nice shortcut though... You have to figure out what the probability is of a suited BJ for each upcard, and if it's not a definite win you also have to account for the conditional probs of dealer BJ for both suited and non-suited hands. It can get very very confusing, but yes I do get a 1.86% exactly by hand. (I did it by hand just for your sake btw for this example to confirm and it worked out as expected). So it's not just as simple as multiplying 1.18*1.5 which is probably close to what you were doing.
I tried posting the details but it doesn't allow tabs so it's unreadable... I'll try a new post to see if it'll let me show them.
Last edited:
Ok - this worked better than an edit - here are the details (I needed the "code" tags):
Code:
N Cards A 10 non A/10
416 32 128 256
Net P(BJ) 0.047451344
Upcard A 10 non A/10
P(UC) 0.076923077 0.307692308 0.615384615
P(Dealer BJ given UC and Player BJ) 0.307506053 0.075060533 0
Relative P(Suited) 0.25 0.25 0.25
Relative P(Non-Suited) 0.75 0.75 0.75
EV(Suited BJ) 3 3 3
EV(Suited BJ Push) 3 3
EV(Non-Suited BJ) 1.5 1.5 1.5
EV(Non-Suited BJ Push) 0 0
EV = P(Suited)*EV(Suited) + P(Non-Suited)*(P(Win)*EV(Win)+P(Push)*EV(Push))
1.52905569 1.790556901 1.875
No Bonus Bonus
Net EV BJ A 0.079903148 0.117619668
Net EV BJ 10 0.426895139 0.550940585
Net EV BJ non-A/10 0.923076923 1.153846154
Net EV BJ 1.42987521 1.822406407
Net Difference EV BJ (Bonus-No Bonus) 0.392531198
Net Difference EV Overall given net P(BJ) 0.018626133
Last edited:
I just meant sometimes I've been fooled as to whether he's using infinite deck assumptions etc. I know you know what your program is doing.MGP said:Yes I do get the same overall prob of suited BJ and please don't be silly and suggest I can't tell how many decks my CA is using
Yeah it was the tied BJ part I have least faith in. No I wasn't doing 1.18*1.5.
Does it matter whether I have a suited BJ or not vs dealer A/10 or just how many pushed BJ's there will be?
I was thinking after there is a BJ, the prob of another BJ is 31/414*127/413.
Times 2 for each way. So .0230258*2=.046516. And .046516*.0118628=.0005463. So subtract that from .0118628=.0113165 would equal the number of times you get a non-tied suited BJ.
I did finally find, since we're into this, and I sincerely hope we are the only ones into this lol, that for a 6D, H17, DAS, DOA game with a suited BJ 2-1 payoff, Wiz gets 0.57% as the benefit. Would your system duplicate that?
MGP - I think I got it now.
Your sim
Base ev -0.55389551
EV after suited bonus .011435844
Dif=+.01697479. So that would be the value of the suited bonus, not 1.86%
My calc
0.0113165*1.5 extra units= the same number to 15 decimal places.
So I agree EXACTLY with your sims. I feel so much better now
Your sim
Base ev -0.55389551
EV after suited bonus .011435844
Dif=+.01697479. So that would be the value of the suited bonus, not 1.86%
My calc
0.0113165*1.5 extra units= the same number to 15 decimal places.
So I agree EXACTLY with your sims. I feel so much better now
It's NOT A SIM!!! I use Combinatorial Analysis (CA). I'm glad you figured out where the disagreement was.Kasi said:MGP - I think I got it now.
Your sim
Base ev -0.55389551
EV after suited bonus .011435844
Dif=+.01697479. So that would be the value of the suited bonus, not 1.86%
My calc
0.0113165*1.5 extra units= the same number to 15 decimal places.
So I agree EXACTLY with your sims. I feel so much better now
6D, H17, DAS, DOA, NS, SPL3
Base TD EV: -0.618148808389512%
TD EV 2:1 Suited BJ Always Wins: 0.056713552455954%
TD EV 2:1 Suited BJ Must Win: -0.051611461938709%
So it looks like the Wizard's value is based on a must win bonus.
OK it's a CA!MGP said:
I don't know what you mean by "Suited BJ Must Win" - to me it makes it sound like it will be paid against a dealer BJ. But the Wiz assumed that a suited BJ would not be paid vs a dealer BJ - it would just be a push.
So he gets the 2-1 bonus is worth the difference between your 1st number and 3rd number.
Did u figure out what was wrong with your 1.86% calc?
You're just trying to kill me aren't you! There is nothing wrong with my value. It's correct if the BJ always wins. If you read my posts you'd see I gave two values - one for when the BJ always pays 3:1 even against dealer BJ, and the second is for when a dealer BJ pushes.Kasi said:
Perhaps then you can tell me why, since you said we both agree on the frequency of suited BJ's at 0.0118628, the value of suited BJ's always being paid regardless of whether the dealer has a BJ, isn't 0.0118628*1.5=0.0177943?MGP said:
Clearly that's the value of suited BJ's if they are always paid. How can it be anything else?
Sorry if I'm killing you. Call it tough love
Perhaps you could read the post that I spent time typing out and formatting for you that explicitly details the calculationKasi said:
? It's because of the changes against A/T upcard and the conditional probailities usually used mixing with the unconditional probabilities needed. If you can show me an error in my calculations and convince me why it's an error then I'll change my CA. I'm not infallible and no one's checked my calculations yet, but I obviously don't think there's an error.LOLKasi said:Sorry if I'm killing you. Call it tough love
Last edited:
I read your post and it gave me a headache. I was hoping being armed with the sure and certain knowledge of the right answer would inspire you to find your mistake.MGP said:Perhaps you could read the post that I spent time typing out and formatting for you that explicitly details the calculation
If you agree that suited BJ's occur 1.18628% of the time, how can u disagree with the conclusion? Can you tell me that? It's axiomatic. How you can continue to think you have not made an error is beyond me.
Put another way, if ALL BJ's were ALWAYS paid 3-1 that would be worth 0.047451344*1.5=0.071177016 wouldn't it? Can you see that as a starting point?
If suited BJ's occur 1.18628% of the time, and we've already agreed they do, then non-suited BJ's occur 0.047471344-0.0118628=0.035588544 of the time. Can we agree on that?
Which would be worth 0.035588544*1.5=0.053382816.
So the difference between the 2 has to be what suited BJ's are worth or 1.7794%.
It's really pretty straightforward and much easier to do since you don't even have to worry about tied BJ's. The only think you need to find out is the frequency of suited BJ's. What else is there to find out if they always win?
Maybe you can do your CA thing with a base EV of all BJ's being paid 3-1 and see if it changes by the first number above and work from there?
Somewhere I think you're counting the 1.5 units twice on the value of a tied suited BJ. A tied suited BJ will occur 0.0005463029 of the time. Multiply that by 1.5 and add it to our original result of 1.69747 and you will also get the 1.7794%. You've added it twice to get your 1.86%.
For 1.86% to be correct, suited BJ's would have to occur 0.0124091 of the time wouldn't they? Since that *1.5 = your 1.86%.