3:1 on suited blackjacks?

[EasyRhino]

So, I saw a sign that a local casino was occasionally offering a promo where suited blackjacks pay 3 to 1. It begs some interesting questions.

- How much of an advantage would this yield off-the-top? It seems like it would be large.
- At what negative count would the game revert to having a house edge? (Let's assume 8D, H17, DOA, DAS, LS)
- What are the odds that the promo would be burned out if I name the casino on the interweb (I mean, they're advertising it in-house, so it's not mega-secret or anything).

 

[MGP]

So, I saw a sign that a local casino was occasionally offering a promo where suited blackjacks pay 3 to 1. It begs some interesting questions.

- How much of an advantage would this yield off-the-top? It seems like it would be large.
- At what negative count would the game revert to having a house edge? (Let's assume 8D, H17, DOA, DAS, LS)
- What are the odds that the promo would be burned out if I name the casino on the interweb (I mean, they're advertising it in-house, so it's not mega-secret or anything).

TD EV: 1.30747531943847%

 

[shadroch]

I wouldn't worry about mentioning it here. I've become convinced there are no more than a handful of people here that actually play the game.I'm sure it will produce a lot of fascinating posts on the advantages of such a game,but little action.

 

[QFIT]

I wouldn't worry about mentioning it here. I've become convinced there are no more than a handful of people here that actually play the game.I'm sure it will produce a lot of fascinating posts on the advantages of such a game,but little action.

1. People that play seriously don't post trip reports.
2. People that play seriously rarely talk about play at all.
3. Most people lurk on forums and don't post at all.

So, how would you know

 

[ihate17]

Broadcasting a positive EV game

If you have a casino promotion that gives you a positive game off the top, broadcasting it on the internet can result in the destruction of the game.

The casino makes a decision. They know that for example, Tuesday morning and afternoon is their slowest time, so they come up with a promotion. They do not advertise it on TV or the local papers, they announce it in their mailers to their existing customers. Those existing customers are the people the casino wants to entice in on Tuesday with the bonus. If the broadcast on the internet results in a bunch of unknowns, who play much better blackjack than their usual customers, someone just about always pulls the plug on the promotion. Sometimes it is even killed while it is supposed to be going on.
So I do not share special games. I have had request to share from several people in the past and if I happen to think they might have something to trade, I might share information, but otherwise spreading the word burns the game out in my opinion.

ihate17

 

[EasyRhino]

That's the concern, and I think your point about how means of advertising is a good one.

The counter-example I can think of would be Siena's (Reno) Happy Hour Blackjack. A really great promotion, that's still running after being blabbed on the boards here. However, that promotion is also structured in such a way as to be fairly well-protected against interlopers (weekend mornings only, max $25 bet I think).

 

[zengrifter]

TD EV: 1.30747531943847%

Suited BJs = +1.3%? Sounds high. zg

 

[MGP]

Suited BJs = +1.3%? Sounds high. zg

I assumed that the suited BJ always won. If it doesn't then I get:

TD EV 3:1 hand must win: 1.14358443558594%

 

[EasyRhino]

I assume the suited BJ is otherwise like a regular one, in that it wins against all but a dealer BJ. I honestly haven't inquired.

And remember, it's three to one... the 1.3% number seems reasonable, if extrapolating from the effects of 1:1 and suited 2:1 payouts from wizardofodds.

 

[Kasi]

I assume the suited BJ is otherwise like a regular one, in that it wins against all but a dealer BJ. I honestly haven't inquired.

And remember, it's three to one... the 1.3% number seems reasonable, if extrapolating from the effects of 1:1 and suited 2:1 payouts from wizardofodds.

Without simming it, I get prob of suited BJ about 1.19%. Prob of pushing BJ about 4.6% so prob of untied (winning suited BJ)=1.13%.

So you're picking up an extra 1.5 units 1.13% of the time. So if the house edge is -.58% your picking up an extra 1.7 % for an overall edge of -0.58+1.7=1.12%. More exactly, I actually get 1.13%.

So pretty close to MGP's sim but not as high as the 1.3% like Zen intuitively thought.

So I would assume around a TC-2 the game might revert to about even.

So does one wong out in a positive expectation game lol? Bet table max off the top lol? How greedy u gonna get ?

 

[EasyRhino]

Yep, the amount to bet off the shuffle is a conundrum, as constructing a "negative bet ramp", which I didn't even know existed until now.

And then I wonder if having a game with 3:1 on suited BJ would substantially increase the variance of the game, to the point where it would affect the appropriate betting amounts. I dunno, maybe.

There's the question of how the casino will deal with a player coming in at exactly 7 oclock and betting the maximum $100 amount of the promotion, while asking how many hands he can play, only occasionally dropping his bets while playing, and then increasing bets with the shuffle? And then leaving as soon as the promo ends?

And then there's the biggest question of if I will even be able to get myself out there on the days and times when it's offered. Too bad it's not a 24/7 thing.

 

[Kasi]

Yep, the amount to bet off the shuffle is a conundrum, as constructing a "negative bet ramp", which I didn't even know existed until now.

And then I wonder if having a game with 3:1 on suited BJ would substantially increase the variance of the game, to the point where it would affect the appropriate betting amounts. I dunno, maybe.

There's the question of how the casino will deal with a player coming in at exactly 7 oclock and betting the maximum $100 amount of the promotion, while asking how many hands he can play, only occasionally dropping his bets while playing, and then increasing bets with the shuffle? And then leaving as soon as the promo ends?

And then there's the biggest question of if I will even be able to get myself out there on the days and times when it's offered. Too bad it's not a 24/7 thing.

Tough controlling one's greed in those "the more you play, the more you win" games :grin:

But think of the cover of flat-betting table max until TC-2! Why be greedy - play the negative counts too and take your 1.13%

Bring your 10 girlfriends and have them all bet half table max if they can find a seat

The only thing you know for sure is that it won't last.

 

[MGP]

Without simming it, I get prob of suited BJ about 1.19%. Prob of pushing BJ about 4.6% so prob of untied (winning suited BJ)=1.13%.

So you're picking up an extra 1.5 units 1.13% of the time. So if the house edge is -.58% your picking up an extra 1.7 % for an overall edge of -0.58+1.7=1.12%. More exactly, I actually get 1.13%.

So pretty close to MGP's sim but not as high as the 1.3% like Zen intuitively thought.

So I would assume around a TC-2 the game might revert to about even.

So does one wong out in a positive expectation game lol? Bet table max off the top lol? How greedy u gonna get ?

Hi - just wanted to say that my value is CA not sim and should be exact for a fixed TD strategy...

 

[Kasi]

Hi - just wanted to say that my value is CA not sim and should be exact for a fixed TD strategy...

Thx MGP.

#1 I never have much confidence in my probability math - I was happy to get with 0.0001 of you

#2 Wouldn't a fixed TD strategy have errors in it? Like it would hit 5-card 16's vs a 10 when the correct play would be to stand - stuff like that?

 

[MGP]

Thx MGP.

#1 I never have much confidence in my probability math - I was happy to get with 0.0001 of you

#2 Wouldn't a fixed TD strategy have errors in it? Like it would hit 5-card 16's vs a 10 when the correct play would be to stand - stuff like that?

#1 - Nice job

#2 - How is it an error when it's defined to be TD? You'll note I specifically avoid using BS in my posts because it's imprecise and can mean whatever you want it to and there have been numberous arguments about what "Basic" means. TD is TD though and so there aren't any errors. The CD values in this case aren't much different anyways.

 

[Kasi]

#1 - Nice job

#2 - How is it an error when it's defined to be TD? You'll note I specifically avoid using BS in my posts because it's imprecise and can mean whatever you want it to and there have been numberous arguments about what "Basic" means. TD is TD though and so there aren't any errors. The CD values in this case aren't much different anyways.

Well I think of total-dependent as being that, given a certain total, say 16, vs a dealer upcard, say a 10, the strategy based on hitting/standing etc will always be the same regardless of how many cards the 16 is comprised of. Like you could have a 16 composed of 8 2's and it would still say hit because it's decision is based on the fact that the total is a 16. Don't know if that's what your program did or not. But I was thinking that might account for our huge difference

Just because BS is based on 2-card total-dependent totals doesn't mean it's 100% accurate for all situations. We all know it's not.

In any case, for questions like this, calculating the exact probabilities thru number of cards etc is exactly accurate and will always be better than any sim. Suited BJ's will only occur a certain amount of time. Likewise pushed BJ's. And the extra pay-off will only effect the house edge by so much. Regardless of whether someone misplays a 16 vs 10, or indeed every hand, or not. Doesn't mean I figured it out right though!

Although I was thinking the other day if maybe Easy Rhino, or the casino lol, might not have meant "3 FOR 1" instead of "3 TO 1". Since I have no idea how a casino could stay in business like this. Unless it runs this promo for an hour with a $5 max lol. Pros would fly in from China for a game like this.

 

[MGP]

Well I think of total-dependent as being that, given a certain total, say 16, vs a dealer upcard, say a 10, the strategy based on hitting/standing etc will always be the same regardless of how many cards the 16 is comprised of. Like you could have a 16 composed of 8 2's and it would still say hit because it's decision is based on the fact that the total is a 16. Don't know if that's what your program did or not. But I was thinking that might account for our huge difference

That is what TD is - but if you are defining a strategy to be that way then it's not a mistake - it's the correct strategy. That is not saying it's the optimal strategy which is what you're referring to.

Just because BS is based on 2-card total-dependent totals doesn't mean it's 100% accurate for all situations. We all know it's not.

I don't remember saying that my CA calculates TD strategies based on the first 2-cards... My CA takes into account every hand played and only if it's played based on the strategy in the correct proportions and determines the correct TD, 2-Card, n-Card and CD strategies. I only posted the results for the TD strat. The strategies I get are 100% accurate for their respective definitions with the exceptions of OBBO/BBO splits which are accurate to 5 decimal places and CD-post-split values since you can only brute-force those and the time required for is prohibitive.

In any case, for questions like this, calculating the exact probabilities thru number of cards etc is exactly accurate and will always be better than any sim. Suited BJ's will only occur a certain amount of time. Likewise pushed BJ's. And the extra pay-off will only effect the house edge by so much. Regardless of whether someone misplays a 16 vs 10, or indeed every hand, or not. Doesn't mean I figured it out right though!

The calculation of the value of suited BJ's can be done by hand as you did easily enough, but the differences come into play based on whether or not you are using the correct EV without suited BJ's to start with.

 

[jimbiggs]

I've been going around in my head trying to figure out at what count one should start taking even money with a suited BJ that pays 3:1 when the dealer is showing an ace. Anyone know how to figure this out? Or have any recommendations to help me find the answer?

 

[Sonny]

I've been going around in my head trying to figure out at what count one should start taking even money with a suited BJ that pays 3:1 when the dealer is showing an ace. Anyone know how to figure this out? Or have any recommendations to help me find the answer?

The insurance bet still pays 2:1 so the insurance/even money play would not change. Insurance/even money is a side bet so the 3:1 BJ payout does not affect it.

-Sonny-

 

[jimbiggs]

The insurance bet still pays 2:1 so the insurance/even money play would not change. Insurance/even money is a side bet so the 3:1 BJ payout does not affect it.

-Sonny-

I think I follow you, but my mind is a little fuzzy on this one. If you take even money, you give up on the 3:1 and you basically end up with 1:1 whether the dealer has a BJ or not. If you buy insurance and the dealer does have it, then you win your side bet and push on your initial wager. So you end up with 1:1. If you buy insurance, and the dealer doesn't have it, you lose the side bet, but you're still paid 3:1, so you end up with 2.5:1. So it would seem like it would be a good idea to buy insurance rather than take even money when the count dictates. The count for taking insurance doesn't change. Is my thinking correct on this?