11 vs A strategies
I've been looking at 11 vs A, and i know the index says you can start insurance at >+3, and i know at >+1 you can start doubling that 11. My question is if you have a deck that is at a TC of +4, and you insure and then also double down your bet, what can you expect your per hand output to be? According tooverall time, (simulation below) if you hit, your per hand net is -.201 units. If you double the 11, your output is -.211 units per hand. I calculated this out as final cash divided by $10 (the per hand bet amount) and dividing that by total hands (25000). I used these numbers to know whether i was on the right track with the experiment i will describe below.
6 decks, double allowed after split, dealer stands on soft 17
Hand: 11 vs A
Player 1 Action: Hit
Starting cash $1,000, starting bet $10
Results:
games played: 25,000
games won: 9,005
final cash: -$51,400
Player 2 Action: Double/Hit
Starting cash $1,000, starting bet $10
Results:
games played: 25,000
games won: 8,871
final cash: -$53,130
Firstly, is there a flaw in this method? And secondly, what would you expect from a +4 tc deck where you insure AND double? I did a small simulation of my own using 2 decks, removing half of one deck, making sure that half deck made the rest of the deck =+6 so the true count was +4 (+6/1.5=+4TC). Then i pulled an ace, a 5, and a 6 from the discarded half deck creating my 11 and A hands. Then i shuffled 6 times, cut the cards randomly, and the first card was the doubled card on the 11, as it would be assuming a random hole card from a +4 deck where already out, and the dealers up card were and A. This ensured the true count was still precisely +4 when the double card came out. Then i played the dealer hand. I then randomly re-inserted the used cards and shuffled 6 times, and cut the deck. I repeated this 50 times. I kept track of wins, losses, losses due to blackjack (therefore winning insurance side bets, and not allowing the double bet), and pushes. Then i calculated my net cash amount if my original bet were $10 (double being $20, insurance side bet being $5) and calculated my net per hand average. I came up with -.176 per hand. Is there anyone that can tell me how accurate that number is? I accept that because i didn't reset the deck for the dealers hole card, and i didn't randomly re-stack the deck for each hand at +4, that there is error involved, but i'm hoping my number is close. In addition, can anyone tell me how i could calculate these numbers more accurately, without having to play 50 hands like this?
I know i've asked a lot of questions, but can anyone answer any of these questions for me?
You said it backwards or are confused.
Double 11 Vs A at >=+1
And Take Insurance at >=+3
Someone smarter will actually answer your question.
Quote: nickolay411mplan68,
You said it backwards or are confused.
Double 11 Vs A at >=+1
And Take Insurance at >=+3
Someone smarter will actually answer your question.
You're right, i did have it backwards. I guess i just had a lot of numbers rolling around in my head while i was writing this.
Quote: mplan68I know it's not a 9% edge in the same sense that the .28% edge is counted, or in raw winning percent, but it's the edge i care about. It's the financial edge over the casino used to predict approximately how financially successful my combined strategies and methods are. I can know that if i put 1000 dollars on the table in a shoe, i can expect to net about $90 given the history of my betting strategies, cover strategies, and even average number of mistakes i make.
You've experienced variance. You do not have a 9% edge in a counting game.
Quote: RSYou've experienced variance. You do not have a 9% edge in a counting game.
But there is nothing wrong with that. I mean, there is no way to otherwise calculate success overall given that i'm designing my game to be as random as possible. I've got over a year, and $12,000 of trials in my overall strategies, so it will continue to become more well defined until i can ultimately determine precisely how much i can earn. As long as that number is positive, i can change my strategies to continue covering. That's why it's important to me. I could technically call it a variance, but it's not calculable like normal variance. Variance is calculated. I would have to average mistakes in basic strategy, the cover mistakes done in situations, and everything else, and then it would just be an estimate, because it is constantly changing and with more trials it will constantly be getting more accurate to what i could expect in the future. It's probably more accurate to say i am, on average, doing 9% better than the house (in terms of $$ in $$ out), but that translates to me as an edge, not the variance.
Are you saying you have a 9% return on your money so far? or 9% on your action?Quote: mplan68But there is nothing wrong with that. I mean, there is no way to otherwise calculate success overall given that i'm designing my game to be as random as possible. I've got over a year, and $12,000 of trials in my overall strategies, so it will continue to become more well defined until i can ultimately determine precisely how much i can earn. As long as that number is positive, i can change my strategies to continue covering. That's why it's important to me. I could technically call it a variance, but it's not calculable like normal variance. Variance is calculated. I would have to average mistakes in basic strategy, the cover mistakes done in situations, and everything else, and then it would just be an estimate, because it is constantly changing and with more trials it will constantly be getting more accurate to what i could expect in the future. It's probably more accurate to say i am, on average, doing 9% better than the house (in terms of $$ in $$ out), but that translates to me as an edge, not the variance.
Quote: mplan68I can know that if i put 1000 dollars on the table in a shoe, i can expect to net about $90 given the history of my betting strategies, cover strategies, and even average number of mistakes i make.
As this is usually studied, you are evidently confusing Expected Value with House Edge. We need to use a common language here, as you are quickly finding.
Quote: mplan68I know it's not a 9% edge in the same sense that the .28% edge is counted, or in raw winning percent, but it's the edge i care about. It's the financial edge over the casino used to predict approximately how financially successful my combined strategies and methods are. I can know that if i put 1000 dollars on the table in a shoe, i can expect to net about $90 given the history of my betting strategies, cover strategies, and even average number of mistakes i make.
How about 1%? That six deck game is as good as it gets rules wise but we don't know details such as penetration. With mistakes and cover you'll have your hands full trying to reach that 1%. You must not play negative counts in shoe games.
If you expect to win $90 per thousand wagered you're going to be disappointed.
Quote: mplan68But there is nothing wrong with that. I mean, there is no way to otherwise calculate success overall given that i'm designing my game to be as random as possible. I've got over a year, and $12,000 of trials in my overall strategies, so it will continue to become more well defined until i can ultimately determine precisely how much i can earn. As long as that number is positive, i can change my strategies to continue covering. That's why it's important to me. I could technically call it a variance, but it's not calculable like normal variance. Variance is calculated. I would have to average mistakes in basic strategy, the cover mistakes done in situations, and everything else, and then it would just be an estimate, because it is constantly changing and with more trials it will constantly be getting more accurate to what i could expect in the future. It's probably more accurate to say i am, on average, doing 9% better than the house (in terms of $$ in $$ out), but that translates to me as an edge, not the variance.
You can figure out how much you cab earn by simulating your game using software (CVCX if you play like a robot, CVData if you incorporate cover and whatnot).
But absolutely do not confuse previous results with your edge or EV.
Quote: mplan68According to http://blackjack-trainer.net/simulator.php
Too many bugs.
6 decks, double allowed after split, dealer stands on soft 17
Hand: 11 vs A
P1 game #0 bet=$10 cash=$1000
Dealer got 6c Ah [-17] open A
Player got 3s 8d [11]
Player: hit Th [21]
Dealer: hit Ac [-18]
Result: player [21] vs dealer [-18], player won $10
Hand: A8 vs 7
Dealer got Ac 7d [-18] open 7
Player got Ah 9d [-20]
Player: stand
Dealer: stand
Result: player [-20] vs dealer [-18], player won $10
Hand: AA vs A
P1 game #0 bet=$10 cash=$1000
Dealer got 6c Ac [-17] open A
Player got Ah Ad [-12]
Player: split 8s [-19] stand
Dealer: hit 2h [-19]
Result: player [-19] vs dealer [-19], draw
Playing hand #2
Dealer got 6c Ac [-17] open A
Player got Ad Ts [-21]
Result: player [-21] vs dealer [-17], player won $15 w/Blackjack
Quote: mplan68I know it's not a 9% edge in the same sense that the .28% edge is counted, or in raw winning percent, but it's the edge i care about. It's the financial edge over the casino used to predict approximately how financially successful my combined strategies and methods are. I can know that if i put 1000 dollars on the table in a shoe, i can expect to net about $90 given the history of my betting strategies, cover strategies, and even average number of mistakes i make.
That doesn't really make any sense. How are you coming up with these numbers? How many billions of simulations did you run?
Quote: 1BBHow about 1%? That six deck game is as good as it gets rules wise but we don't know details such as penetration. With mistakes and cover you'll have your hands full trying to reach that 1%. You must not play negative counts in shoe games.
If you expect to win $90 per thousand wagered you're going to be disappointed.
For some reason he is calculating it as a percentage of buy-in.
Except, he's actually only calculating his results over a small number of trials.
Once I got a royal flush in video poker. I guess my edge is 80,000%. I should play more; that edge is too good to pass up.
Quote: AxelWolfAre you saying you have a 9% return on your money so far? or 9% on your action?
9% on action. Of the more than $12,000 i have put across the betting circle over more than a year, i have returned 9% above that, which is to say about $13,100.
As for other comments on here, i'm not calculating 9% return over a small number of trials. With only $12,000 at small tables ($2-$5 min), i certainly can't say i've had more than a thousand trials, but a thousand hands is not unsubstantial. After a thousand total trials, 5 or 10 big hands can easily be nullified by 5 or 10 bad hands. Even if i did have a blackjack equivalent of a royal flush, it would have been entirely destroyed by now. Especially if i had assumed that were the norm.
No i'm not confusing house edge or ev with my 9% but i will wholeheartedly refuse to call that 9% anything but an edge, because as discussed before, it is not the variance, and it is what i'm consistently edging the house for. It is simply what i am using to measure the success of my strategies and game overall.
My single biggest night win came in February where i had a TC of what i figured to be +14 (the dealer seemed to be dipping below 1/4 of the deck, a huge mistake, so i took advantage) on the last hand of the deck (single deck, H17, DAS, no RSA, no surrender, not sure about how many split hands allowed or whether doubling was limited to 9-11, but either way it was too hard to resist, and is one of the only none 6-deck games i've played). I was betting $30 a hand (the most i have ever bet on a single hand) and remarkably got 9,9. There were 2 other people at the table, and after they got dealt their hands the TC remained around +6 or +9 including the dealers up 5. Of course, i split the 9's. On the first hand i got a face card and on the second i got an ace. I doubled the soft 10 because the deck was still hot and the dealer had a 5, and i got a 20. The dealer turned over a face card and had to hit and then bust. That was my biggest single hand ($90), and also my single biggest night (about $325 after PROMPTLY cashing out). On the other end of the spectrum, my single worst night was about $350. I'd get good hands, and the dealers would just get better hands. She pulled an 8 on a 13 about 5 times and probably the same on a 15 pulling a 6. This all to say, no, i haven't had any royal flush nights.
As far as penetration, i have no "only" buy in decision, like only buy in at tables with a pen of 75%. I don't want to create patterns. The only rule i have is the common sense one that it's not even worth it at a table that doesn't go at least 2/3. I do play negative counts. I play them in order balance the positive counts, but i try to get out if it looks like it's going to stay there a while. The difference of course is the strategy changes and the bet amounts. I would like to do a little less negative shoes and a little more walking when the deck is hot to avoid suspicion.
All in all, no one has answered any of my original questions. Can anyone tell me about what the odds say about 11 vs A if you take insurance and double at TC +4? That's primarily what i'm here for.
Although I'd guess you've played more than that, as that doesn't take long at all.
Quote: IbeatyouracesBTW, you'll consistently have a 9% edge if you only play with a TC of approximately +19 and higher.
Happens all the time!
Quote: mplan689% on action. Of the more than $12,000 i have put across the betting circle over more than a year, i have returned 9% above that, which is to say about $13,100.
As for other comments on here, i'm not calculating 9% return over a small number of trials. With only $12,000 at small tables ($2-$5 min), i certainly can't say i've had more than a thousand trials, but a thousand hands is not unsubstantial.
Yes it is. The fact that you have won at about 9x your expectation (if you are lucky enough to have an expectation of 1% of your action, which I doubt) should tell you just how unsubstantial it is.
1000 hands would be a significant amount of data if you were flat betting. But, you are not flat betting. You have a very small number of hands with big bets out, and those have a disproportional effect on your outcome.
Quote:No i'm not confusing house edge or ev with my 9% but i will wholeheartedly refuse to call that 9% anything but an edge
In that case, you have no clue what "edge" means. "Edge" is unrelated to past results. "Edge" is simply the expectation of the bet, expressed as a percentage of the initial bet. If you insist on re-defining commonly-used terms, you can't expect people to take you seriously. Speak English, not some random language that you made up where common English words are mapped to your own definitions.
Quote:because as discussed before, it is not the variance, and it is what i'm consistently edging the house for. It is simply what i am using to measure the success of my strategies and game overall.
Nowhere near consistently. 1000 hands is nothing. Are you new to gambling or something? You are currently running well. Good for you. Expecting to continue to win at that rate is foolish.
9% would be a reasonable edge if you could see the hole card every single hand, and could play with minimal cover. If you think that you are getting anywhere near that edge from simple counting, that's just insane.
Quote:My single biggest night win came in February where i had a TC of what i figured to be +14 (the dealer seemed to be dipping below 1/4 of the deck, a huge mistake, so i took advantage) on the last hand of the deck (single deck, H17, DAS, no RSA, no surrender, not sure about how many split hands allowed or whether doubling was limited to 9-11, but either way it was too hard to resist, and is one of the only none 6-deck games i've played).
You don't even know the rules of the game that you are playing and you think that you have a 9% edge? That's funny.
Even with a TC of +14 you do not have a 9% edge, by the way. Do you see how crazy your claim is? If you could play EVERY SINGLE HAND in that situation, your edge would still not be 9%.
Quote:I was betting $30 a hand (the most i have ever bet on a single hand) and remarkably got 9,9.
Is that remarkable?
Quote:There were 2 other people at the table, and after they got dealt their hands the TC remained around +6 or +9 including the dealers up 5. Of course, i split the 9's. On the first hand i got a face card and on the second i got an ace. I doubled the soft 10
1: There is no such thing as a soft 10, unless your other card is a -1. You had a soft 20.
2: Was it +6 or +9? Not only do you not know the rules, you don't know the count? And you think you have a 9% edge?
3: I play almost no single deck but I'm pretty sure that doubling is wrong here. I am not 100% sure, of course (I won't pretend to know all the index plays for "+6 or +9")
Quote:because the deck was still hot and the dealer had a 5, and i got a 20. The dealer turned over a face card and had to hit and then bust. That was my biggest single hand ($90), and also my single biggest night (about $325 after PROMPTLY cashing out). On the other end of the spectrum, my single worst night was about $350. I'd get good hands, and the dealers would just get better hands. She pulled an 8 on a 13 about 5 times and probably the same on a 15 pulling a 6. This all to say, no, i haven't had any royal flush nights.
So you can win or lose $350 in a night, but you think that your edge is 9% because you have won $1100 over $12k of action? 3 losing nights in a row and you are back to even. Then where's your 9%? Don't think it can't happen. It even happens to people who know the rules AND the count.
Quote:I do play negative counts.
HUGE mistake in shoe games. Unavoidable in single or double deck.
Quote:I play them in order balance the positive counts, but i try to get out if it looks like it's going to stay there a while.
How do you know if it looks like it's going to stay a while? Is the dealer kind enough to show you what cards are coming next? If so, your edge should be a lot more than 9%. If not, all you have to go on is that the expected value of the count after the next hand is the same as the count before the hand.
Quote:All in all, no one has answered any of my original questions. Can anyone tell me about what the odds say about 11 vs A if you take insurance and double at TC +4? That's primarily what i'm here for.
What exactly do you want to know about it? Taking insurance and doubling are both correct. Insurance is close (the index is +3) and doubling is not (the index is +1). From an EV standpoint the two bets are unrelated, but it doesn't matter, because you want to do them both.
Once, I took a girl to a local casino. She wanted to play blackjack well, so I taught her basic strategy on the car ride over. She did not count, and she flat-bet the table minimum the entire time. She played close to perfect basic strategy, mostly because when she hesitated I would point her in the right direction. Occasionally she made a mistake without hesitating and giving me a chance to correct it. The house edge in that game was about 0.4%; she was probably playing at about -0.5%. She was up 20 bets after an hour or two of play (she got really lucky -- at one point, she got 3 blackjacks in a row). She played, maybe, 100 hands. Do you think she had a 20% edge? Of course not. She just got lucky. So did you.
The point of this story is, don't fall into the trap of thinking that your past results are a good indication of your actual edge. You have nowhere near enough data. 9% is insane. 1% would be good -- you could get that if you played perfectly and didn't play through those negative counts. MAYBE you could get more at that juicy single-deck game that you're not sure of the rules to. If you are playing-all at 6D I'd be surprised if you are even playing a break-even game.
Quote: IbeatyouracesBTW, you'll consistently have a 9% edge if you only play with a TC of approximately +19 and higher.
I don't think that's true. Doesn't the linearity start to break down at that point?
At some point the large number of pushes (20 v 20) starts to significantly deteriorate your edge -- it actually starts to drop with a higher count. Note that in a deck of all 10s, your edge is 0.
Quote: IbeatyouracesI'll put it this way also. When holecarding, most BP's are only playing with about an 8% edge give or take due to discreet play. 9% consistently while counting? No way Jose.
Yeah, that is the point that I made.
Quote: AxiomOfChoiceI don't think that's true. Doesn't the linearity start to break down at that point?
At some point the large number of pushes (20 v 20) starts to significantly deteriorate your edge -- it actually starts to drop with a higher count. Note that in a deck of all 10s, your edge is 0.
IDK why I even bothered because I know you're right, but boredom and curiosity got me to calculate the odds of getting a blackjack if there were 20 cards, only T through A in a deck. (TC of +52) It's 33.68ish%.
So with a ton more decks (let's say infinite for simplicity) and the same cards, the odds of player getting blackjack are around 1/3, minus the 1/9 of the time you push b/c the dealer gets a blackjack. So about 2/9 times you'd get an extra half unit. So you're edge from that is like 1/9, or like 11.11%. Then AA pair adds some, which I've had a couple too many drinks to bother estimating.
So an infinite deck with a TC of 52 your edge is somewhere in the 11-15% region. Obviously this is all useless information but I was bored and it really just started with me being curious about the odds of getting a blackjack if there were only T though A lol.
(FWIW, I'm not an AP in blackjack generally and I've never put card counting into practice and don't really intend to. But I did start my study in AP w/ blackjack and I've had a few thoughts on the subject I feel nobody has studied. I have a strong feeling that the non linear aspects of deck composition call for some more strategy deviations deeper into shoes with more neutral counts but I have no interest in proving it since I don't ever run into a situation where my theory would be useful. Anyway if anyone is interested in hearing what I had thought about, it had to do with the floating count and the logic behind why it exists, along with the basic strategy deviations. It's totally possible that my thinking was flawed, but I really think it's something that has just been overlooked either because it hasn't been thought of by many or because the math may have been done by someone and shown to barely add profit. Feel free to PM me and I'll get back to you probably within a few days, as I don't check WoV everyday.)
In a deck with all 10s, you have no edge. But in a deck with only 10s and As, in a 4-1 ratio, your edge is massive. Not only do you get a natural 8/25 hands, but insurance has a 140% edge, and you can split all your 20s.
