learningtocount said:These are the only rules that a card counter should except!!! First of on their single deck blackjack games blackjack pays 7 1/2 to 5 instead of 6 to 5!
Second you can double down on any two cards second most important rule!! You are allowed to split aces. The dealer does hit soft 17 giving them a non card counting basic stragedy player only .18% of a disadvantage!!! A card counters advantage can be as high as 1%!!! These rules are usually reserved for the double deck game!!!
learningtocount said:... I don't care how much you spread on 6 to 5 you can not make money on it unless you are the casino or the dealer ...
jimpenn said:Alex...Interesting concept...you may have something here concerning heat. What is your opinion concerning the insurance strategy verses 6D games paying 3/2, Stand or Hit 17, etc. Recently I caught two insursed BJ's in a 6D game in the mid-west and immediately faced early shuffles thereafter.
Have you ever read any professional player's posts that agree with your 6/5 strategy? If so, please forward link.
Best,
-J.P]
P.S It was a boy!
AlexD30 said:How can I over bet the bank when I bet $0 on 75% of the hands and have an average of exactly $200 for the flat bets on +2 or above? Actually the overall average for the 100 hands observed but only 25 played is $133. During the hot shoe (best-half) I have an average EV of about 1.5% with an average bet of $200. During the cold shoe I have EV and an average bet of 0 (Zero).
How can I be wrong?
Sonny said:The optimal bets will be based on the Kelly Criterion. Overbetting by a factor of 2 will give you a 100% risk of ruin despite the fact that you are only playing positive counts. You are suggesting that people overbet their bankroll by a factor of 4 in order to make up for "lost" hands (which technically aren't lost at all). Your system is guaranteed to bankrupt players in the long run.
Although I do agree that you can safely raise your bet sizes when you are Wonging, you need to be very careful not to get carried away. As SSR mentioned, the variance will be huge no matter what the count is, and making bigger bets will only make it worse.
-Sonny-
stainless steel rat said:how can anyone assess your RoR when your bankroll and playing methodology change on a daily basis? One day you are a pure progression player. Next day an advanced level-2 or level-3 counter. Betting 1/12th Kelly makes absolutely no sense to anyone that is serious about advantage play. Playing with such an incredibly small RoR also means you are playing such that it will take forever to double your B/R... Something like investing $250,000 at a .1% annual rate of return... It works, but it isn't the smartest thing to do...
Personally, I don't believe a word of any of this. Way too many inconsistencies and change of tunes...
stainless steel rat said:Somehow I didn't think you would "get it". No big deal. BTW nobody plays with a 0.000% risk of ruin. The only way to do that is bet zero... but since none of the other math or statements are credible, no need for that one to be either...
AlexD30 said:What would be a meaningful measure of success? Winning a few sessions?. We can establish the session to be 20, 100 or 200 hours. I would say this: One week should be sufficient time to prove my play. Now, the suckers will surface. Some would claim that my methods would validate only in one million hands or even more. I would tell them, thats not math, thats cowardice. There are two methods in mathematics and logic: the complete induction and the incomplete deduction. Based on the latter, if a relation proves to be true for 1 case, for 2 cases, . . . for n cases, then it must be true for (n+1) cases. You dont need to play 1,000,000 hands of blackjack. The hands played in one week will suffice to prove that my trend is going up.
AlexD30 said:Well, I have $250,000 BJ bank, no mortgage, no rent to pay, single (divorced), BMW 750IL paid off .. etc.., and am betting 0.08% of the bank that is K/12.5 when I have more than 1% edge. I am looking for an average $200-$250/hour income. I honestly cannot see the ruin. How can I get ruined when I have 2,500 units and am betting K/12.5 with edge and 0 (Zero) with -EV.
And, have over 30 years experience at the tables
AlexD30 said:Well, I have $250,000 BJ bank, no mortgage, no rent to pay, single (divorced), BMW 750IL paid off .. etc.., and am betting 0.08% of the bank that is K/12.5 when I have more than 1% edge. I am looking for an average $200-$250/hour income. I honestly cannot see the ruin. How can I get ruined when I have 2,500 units and am betting K/12.5 with edge and 0 (Zero) with -EV.
And, have over 30 years experience at the tables
NYB said:..your 15 inch dick
Ray said:Alex- As I see it, when you wong in a shoe game, you are just playing at some theoretical advantage with some average bet. If you flat bet $200, that is your unit and average bet. If you play with an advantage and never vary from that you will win money in the long-run. Otherwise the word bias has no meaning. The only real consideration for you is time/hands played and ave bet must be supported by the correct BR.
What you can't do is relate your method to some other strategy that uses a proportional betting method with a spread. Both are valid but different.
AlexD30 said:
Betting optimum means betting Kelly to maximize the expected logarithm or the growth of your bank. But when the bank is big beyond any reasonable chance to ruin than you looking to generate income as safe as possible. Like having ROR = 0.0000000001%
AlexD30 said:Yes, you are right. You got it
Well, my average bet is always a small fraction of the bank. I actually under bet the bank all the time. I use the bank as leverage. The absolute number of units in my bank let me do that. I am looking to make a certain amount per hour, week, month, year and not to double the bank. If I you have a million bank and 1.5% edge you should be looking to make 100K/year while betting K/25 during the 25 hands of the hot shoe. You will be betting $400 per hand for a total of $10,000 action per hour during the hot shoe and expecting to make over the long run $150/hour. In less than part time work like 660 hours you will be up there to 100K.
You dont have to double your bank. What is the point of doubling? You already are at the saturation point in your bank. Doubling it will not give you more protection from the ROR. Youre already supper protected.
Betting optimum means betting Kelly to maximize the expected logarithm or the growth of your bank. But when the bank is big beyond any reasonable chance to ruin than you looking to generate income as safe as possible. Like having ROR = 0.0000000001%
Ray said:Sounds like Alex leverages his BR like income producing mutual funds. I'll have to agree with Sonny. That approach want get it for us church mice (poor, hungry and invisible).
AlexD30 said:The bankroll is so important such that if you plan to play a 16 hours of BJ on a trip and you know that you have an hourly EV of 1.5 average units with a SD of 20 and a DI = 7 but you only bring with you only 60 units of Bank. You also plan to stop playing when the cash runs out. What are the Expected Earnings for the Trip? One might answer that it is 16 hours of play times 1.5 units equals 24 units. However that assumes that you will get to play 16 hours. With your low bank, your chances of tapping out are approximately 36%! The net result of this is that your EV for the trip is only 19 units. Your low bank has cost you about 20% of the EV potential.
In a nutshell, I would say that one has to use his bank as leverage against the ROR to be able to take full advantage of the entire EV potential
stainless steel rat said:That is so incredibly convoluted... Your bankroll is your bankroll. Your session bankroll is your session bankroll. .....
......., please stop just tossing out buzzwords and instead try to say something that makes some sense...
AlexD30 said:
Blackjack strategies are based upon optimizing Expected Value However, in borderline cases standard deviation ought to be looked at.
AlexD30 said:
2. Another one is: BS calls for doubling Ace-2 vs. 5 in <=9 decks. However in 8 decks, the critical fraction to .0005% of the bank. I calculate that more than 200,000 units of bank are required to make this play. Even in 6 decks, the required bank has to be 576 units.</font>
stainless steel rat said:Hey Sonny:
Ask a proctologist, he can tell you where that number came from. :)
Much of what Alex posts appears to be information taken from a variety of blackjack books, then juxtaposed in ways that make absolutely no sense. It looks good on the surface, but makes no sense when looked at with a critical eye...
Any play that would require a 200,000 unit bank would be _some_ high-risk play. Certainly not A2 vs 5. Perhaps doubling a hard 20 with a huge bet out might require that.
As you have seen, you can get sucked into some impossible discussions when you enter "certain threads". :) Not unlike the other web sites we visit, right? :)
Somehow reading this, one would have to conclude that the standard deviation for doubling A2 vs 5 is _far_ greater_ than 2, even though the best that can happen is you win twice your original bet and the worst is that you lose twice your original bet... Just makes you wonder when Rod Serling's name is going to appear and we begin to hear the theme from "The Twilight Zone" coming from the speakers... :)
AlexD30 said:
f = 2k(e1-e2)/(SS1-SS2)
SS1 = V1+e1^2
SS2 = V2+e2^2
for number of units solve 1/f
stainless steel rat said:That is simply wrong. Sonny pointed out why. A2 vs 5 comes up only once in every 1100 hands roughly (You could get A-2 vs 5 or 2-A vs 5, probability of getting those three cards is 1/13 ^ 3 and since there are two ways to get 'em you multiply by 2.) No way making that play needs a 200K+ unit bankroll. The EV is +.134 for hitting, +.126 for doubling.
Next, nobody bets full kelly. No one advises anyone to bet full kelly. The bankroll fluctuation is too wild to deal with. In your case, you are predicting that the player will bet beyond full-kelly to require that kind of bankroll, according to Sonny's post. When you get to kelly * 2, you will eventually bust out anyway. But even a 13.5% RoR is not acceptable for a professional player, so they lower that by betting a fraction of kelly.
If I could "wong in" every time I would be dealt a hand of A2 vs 5, I'd play non-stop and double every time. +.13 * 2 units is a nice return on an investment for less than a minute's time... If you want, I can run a sim on A2 vs 5 for a couple of billion rounds to show how far one's bankroll could swing in either direction, which would show exactly how big your bankroll has to be (in terms of units) to play that one hand over and over.
In fact, I modified my code and ran this sim for 100M rounds on my slow laptop. The result was:
total hands = 100,000,000 win = 50900122 push = 4490697 lose = 44609181
Which gives an ev of +.124, based on infinite deck. More rounds would likely bring the EV down slightly to the Wizard's number of +.123. I see nothing alarming in the above numbers that would suggest an impossibly large SD to make this play too risky unless one owns half of Fort Knox. You are simply mis-using Cac's post. If you think A2 vs 5 is "too risky" then by all means never double. I am always going to double the hand because every bet I make is going to be less than full-kelly-based...
NYB said:Alex,
Just to clarify; You would rather save the money by not doubling a hand you have an clear advantage on, and "invest" it later when you have a smaller advantage?Even though you have a 2,000 unit BR?
Is that the gist of your latest post?
AlexD30 said:The decision to put out more money should simply be based on upon the same criteria that you made the original bet. Suppose you had a true count that gave you an advantage of close to 6%, and you betting about 75% of your advantage, have a bet of about .045 of your bank on the table. You get a hand that calls for a double in your playing index. However, simply hitting also has an advantage. The question is, how much addititional advantage must you gain from the double in order to justify putting out an additional 4.5% of your bank? You certainly wouldn't want to do it for a gain of .005 difference over hitting or standing.
When only a small capital is wagered, then the difference between two strategies is negligible. However Kelly breaks down when larger amounts are bet. For example, a Kelly player with a big bet out should insure "good hands" even if the expectation on the Insurance Bet were negative. (There is an excellent discussion of this point in the 1986 edition of Griffin's "Theory of BJ"). Similarly, Kelly strategy would dictate passing up many double down bets when the gain from doubling did not justify the increased risk.
stainless steel rat said:What idiot is going to bet 4.5% of their bankroll on a single hand? The RoR is 100%. none of this makes any sense...
AlexD30 said:Of course, Rat, you are again completely in error. What you are saying is that based upon your studies of your "kitchen-table rules and game". At the same time you **think** that I am wrong while you are correct on card counting and beating your kitchen table setup game. Well, still have to bet Kelly. If you get to have a bet 4.5% of your bank when you have 6% edge youre going to be at 13.5% ROR.
You either play Kelly or have no business in BJ. If you want less risk then double your bankroll and have K/2 K/3 or whatever you are comfortable with. Get more units or play at half units of what you have now. You definitely got to bet proportionally with your edge/variance if you want to call yourself a professional. If not, then play slot machines or Keno instead.
It looks like you're new here. If you want to get involved, click one of these buttons!