dhanson perhaps you know the answer. I have heard that, in Vegas, card counting is not illegal (although casinos, as “private” clubs, can bar you), but that the use of any mechanical device to aid gambling is illegal. Further, I have heard the “mechanical device” includes pen and paper (for writing down the # of face cards, etc., that have been played). Do you know anything about this? I’ve learned to hide my counting by constantly “nervously” playing with my chips, and using the chips to keep count, but I’ve only played in East Coast and Caribbean casinos.
Yes, using a ‘device’ is a felony in most gambling jurisdictions. “Device” is rather loosely defined, but typically refers to an electronic or timing device of some kind. Paper and pencil have never been subject to device laws - in fact, if you play baccarat the house will routinely hand you a pad of paper and a pencil for tracking wins and losses (which, of course is useless - they do it to give the player the feeling that there is strategy involved, when really baccarat is just a coin flip).
Anyway, don’t worry about pen and paper, or using your chips to count. Just don’t head into your local casino and pull out your Palm Pilot at the blackjack table.
Not only does the dealer have a 5/13’s chance of having a winner from the first two cards, if he doesn’t have a winner, he will have the opportunity to hit in order to get to 17 or above and beat your sixteen. You, on the other hand, have 0% chance of getting seventeen or above if you stand.
I rely on the authors of blackjack books who swear they’ve run the odds through computers, but it makes common sense as well. You have a 5/13 chance of improving your hand. The dealer with the 7 has a 5/13 chance of already beating your hand. Your odds are even right there, and then the dealer has the opportunity to hit if he doesn’t have the winner, thus increasing his odds.
It’s been 9 years since my one and only stats course, but I think I’m right here. Someone correct me if I’m wrong.
Sua
And this is what makes most counting strategies less than effective.
Even on single- and double-deck blackjack tables, they chop so many cards off and reshuffle when they reach them, the strategic advantage is essentially eliminated.
MadHatter, trust the 16 thing. It’s right.
The maneuver I have far more difficulty with than that is staying on 12 if the dealer has a bust card showing. Goes against every fiber of my being.
You are going to have to excuse my ignorance… and I appreciate the time you are taking to explain these things to me, but… in your above statement, what you are saying is:
“Chances are, you will not improve your hand if you hit, and chances are the dealer does not beat your hand anyway.”
If you don’t hit, the only question is whether the dealer busts or not… because if you hold 16, and he doesn’t bust, you lose (he has to go over 17).
For the record, I’m assuming a basically infinite deck here, and dropping out several terms which are basically insignificant (and thus only using two sigfigs).
The dealer’s chance of busting while he has a 7 showing is about 0.23
If you don’t hit, you lose 77 out of 100 times.
If you do hit…
Your chance of busting if you hit on 16 is 0.68
You bust 68 times out of 100. The dealer beats your final hand 6 times out of 100. (Please trust me; I’m not going to attempt to show the math. Again, I’m using an infinite deck and dropping out terms which require multiplying by 1/13 more than twice (And thus would be in the third sigfig).)
68 + 6 = 74
You lose 74 times out of 100.
Let’s recap:
If you hit, you lose 74 times out of 100.
If you don’t hit, you lose 77 times out of 100.
You’re pretty well screwed in this situation anyway… but you’re less screwed if you hit, by three hundredths.
One note on my above calculations: They do not take counting into account (you can’t count on an infinite deck anyway). If the count is high (many high cards in the deck), the count is stacked even better in your favor. If the count is low, these numbers will creep toward one another. With a very negative count, you may well be better off not hitting… but you’ll also be better off not betting at all.
Standing on a 16 against a dealer’s 10 will cost you an extra 3% of your money. Standing on a 16 against a dealer’s 7 is MUCH worse. I’ll leave it as an exercise for you to figure out why.
The 16 vs 10 play is actually very close. So close that if you are counting the correct play is to stand as soon as you see one extra low card come out.
There are a lot of basic strategy plays that may seem counter-intuitive. A lot of people have trouble with the notion of splitting 8’s against a face card, hitting a soft 18 vs a ten, etc. And some of these plays are pretty hard to analyze without the benefit of computer simulation.
But trust us on this - there is nothing unknown about basic strategy. We know the mathematical expectation of EVERY playing option, and have since the 1960’s. There’s simply nothing to debate. Deviate from it at your peril.
If you want to learn more about blackjack, I suggest the following references:
Professional Blackjack by Stanford Wong Blackbelt in Blackjack by Arnold Snyder The World’s Greatest Blackjack Book by Humble and Cooper Blackjack for Blood by Bryce Carlson Blackjack Attack by Don Schlesinger
For a detailed look at what it takes to count cards for big money in a casino:
Turning the Tables on Las Vegas, and Burning the Table in Las Vegas both by Ian Anderson
A more mathematical treatise on blackjack and gambling in general is The Theory of Blackjack by Peter Griffin
A general gambling theory book (graduate university level): The Theory of Gambling and Statistical Logic by Richard Epstein
These books together should constitute the Canon for a serious blackjack player. At the very least, you should own Wong’s book, Griffin’s book, and Schlesinger’s book. The money spent on them is trivial compared to what you’ll lose if you don’t read them and take their lessons to heart.
The chart is a variation of one I found in a 35+ year old book. Using it alone, it brings your odds against the house almost to 1:1.
The author of the book I mentioned introduced card counting, of which he had several stories about how well it worked. So well, that he was barred from many casinos. (He won thousands of dollars).
His counting strategy was all-too-obvious: when he calculated that the remaining deck was bad for him, he’d bet $1; if it was good, he’d bet the table max. It also worked best for single-deck, multiple-round tables. Multiple decks (like the 8 used in Atlantic City) dilute the counting scheme. And of course, one game per shoe completely negates counting.
I don’t have a generic BJ game, but I did write a simple DOS game that does my betting for me, plus has several modes for me to use (manual, automatic, cheating , and automatic/cheating). For the cheating mode, it either: [ul][li]shows me the next 15 cards (allowing me to see that I should double on my K 5 because the next card is a 6, something one should never do based on the tables)[/li][li](in auto/cheat mode) it first sees what the tables suggest. If it’s a ‘hit’, it then sees if that would put you over 21. It changes it to a ‘stand’ if you’d bust, even if you’ll lose anyways.[/li][/ul]
From repeated empirical runs with auto/cheat, I see that if only you could see just the next card, you’d clean up. (Now where are my X-Ray specs?)
MadHatter: Geez, are you going to make us go through every bloody decision on the table and prove it to you? If you won’t take our word for it, why don’t you do the math yourself?
In this specific case (12 vs 4,5,6), the dealer is going to bust roughly 43% of the time, if memory serves. If you hit your 12, you will bust if you hit any face card, and you’ll still make a stiff if you hit a 2,3,4. That only leaves you a 5,6,7,8, or 9 as possible winning cards. That’s 20 out of 49, and the dealer can still outdraw you if he doesn’t bust.
Quit trying to pick holes in the basic strategy: You won’t be able to. As I said, this stuff has been dissected and PROVEN decades ago. You’re wasting your and everyone else’s time. If I had my references here, I could give you the absolute mathematical expectation down to 1/100th of a percentage point of all options. There’s simply nothing to debate.
Settle down would you Sam? I am not wasting ANYBODY’s time here. If you don’t want to post to this thread… then don’t. No one is FORCING you to come in here and WASTE your time by posting. I am just asking for some clarifications. NOWHERE have I said the table is flawed, in some points, however, I have asked the SD as to exactly WHY the odds are more favorable for ‘x’ then they are for ‘y’. Please don’t flame me for wasting your time. If you feel your time is being wasted… there is a simple solution. Don’t post.
A very simple, but effective, counting technique to augment your chances is “counting fives”. Simply keep count of how many 5’s come out,and when more than half are out, increase your bets.
This is based on the fact that a five card is the dealer’s best friend, turning his/her “must-hit” 16 into an unbeatable 21.
You still have to use good basic strategy, though.
And,dlhanson, the best bet in the house is the back-up “odds” bet on the craps table. They are paid at true-odds, and are removeable after every roll.
Sam, MadHatter, hitting on a 16 is the “Monty Hall Question” of blackjack. It’s a very frustrating issue for newcomers to the game, because hitting on a 16 is counterintuitive to the point of causing nausea. It’s a very frustrating issue for experienced players because (a) we’ve explained it a dozen times to other people already, and (b)we still get a little nauseous hitting on the 16, so we rely religiously on the statisticians and get defensive when someone challenges us.
Is there anywhere I can find a table showing the odds of winning and losing on each possible hand shown in the BASIC STRATEGY TABLE?
What action shoukd I take if I hold 3 cards that add up to 17, but one of the cards I hold is an ace? Should I stand on ‘17’ or hit my ‘7’ ? Likewise for 18 - 19?
Not sure if anybody wants it, but a long time ago, I wrote a simple C++ program that would compute what the basic strategy would be for a given set of rules and number of decks.
The way it works is very simple:
When you stand - calculate all the permutations of cards that the dealer can get when he shows his hole card and then all the permutations of his draws. Calculating the percentage win is very easy.
When you draw, repeat the permutation calculation above.
Now it’s really simple to compare the winning percentage of the two options above and take the one that gives you the best odds. A side benefit is that the program can give you the house odds for a given set of rules.
Suffice it to say that the basic strategy posted is right. The only thing you have to do is make sure that you are using the correct strategy for your set of rules and number of decks.
By the way, if you are going to have anything nailed down pat, make sure that you know when to double and split - those are the actions that give you the big money. Screwing up on some of the esoteric stand/hit’s won’t cost you nearly as much as if you double when you shouldn’t or hit when you should double.
Mad: “soft” refers to the ace. An ace can be either 1 or 11, whichever it to your advantage. So if you have an Ace and a six, you have wither 7 or 17. It’s a great situation to be in, because you can either stand if the higher value is good enough, or hit without the risk of busting if it’s not.
, and automatic/cheating). For the cheating mode, it either: [ul][li]shows me the next 15 cards (allowing me to see that I should double on my K 5 because the next card is a 6, something one should never do based on the tables)[/li][li](in auto/cheat mode) it first sees what the tables suggest. If it’s a ‘hit’, it then sees if that would put you over 21. It changes it to a ‘stand’ if you’d bust, even if you’ll lose anyways.[/li][/ul]