Multiple Hands and adjusting strategy...and Probablity vs Odds

Hi, total newb here but I was searching the ask the wizard archives and couldn't get a straight answer.

There are two parts to this topic, two unrelated questions.

1)

the Wizard states that playing multiple hands poses no advantage to the player, yet I've heard this is not true for card counters. If you were playing one hand and counting cards, naturally, you would open up as many hands as possible if the count were to jump incredibly in your favor? This would greatly increase your chances of winning as well as obtaining blackjack on one of said hands netting you the extra 50%.

Now if I am mistaken regarding the above, I was wondering if it is correct to alter your strategy for 2nd hand based on one hands holdings. To give you an example, lets say hand number 1 you've hit to 21. On hand number 2 you have 16. Dealer's up card is 7-A. Is the logic that you have one very likely winning hand enough to stay on the 16 and hope that the dealer busts? I've seen this on a number of occasions as a dealer. The player thinks that one hand is solid, and doesn't risk killing the other hand, otherwise, its a definite push or worst case, push/lose.

2)

I am very confused with probability vs odds. This pertains to the martingale strategy and the logic behind it.
I understand that the odds of flipping a coin and resulting in heads is 1:2. This never changes. But I also know that if you consider a set of two flips, the probability of obtaining heads twice is 1:4. However, the odds are still 1:2 that it will be heads on either occasion.

If we consider a session of blackjack to be 100 hands, assuming it was a 50/50 game, and I've lost 99 hands in a row, how the hell can I not bet the farm on the next hand with the notion that the probability of me losing 100 hands in a row is so ridiculous, that it is highly probable I will win on the next bet?

Otherwise if I didn't win, I'd just label myself as the unluckiest player in the world and quit playing BJ forever.

Its like those players on Bacc who always bet against the run. Eventually, they win.

Thanks very much for your assistance!

Quote: waznboi03

I understand that the odds of flipping a coin and resulting in heads is 1:2. This never changes. But I also know that if you consider a set of two flips, the probability of obtaining heads twice is 1:4. However, the odds are still 1:2 that it will be heads on either occasion.

If we consider a session of blackjack to be 100 hands, assuming it was a 50/50 game, and I've lost 99 hands in a row, how the hell can I not bet the farm on the next hand with the notion that the probability of me losing 100 hands in a row is so ridiculous, that it is highly probable I will win on the next bet?

Read your second paragraph and then read your first again and you will have answered your own question.

Quote: waznboi03

I understand that the odds of flipping a coin and resulting in heads is 1:2. This never changes. But I also know that if you consider a set of two flips, the probability of obtaining heads twice is 1:4. However, the odds are still 1:2 that it will be heads on either occasion.

If we consider a session of blackjack to be 100 hands, assuming it was a 50/50 game, and I've lost 99 hands in a row, how the hell can I not bet the farm on the next hand with the notion that the probability of me losing 100 hands in a row is so ridiculous, that it is highly probable I will win on the next bet?!

Cool post. I'll try your 'fun with coin flips' thing.

My first observation would be that colons -- these guys, ":" -- cause trouble in probability analysis. The odds on flipping a coin and getting heads is "one in two", but the colon odds would be "one to one," normally written like this: 1:1.

Likewise, for us to flip a coin twice and have it come up heads both times (or to win a 2 gm sports parlay) is "one in four," but the odds against it happening are only "three to one," or 3:1 (unless I'm the bettor, then it's completely impossible to win).

To see that's true, consider the 4 possibilities separately: H/H (our winner), H/T, T/H, and T/T. If we bet $100 on each parlay, we'd need to make profits of $300 on our H/H winner to offset the three $100 losers for the other possibilities. To insure they turn a profit, most sportsbooks only pay out a 2 gm parlay at 2.6 to 1.

Now the thing about your 99 straight losing hands of blackjack hypothetical is that the weird, odds defying stuff has ALREADY happened. The odds against a player (playing perfectly) losing 99 straight hands are astronomical, but whatever those odds are, they've already happened. True, lightning's struck that casino, but it's been there and gone by the time our crying, broke player starts his 100th hand.

Even though it doesn't feel right, his odds on winning his 100th hand are no better than his odds were on winning his first hand this morning, back before he had to take out that second mortgage.

deleted

Yea i see what you're saying about the correct way of stating the odds, etc.

In regards to this: "Even though it doesn't feel right, his odds on winning his 100th hand are no better than his odds were on winning his first hand this morning, back before he had to take out that second mortgage."

I understand the actual odds are no better than the first hand, but my confusion lies with the term "probability/probable". Isn't probability different than odds?
At the point of losing 99 hands, can't I state with validity that it is highly unprobable to lose the 100th hand considering ive already lost 99 in a row.

OR does probability directly correlate with odds, as in, its practically equal probable that I will win or lose that 100th hand...

Quote: waznboi03

Yea i see what you're saying about the correct way of stating the odds, etc.

In regards to this: "Even though it doesn't feel right, his odds on winning his 100th hand are no better than his odds were on winning his first hand this morning, back before he had to take out that second mortgage."

I understand the actual odds are no better than the first hand, but my confusion lies with the term "probability/probable". Isn't probability different than odds?
At the point of losing 99 hands, can't I state with validity that it is highly unprobable to lose the 100th hand considering ive already lost 99 in a row.

OR does probability directly correlate with odds, as in, its practically equal probable that I will win or lose that 100th hand...

Try it, action speaks louder than words.....

Quote: hmmm23

Now the thing about your 99 straight losing hands of blackjack hypothetical is that the weird, odds defying stuff has ALREADY happened. The odds against a player (playing perfectly) losing 99 straight hands are astronomical, but whatever those odds are, they've already happened. True, lightning's struck that casino, but it's been there and gone by the time our crying, broke player starts his 100th hand.

Even though it doesn't feel right, his odds on winning his 100th hand are no better than his odds were on winning his first hand this morning, back before he had to take out that second mortgage.

Now, hypothetically, to keep losing straight all the way IS NOT Blackjack. It can either be bac or Roulette BUT NOT BJ. Even the world unluckiest person would get a BJ once a while or the Dealer could bust every other hand. To even consider the Dealer "Unburstable" is way outa line here.

I bet my last dollar.

Quote: waznboi03

Hi, total newb here but I was searching the ask the wizard archives and couldn't get a straight answer.

There are two parts to this topic, two unrelated questions.

1)

the Wizard states that playing multiple hands poses no advantage to the player, yet I've heard this is not true for card counters. If you were playing one hand and counting cards, naturally, you would open up as many hands as possible if the count were to jump incredibly in your favor? This would greatly increase your chances of winning as well as obtaining blackjack on one of said hands netting you the extra 50%.

Now if I am mistaken regarding the above, I was wondering if it is correct to alter your strategy for 2nd hand based on one hands holdings. To give you an example, lets say hand number 1 you've hit to 21. On hand number 2 you have 16. Dealer's up card is 7-A. Is the logic that you have one very likely winning hand enough to stay on the 16 and hope that the dealer busts? I've seen this on a number of occasions as a dealer. The player thinks that one hand is solid, and doesn't risk killing the other hand, otherwise, its a definite push or worst case, push/lose.

2)

I am very confused with probability vs odds. This pertains to the martingale strategy and the logic behind it.
I understand that the odds of flipping a coin and resulting in heads is 1:2.

Your Paragraph (1)
Your experience with BJ probably gave you a clear prospective of the way it should be played. You are new and fresh, unclouded by "ways & methods" Basically, all methods are worthless. The little advantage is to play Basic S...which is the most logical and to player's small advantage.

I would say your perspective of multiple hands BJ is workable and profitable. You mentioned playing only 2 hands which is "controllable and conservative" small BR. We have even played up to 6 or 7 hands, coming out better for it. But it should not be played on hours this way. It could burst your BR pretty fast too. 2 hands is good considering you just want to make some money and not bust the place. You'll make when you win the few hands of doubling and splitting. wait for such chances. If you cannot get through those few chances, leave the table. Some good hands will cover the few bad. Hedging..

Your Paragraph (2)
Negative martingale...No..you cannot make much and perhaps a marginal sum or just break even. Not worth the risk and long run. Just increase the bets by 50% if the going is good "for a few hands only". Most of the time....Bet flat....you will come out tops.

Conclusion...Play BS.....2 hands....increase a little and NEVER stay too long....

Any way you look at it, gambling is still gambling.....50% luck, 30% brains and 20% chances.

Good luck!!!

Quote: Casinoraider

Your Paragraph (1)
Your experience with BJ probably gave you a clear prospective of the way it should be played. You are new and fresh, unclouded by "ways & methods" Basically, all methods are worthless. The little advantage is to play Basic S...which is the most logical and to player's small advantage.

I would say your perspective of multiple hands BJ is workable and profitable. You mentioned playing only 2 hands which is "controllable and conservative" small BR. We have even played up to 6 or 7 hands, coming out better for it. But it should not be played on hours this way. It could burst your BR pretty fast too. 2 hands is good considering you just want to make some money and not bust the place. You'll make when you win the few hands of doubling and splitting. wait for such chances. If you cannot get through those few chances, leave the table. Some good hands will cover the few bad. Hedging..

Your Paragraph (2)
Negative martingale...No..you cannot make much and perhaps a marginal sum or just break even. Not worth the risk and long run. Just increase the bets by 50% if the going is good "for a few hands only". Most of the time....Bet flat....you will come out tops.

Conclusion...Play BS.....2 hands....increase a little and NEVER stay too long....

Any way you look at it, gambling is still gambling.....50% luck, 30% brains and 20% chances.

Good luck!!!

I'll say this as nicely as I can and I hope you don't take it personally. Your posts contain bad advice, superstition and myth. Please stop. New members seeking guidance should not be subjected to this.

Quote: Casinoraider

Now, hypothetically, to keep losing straight all the way IS NOT Blackjack. It can either be bac or Roulette BUT NOT BJ. Even the world unluckiest person would get a BJ once a while or the Dealer could bust every other hand. To even consider the Dealer "Unburstable" is way outa line here.

I bet my last dollar.

Heh, cool. I mean, our broke friend's chances of winning his next hand are also no LESS than they were for his first hand (about 42.5% including pushes, 46.4% without), so if it's a wager you'd generally make on BJ, it's perfectly justified. But me, I'd be nervous about making a bigger than normal bet.

This is just a random thought, but would it matter to you if you'd been present to watch him lose every hand vs. whether you just walked into the casino after hand 99 to investigate the sobbing? (let's assume here he's a perfect player and that's verified for sure). What about if you were just informed about a player playing perfectly just having lost 99 hands in a row by someone calling you on the phone from the casino (let's assume he can't be lying and you know that), would you still bet your last buck on that guy winning his 100th hand? If any of those things would affect your bet size, why would it?

From the fun facts file: With math I could never even hope to follow, I once saw a mathematician calculate that based on the amount of blackjack dealt in casinos worldwide, the average amount of times in the last 50 years that a dealer's gotten a hand of 17 by dealing himself 17 straight Aces is about 9.5. Can you imagine being at one of those tables? I think I'd assume the game was rigged, even though it's really just the law of large numbers in action.

(Still, though, I'd def leave that casino)

This negative progression martingale system is flawed. What is different about 99 and 100? Nothing, when you consider that statistics apply to numbers much much greater than 99 or 100. With this system, if you doubled your bet every time you lost you would be wagering "the farm" on a net 1 unit win. I guarentee you if 9.5 dealers have drawn a hard 17 with 17 aces, that more than 9.5 gamblers have lost their farms on 50/50 bets that were "bound" to win. People run simulations because a sample size of 100 is too small a population simply because yes that 100th time is just as likely or unlikely as the 1st or the first 99. Personally I have lost 17 bets in a row using the martingale system. (Before I was too broke to continue. Excruciating. (Lucky I didn't own the farm at that point).

Quote: waznboi03

I understand the actual odds are no better than the first hand, but my confusion lies with the term "probability/probable". Isn't probability different than odds?
At the point of losing 99 hands, can't I state with validity that it is highly unprobable to lose the 100th hand considering ive already lost 99 in a row.

OR does probability directly correlate with odds, as in, its practically equal probable that I will win or lose that 100th hand...

The Probability vs Odds Thing

Your last sentence is completely correct. The terminology's confusing, but don't let it throw you. Probability and Odds are the exact same thing. Odds are just a numeric expression of probability. Once you get the numbers for one, you already have the other, you just might not know it yet.

Plus, I can help with a trick. For any 50/50 type event, whether it's flipping coins or picking NFL winners vs the spread or whatever else, there's an easy way to work out both probabilities and odds. Just follow this recipe:

1. Start with the number 1.
2. Double it however many times the 50/50 event will happen (how many coin flips? or how many games in your parlay?)
3. That's it, you're done.

So, if we're asked "What's the probability (or odds) that a coin will come up Heads four straight times?" Just crank it into the machine. Start with the number 1 and double it 4 times:

1. The number 1 doubled once = 2
2. Doubled a second time = 4
3. Doubled a third time = 8
4. Doubled a fourth time = 16

That last number you get, here it was 16, gives you the probability. So, there's a 1 in 16 chance that a coin flipped 4 times will come up Heads all four times.

And here's the cool part of the trick: to turn that into Odds, all you have to do is subtract 1 from the probability number. Therefore, the Odds are 15 to 1 against a coin coming up Heads 4 times.

If we'd been asked above what the probability of flipping 3 straight Heads was, then we would've stopped after step 3 and reported there's a 1 in 8 chance of flipping 3 straight Heads and that therefore, the odds against doing that are 7 to 1.

Not too bad, y?

Quote: 1BB

I'll say this as nicely as I can and I hope you don't take it personally. Your posts contain bad advice, superstition and myth. Please stop. New members seeking guidance should not be subjected to this.

I think you may be too engrossed with maths on blackboards to face reality.....I have done some reading, and checked some facts.....please look at the "Spread Bettings" forums from some experienced players and in the past few days....then tell me...was I all that wrong???

Anyway, no hard feelings and no offence, never did liked your postings of selfish views and mis-guided opinions. Perhaps, you are working with the Casinos?

deleted

Quote: Casinoraider

I think you may be too engrossed with maths on blackboards to face reality

Sooner or later, the reality matches the math.

Quote: waznboi03

I am very confused with probability vs odds.

Although I tend to prefer probabilities stated as percentages (1 in 16 = 6.25%), hmmm23's explanation above holds water. The reason for my preference is that it is easier to ensure that my probabilities all fall within 0 and 1 (or 0% and 100%).

Quote: waznboi03

If we consider a session of blackjack to be 100 hands, assuming it was a 50/50 game, and I've lost 99 hands in a row, how the hell can I not bet the farm on the next hand with the notion that the probability of me losing 100 hands in a row is so ridiculous, that it is highly probable I will win on the next bet?

Sure you could bet the farm. But if you're going to do that, you might as well do it on the first hand, before those 99 losses - at least that way you would win more if you win the hand!

The answer to your question involves conditional probabilities or conditional odds, whichever format makes more sense to you (as hmmm23 indicated, you can calculate one from the other - there is a 1 to 1 relationship between probability and odds). So, while the probability of losing 100 hands in a row is nearly zero (and the odds against nearly 1:1), the conditional probability of losing the 100th hand given that you have lost the first 99 hands is the same as losing a single hand. This assumes no card counting or playing the 100th hand immediately after a shuffle. And it makes no difference whether you were there to witness the 99 losers or not, except with respect to your believing that it has happened!

Unless you have other information (such as having a high TC), you would not want to bet any more on the 100th as on the first 99.

Another way to look at this - why not bet the farm on the 50th or 90th hands? Why is the 100th hand so much more important than any other before or after it? The answer is it is not - it is just another hand.

Quote: camapl

Another way to look at this - why not bet the farm on the 50th or 90th hands? Why is the 100th hand so much more important than any other before or after it? The answer is it is not - it is just another hand.

Yes, this is the thought that should scare you away from this line of thought. By the 99th hand, you've had a lot of "sure things" go bad.

I had a friend who was generally pretty sharp, but was been bitten by the martingale bug back when partypoker was still available in the US. He would win a dollar at a time, but double his bets on a loss to get even.

I waited for the satisfaction of seeing a bad run, but instead, he a had a good run and saw that the opposite could've happened, so he quit.

So u get to hand 99 and decide to bet the farm everything you have. The dealer deals you a nine and a two you look up and see the dealers up card is a six...
now what do you do? or you get Aces or eighths vs some up card that by not splitting gives you a greater chance to lose.

Quote: AxelWolf

So u get to hand 99 and decide to bet the farm everything you have. The dealer deals you a nine and a two you look up and see the dealers up card is a six...
now what do you do? or you get Aces or eighths vs some up card that by not splitting gives you a greater chance to lose.

In the case of 11 vs. 6 and you're broke, you can still take a hit and have the same chances of winning.

In the case of getting a pair of aces or eights when you're broke, that's a case where being unable to split does hurt your chances of winning.

As a side-note, a case where this may be of some practical value is if you're playing a blackjack tournament when you need to bet everything or almost everything to catch up at the end. If you're deciding whether to bet half or almost half (and leave some room to raise your bet with a double or split) or to go ahead and bet the farm or almost all of it, the latter is usually the better choice. The only time that strategy hurts you is if you get a pair like aces or eights which means you're playing with a weaker hand than you would if you could split (getting a good double hand isn't a bad thing though since you can still take a hit and get the same results as if you bet only half as much initially and then doubled). By contrast, there are more (and more frequent) hands which are good alone but are severely hurt if forced to double and take another card (e.g. 10-8 and 10-9 which have very few cards that won't bust the hand if drawn to, and smaller hands which leave you unable to take another card).

There is a world of difference between:
I will lose 99 in a row and then win 1
And
I have lost 99 in a row and will win the next 1.

About 49.9999%.

Quote: waznboi03

I understand the actual odds are no better than the first hand, but my confusion lies with the term "probability/probable". Isn't probability different than odds?
At the point of losing 99 hands, can't I state with validity that it is highly unprobable to lose the 100th hand considering ive already lost 99 in a row.

OR does probability directly correlate with odds, as in, its practically equal probable that I will win or lose that 100th hand...

Probability and odds are two different ways of stating the same thing. That's not the issue.

The issue is whether you're dealing with independent events. If I'm dealing cards from a shuffled deck and I get through the first 50 without seeing the ace of spades, I know my chances of seeing the ace is now 1 in 2: there are two cards left and one of them is the ace. That's an example of a situation where events are not independent. If the ace had already been dealt, the probability of seeing it later in the deck is zero.

However, coin flips and dice tosses are not such events -- there is no causal relationship between the odds of one flip of a fair coin and the next one. It doesn't matter whether the coin is freshly-minted or has been flipped 99 times (all heads).

People are generally terrible at comprehending randomness. The trick is to learn to separate your intuition (which is often wrong) about what happened in the past with a proper understanding of what is likely to happen in the future. In short, you need to (a) learn to suppress your natural instincts about looking at the past in all cases to help comprehend the future, and (b) learn to distinguish the cases where the past matters vs. where it doesn't. That's a tall task for most people -- even people in the casino industry. I've worked with plenty of casino technologists who nevertheless misunderstand gaming mathematics.

For coin flips, what happened in the past doesn't matter. Same for dice tosses or roulette spins. For dealing down a deck of cards, it does. That's why card counting works in blackjack, but there is no equivalent "dice counting" technique for craps. It's also why -- despite the centuries of roulette system hawkers -- there's no way to divine an exploitable pattern of past outcomes on a fair roulette wheel.

It turns out that the changing composition of a depleting shoe of cards allows card counting to be effective in blackjack but not in baccarat. Moreover, in neither game can you discern an exploitable pattern from the past sequence of wins and losses. The best you can do -- and this is only in blackjack -- is understand that the next hand is slightly more or less likely to win (and bet more or less accordingly). It's nothing nearly as cut-and-dried as "I just lost 99 hands, so I'm due to win the next one" or "I just saw 99 banker wins in a row, player is due next." Casinos are happy to give you scorecards to track past roulette or baccarat outcomes because doing so can't help you beat the game.

Does this mean a casino will not give you a card to keep the count in blackjack ? That hardly seems fair !

Quote: MathExtremist

The issue is whether you're dealing with independent events.

First, should you, respected reader, find anything of worth in this post, I humbly request that you remember it was Hmmm23 who turned you on to this and should you be fortunate enough this Fall to find a sportsbook still accepting this type of wager (or even its diluted weaker sister, IF bets), then I ask only that you alert my humble self to its existence. Thanks very much.

Correlation has a real world application that's been very profitable for sports gamblers. Until recently, some sportsbooks allowed early season college football parlays with a team and an OV/UN total from the same game.

In the early weeks of the season, when teams like Texas play teams like Southwestern Upstairs Barber College Night School, they'll be 45 pt favorites, with an OV/UN of maybe 48. To see correlation in action, look at the game from the favorite's point of view.

If Texas covers it's - 45 pt spread, then this game's also extremely likely to go OV 48. Since a 2 gm parlay (usually) pays out 2.6:1, your $100 winner on Texas - 45 suddenly becomes a $260 winner, provided only that the final score's not 46 or 47 to 0 (although that happens more often than a bright young gambler expected it would).

Slightly less reliable but still in profit is taking the reverse: Barber College + 45 and UN 48. Since scissors aren't allowed on the field, the aspiring barbers aren't likely to score much against Texas's defense, so if they do manage to cover + 45, the total's also very likely to be UN 48. Again, a $100 bet becomes a $260 winner.

Alas, as far as I know, this opportunity is no longer available. But if it is available somewhere this Fall, I hope you'll remember it was your old pal Hmmm23 who turned you on to the idea and send his humble self a PM to that effect.

Quote: Buzzard

Does this mean a casino will not give you a card to keep the count in blackjack ? That hardly seems fair !

lol (I actually did laugh out loud - thanks!)

@math extremist:
Now I'm totally screwed up. You said : However, coin flips and dice tosses are not such events -- there is no causalrelationship betweenthe odds of one flip of a fair coin and the next one. It doesn't matterwhetherthe coinisfreshly-mintedor has been flipped 99 times (all heads).

Well if thats the case, what do you call the notion of flipping 10 heads in a row? and even though the flip itself is 50 percent heads, how can the likely hood of heads occuring x times in a row not add some sort of meaning and value towards the choice of the next flip as x approaches infinity?

This is where I am struggling. Because I know that if its extremely unlikely it'll happen, it probably wont. And this principle should apply to blackjack, no?

deleted