Unwinnable Hands
What I'd be fascinated to see is the results of only winnable hands, with those unwinnable removed from the equation. I'd love to see the odds of Winning/Losing for each of the hard hands where a decision will actually affect the outcome.
Is there someone who can provide that information?
I always appreciate those willing to help.
try to get your head around this...Quote: bcmarshallIt's clear that some hard hands are unwinnable. For example, 16 vs. 10. Next card in the deck is a 10, while dealer has 20. Hit or stand makes no difference. It's unwinnable.
What I'd be fascinated to see is the results of only winnable hands, with those unwinnable removed from the equation. I'd love to see the odds of Winning/Losing for each of the hard hands where a decision will actually affect the outcome.
Is there someone who can provide that information?
I always appreciate those willing to help.
https://www.youtube.com/watch?v=jCF-Btu5ZCk
Quote: bcmarshallIt's clear that some hard hands are unwinnable. For example, 16 vs. 10. Next card in the deck is a 10, while dealer has 20. Hit or stand makes no difference. It's unwinnable.
What I'd be fascinated to see is the results of only winnable hands, with those unwinnable removed from the equation. I'd love to see the odds of Winning/Losing for each of the hard hands where a decision will actually affect the outcome.
Is there someone who can provide that information?
I always appreciate those willing to help.
So, let me ask questions first:
This is like a version of European No Peek: No card is initially dealt to the dealer for his down card? You are one-on-one, or at third base at the table, so the next card in the deck either goes to you or, if you stand, to the dealer? And if the dealer gets an Ace, dealer has blackjack and you lose? Do you care how many decks?
If you'll answer those questions, I will do the full calculation for 10-6 vs 10.
| Next Card | STAND EV | HIT EV |
|---|---|---|
| Ten | -1 | -1 |
| 9 | -1 | -1 |
| 8 | -1 | -1 |
| 7 | -1 | -1 |
| 6 | 0.3125 | -1 |
| 5 | 0.17001 | 0.8102 |
| 4 | 0.11186 | 0.4354 |
| 3 | 0.03781 | -0.01701 |
| 2 | -0.01959 | -0.2411 |
| Ace | -1 | -0.4627 |
This calculation assumed 8 decks. The probabilities of the next card being a 10 or 6 are slightly lowered by the removal of two Tens and one 6 from the deck to form the 16v10 scenario.
Quote: gordonm888
This calculation assumed 8 decks. The probabilities of the next card being a 10 or 6 are slightly lowered by the removal of two Tens and one 6 from the deck to form the 16v10 scenario.
| Average | STAND EV | HIT EV |
|---|---|---|
| Average | -0.575 * | -0.56826 ** |
* (0.8102+0.4354 -0.01701 -0.2411 -0.4627-8)/13
** (0.3125 +0.17001 +0.11186+0.03781 -0.01959 -8)/13
Is that correct? Seems a very close call with hit being favoured as the least worst option.
But one must assume the dealer has a 10 and the next card out is a 10 (-:Quote: mosesThe are only 12 cards in a deck of 52 that will take your 16 to a 19, 20, and 21 which are considered winnable hands. Therefore, some theories suggest always stand vs 10.
Not so fast my friend. Dealer could also have a 7,8,9 in the hole to stop. But if you draw anything above a 6. The fat lady is singing.Quote: AxelWolfBut one must assume the dealer has a 10 and the next card out is a 10 (-:
I can't control what the dealer has in the hole. But it is in my control to take the next card or not.
Quote: mosesThe are only 12 cards in a deck of 52 that will take your 16 to a 19, 20, and 21 which are considered winnable hands. Therefore, some theories suggest always stand vs 10.
Aces and Twos do not bust your hand, so they have some influence as well, just not as much as 5,4,and 3..
In this game, you do not know that the dealer's "down-card is NOT an Ace" - that flips the odds slightly towards hit.
So the 10 is up. Lets say you have 10,6. 12 cards will likely get you to the winners circle. The Ace and 2 improve your hand. But not enough to win. So the remaining 29 cards will have the player saying "o, crap" before the dealer does a thing.Quote: gordonm888Aces and Twos do not bust your hand, so they have some influence as well, just not as much as 5,4,and 3..
In this game, you do not know that the dealer's "down-card is NOT an Ace" - that flips the odds slightly towards hit.
So you hit a 5 for 21 and dealer has Ace in hole? That's going to have mamy folks rubbing poop in their hair.😲 Now wouldnt that be a whizzer.
Quote: mosesThe are only 12 cards in a deck of 52 that will take your 16 to a 19, 20, and 21 which are considered winnable hands. Therefore, some theories suggest always stand vs 10.
As I see it.....
If you stand, the only way to win is if the dealer busts. I used to know these numbers better but it has been years, if not decades since I studied them, but the dealer will bust with a ten about 20-21 percent of the time, so standing loses 4 out of 5 hands.
If you hit, you can win, lose or tie. If you stand you can only hit or lose.
In the long run, hitting is better. You will lose less than eighty percent of the hands.
Well said. If memory serves me correctly, the difference between always standing vs always hitting is minimal. So the happy medium for best results is RC 0. Hit if below. Stand if above.Quote: billryanAs I see it.....
If you stand, the only way to win is if the dealer busts. I used to know these numbers better but it has been years, if not decades since I studied them, but the dealer will bust with a ten about 20-21 percent of the time, so standing loses 4 out of 5 hands.
If you hit, you can win, lose or tie. If you stand you can only hit or lose.
In the long run, hitting is better. You will lose less than eighty percent of the hands.
I have my own way. But dont want to get hit by a rifle ball from one of the bjtf'ers taking a pot shot my direction.😄
If you understand probability theory, you can calculate and compare the expected values for HIT and STAND, and whichever has the highest expected value is the best option. HIT is the best option.
Anyone saying "some theories suggest. . . STAND is the better option" has some explaining to do.
There were self professed APs at BJTF who suggested always standing. There a few 16vs10 threads with many posts. One would have to go back and review as to who said what.
Most of these theories are thresholds which amount to very little in the long run. A one size fits all? Hmmm. I dont think so.
Of course, free play on Electronic Blackjack is the only time I’d ever play Blackjack without being able to surrender, so it’s a moot question as far as physical tables go.
I don’t care for Blackjack, but Surrendering, when appropriate, is at least fun.
Quote: unJonStanding on 16 v 10 is a cheap cover play if you are counting hi lo.
The ones stating this were shoe players. East coast for the most part.
In a pitch game, passing up on hitting in deep negative territory is very expensive cover. Not sure about HiLo. Bad count for a pitch game.
Quote: Mission146I understand the negligible EV costs and conclude that busting on a hit p***** me off enough that I am willing to stand to avoid it on this one decision point.
For US games with dealer peek, HITTING a 10-6vs10 is better by 0.5% than STANDING (From a fresh shoe, i.e. without a skewed count.) However STANDING on a 5-4-5-2vs10 is about 1% better than hitting (again without removing any other cards from the deck.
And then consider that in U.S. games dealt from a shoe that don't offer surrender, the most frequently occurring strategy decisions that one needs to make are '16 vs 10' and '15 vs 10.' (I am not counting hands like '20 vs 10' or '17 vs 10', which are decisions that are never altered due to the removal of other cards from the deck.)
So, I have always firmly believed that 16v10 is the one strategy decision that is most profitable to study and master, because it occurs so frequently and because it is close, i.e., it is so susceptible to influence by changes in card composition of the shoe. The EOR for 6 influences you in the direction of hitting and the EOR for 5 influences you in the direction of standing, so Hi-Lo and other systems that count 5 and 6 as similar cards are inefficient tools for exploiting the 16 vs 10 decision. But if you find ways of getting closer to computer-perfect play on 16 vs 10, and mix in a little bit of acting when at the live tables, I think its a worthwhile tool in your BJ arsenal, particularly at significant shoe penetration.
Quote: gordonm888Quote: Mission146I understand the negligible EV costs and conclude that busting on a hit p***** me off enough that I am willing to stand to avoid it on this one decision point.
For US games with dealer peek, HITTING a 10-6vs10 is better by 0.5% than STANDING (From a fresh shoe, i.e. without a skewed count.) However STANDING on a 5-4-5-2vs10 is about 1% better than hitting (again without removing any other cards from the deck.
And then consider that in U.S. games dealt from a shoe that don't offer surrender, the most frequently occurring strategy decisions that one needs to make are '16 vs 10' and '15 vs 10.' (I am not counting hands like '20 vs 10' or '17 vs 10', which are decisions that are never altered due to the removal of other cards from the deck.)
So, I have always firmly believed that 16v10 is the one strategy decision that is most profitable to study and master, because it occurs so frequently and because it is close, i.e., it is so susceptible to influence by changes in card composition of the shoe. The EOR for 6 influences you in the direction of hitting and the EOR for 5 influences you in the direction of standing, so Hi-Lo and other systems that count 5 and 6 as similar cards are inefficient tools for exploiting the 16 vs 10 decision. But if you find ways of getting closer to computer-perfect play on 16 vs 10, and mix in a little bit of acting when at the live tables, I think its a worthwhile tool in your BJ arsenal, particularly at significant shoe penetration.
Point taken. I maybe shouldn't have said, "Negligible," but should have said, "Acceptable to me in the rare event that I play."
I was referring to the two-card 16 v. dealer 10 scenario specifically in my post, and by, "Negligible EV," I was referring to the EV cost in the overall game of that decision point combined with the probability of being faced with that decision in the first place.
I also consider your second paragraph and agree with what you have said. For me, in the extraordinarily unlikely event that I would ever play Live Blackjack for recreational purposes, I would not play at a table in which I could not Surrender anyway, so, in effect, my decision against a Dealer 10 will always be to Surrender.
In theory, I would be willing to play Blackjack at an advantage, even though it bores me to tears after about ten minutes, but I have not been presented with such an opportunity in recent years.
I also agree with your final paragraph, especially if someone is going to be playing Blackjack frequently. Speaking in terms of recreational play only, I would never play without Surrender and I would never play with a base bet of more than $5, so recreationally-speaking, this situation will never apply to me anyway because I would just Surrender.
Quote: gordonm888Quote: Mission146I understand the negligible EV costs and conclude that busting on a hit p***** me off enough that I am willing to stand to avoid it on this one decision point.
For US games with dealer peek, HITTING a 10-6vs10 is better by 0.5% than STANDING (From a fresh shoe, i.e. without a skewed count.) However STANDING on a 5-4-5-2vs10 is about 1% better than hitting (again without removing any other cards from the deck.
And then consider that in U.S. games dealt from a shoe that don't offer surrender, the most frequently occurring strategy decisions that one needs to make are '16 vs 10' and '15 vs 10.' (I am not counting hands like '20 vs 10' or '17 vs 10', which are decisions that are never altered due to the removal of other cards from the deck.)
So, I have always firmly believed that 16v10 is the one strategy decision that is most profitable to study and master, because it occurs so frequently and because it is close, i.e., it is so susceptible to influence by changes in card composition of the shoe. The EOR for 6 influences you in the direction of hitting and the EOR for 5 influences you in the direction of standing, so Hi-Lo and other systems that count 5 and 6 as similar cards are inefficient tools for exploiting the 16 vs 10 decision. But if you find ways of getting closer to computer-perfect play on 16 vs 10, and mix in a little bit of acting when at the live tables, I think its a worthwhile tool in your BJ arsenal, particularly at significant shoe penetration.
This is a pivitol hand. I dont think it's that big of a tell. If so, you've aleady done something to get the EITS attention.
Stand on positive. Hit on negative. Flip on 0. Hitting at deep negative is where most 16's are won (in that 20%). Even in a loss, it's likely your move will take one or more smaller cards out of the deck. Shoes go on your feet.
Quote: OnceDeartry to get your head around this...
/watch?v=jCF-Btu5ZCk
I thought that was an excellent video, and I was amazed at the facility that you have with Excel.
Maybe I didn't understand the results presented, or perhaps my question wasn't clear. It's pretty obvious that the outcome of a 16 v 10 is dire no matter what. The math shows it's better to hit than to stand, and that math presented takes those unwinnable hands into consideration.
It wouldn't change anything, but for myself I'd like to know what the actual percentage of wins vs. losses would be on only those hands were a decision makes a difference. Or said another way, it would be nice to know that no matter what I do I'm gonna lose x% of the time. Knowing that would temper expectations a bit.
Your own calculator shows odds of -0.540954 for stand and -0.534676 for hit. There's not a huge difference but odds do favor the hit. If one imagines that the unwinnable hands were eliminated from the mix so only those where a decision matters are considered, then it would be a much more simple calculation. Throwing out numbers for example's sake without basing them on anything, there might be a 55%-45% benefit to hitting over standing for example, something much easier to visualize
Sure, it all goes back the actual odds from the calculator, but it seems to me that one could see more clearly the advantages
Quote: gordonm888So, let me ask questions first:
This is like a version of European No Peek: No card is initially dealt to the dealer for his down card? You are one-on-one, or at third base at the table, so the next card in the deck either goes to you or, if you stand, to the dealer? And if the dealer gets an Ace, dealer has blackjack and you lose? Do you care how many decks?
If you'll answer those questions, I will do the full calculation for 10-6 vs 10.
In all honesty, I wasn't envisioning any of those scenarios. I'm talking about a straight blackjack game, dealer's hole card buried, dealer has peeked for A, 6 decks, and no consideration of position at the table. It's just a general question.
As I said in my previous reply to another post, I think the end result would tell a simplified actual percentage advantage of the 100% of potentially winnable hands between hitting and standing. 40% vs. 60% of winnable hands, for example.
Now I have the ability to know exactly what's been played and what still remains by grouping or column counting in one or two decks. Not sure if that's a blessing or a curse.
Point is, Ive been in the position of power to where only one card remaining could hurt me. It did.
For shoes? Your question is a needle in a haystack of minimum bets.
The sim uses 2 card 16. Pecentage of hands dealt is 1.916%
Always Stands Results. 21.20% wins. 78.80%. lost.
Always Hit Results. 19.50% wins. 74.80% lost. 5.7% ties.
So IF you played 100k hands. This would occur 1,916 times.
Standing would win 406 and lose 1510 times.
Hitting would win 374 times lose 1433 and push 109.
An average bet of $25 per hand is $27,600 lost by always standing.
$26,475 lost by always hitting. A difference of $1,125 over the courae of 100k hands. Or about two years of play for the most intense competitors.
This is why card counting is important. The more positive the deck, the more you bet. So the average bet goes up. At a time where hiting will mean a higher percentage of losses.
So SCORE was significantly higher by always standing as opposed to always hitting.
Quote: mosesI have my own way. But dont want to get hit by a rifle ball from one of the bjtf'ers taking a pot shot my direction.😄
I'd be interested in hearing FWIW.
Quote: bcmarshallI'd be interested in hearing FWIW.
Lol. No...you wouldn't. You'd have to break out a bottle of Excedrin.
But I would suggest looking into products offered by Qfit.
I like CV Data. Norm suggested it years ago for what I wanted to accomplish. He was spot on. But I think the more popular sim product is CVCX.
FWIW. Might not want to mention the word Moses in those parts.😉😄
Quote: mosesLol. No...you wouldn't. You'd have to break out a bottle of Excedrin.
But I would suggest looking into products offered by Qfit.
I like CV Data. Norm suggested it years ago for what I wanted to accomplish. He was spot on. But I think the more popular sim product is CVCX.
FWIW. Might not want to mention the word Moses in those parts.😉😄
Tell Norm KewlJ sent you.
Quote: mosesI was curious and had forgot the results. So I ran a sim.
The sim uses 2 card 16. Pecentage of hands dealt is 1.916%
Always Stands Results. 21.20% wins. 78.80%. lost.
Always Hit Results. 19.50% wins. 74.80% lost. 5.7% ties.
So IF you played 100k hands. This would occur 1,916 times.
Standing would win 406 and lose 1510 times.
Hitting would win 374 times lose 1433 and push 109.
An average bet of $25 per hand is $27,600 lost by always standing.
$26,475 lost by always hitting. A difference of $1,125 over the courae of 100k hands. Or about two years of play for the most intense competitors.
This is why card counting is important. The more positive the deck, the more you bet. So the average bet goes up. At a time where hiting will mean a higher percentage of losses.
So SCORE was significantly higher by always standing as opposed to always hitting.
Wait! I'm confused! It's not possible that less than 2% of hands are a 16. I must be missing a point you're making.
Also, you're saying standing is better than hitting.
Please explain again!
Quote: mosesI was curious and had forgot the results. So I ran a sim.
The sim uses 2 card 16. Pecentage of hands dealt is 1.916%
Always Stands Results. 21.20% wins. 78.80%. lost.
Always Hit Results. 19.50% wins. 74.80% lost. 5.7% ties.
So IF you played 100k hands. This would occur 1,916 times.
Standing would win 406 and lose 1510 times.
Hitting would win 374 times lose 1433 and push 109.
An average bet of $25 per hand is $27,600 lost by always standing.
$26,475 lost by always hitting. A difference of $1,125 over the courae of 100k hands. Or about two years of play for the most intense competitors.
This is why card counting is important. The more positive the deck, the more you bet. So the average bet goes up. At a time where hiting will mean a higher percentage of losses.
So SCORE was significantly higher by always standing as opposed to always hitting.
Using your numbers, someone who always stands will make approx. 4% less cash over that period.
Quote: bcmarshallWait! I'm confused! It's not possible that less than 2% of hands are a 16. I must be missing a point you're making.
Also, you're saying standing is better than hitting.
Please explain again!
The sim results did not include 8,8 or 3+ card 16s. So yes, the percentage of hands played would be more. But I doubt the end result would differ much.
IF you are flat betting $25 a hand. Then you'd have $1,125 more if you always hit as opposed to always standing...after 100k hands.
Now if you're counting cards, it stands to reason, you'd be betting at a higher TC. No? Therefore, your bet per hand average would go up at a time when even more 10s are in the deck than the start. So, you have a higher percentage of losses than the start of the deck by hitting and getting 26.😉
The sim only processes the data it is fed. SCORE suggests standing IF you are increasing your bets in a positive count.
The purpose of counting is to know when to hit or stand in order to maximize the return of your investment.
I will have sim results in a few minutes.
You win 21% lose 75.40%. Push 3.6%.
win 402 hands. lose 1,444. tie 69.
You lose $26,050 over the course of 100k hands.
So this puts another $425 in your pocket.
However, the sim will factor in your average bet by the amount you decide to adjust your bet for positive counts.
In other words SCORE will soar. THIS, my friend, is why counting is advantageous to basic strategy.
You may have missed the smiley face.Quote: mosesNot so fast my friend. Dealer could also have a 7,8,9 in the hole to stop. But if you draw anything above a 6. The fat lady is singing.
I can't control what the dealer has in the hole. But it is in my control to take the next card or not.
