scoring hard double downs

[sagefr0g]

from looking at QFIT's chart:
http://www.blackjackincolor.com/blackjackfirstcards1.htm
it appears that our chances of having the opportunity to make a hard double down are less than 11% of the time playing a six deck shoe.
and it seems ironic that we are most hopeful for these hard double downs when we have more money out ie. in positive true counts. a time when there are less low cards remaining to be dealt.
interesting to note that those hard doubledowns come principly from a two card initial hand of 11 made up of 9,2 8,3 7,4 6,5
or 10 made up of 8,2 7,3 6,4 5,5
or 9 made up of 7,2 6,3, 5,4
or 8 made up of 6,2 5,3 4,4 (assuming index plays)

in the case of hard 11 we get to double against almost every dealer upcard
in the case of hard 10 we get to double on only the low and neutral upcards
in the case of hard 9 we get to doulbe on only the low dealer upcards
excepting for index plays.

no big deal i just think it's interesting and seems ironic as these are a big part of our money making plays.

 

[sagefr0g]

just wonder what anyone thinks about double downs

have a few questions in case anyone knows about these things:

still in referance to the first post http://www.blackjackinfo.com/bb/showpost.php?p=69665&postcount=1
where it just seems so ironic, well even parodoxical to me.
but what i wonder because i don't really know has to do with the thing we all believe and that is as far as winning money off of types of hands over the long run is:
1. we lose money on the normal hands
2. we win money on our double downs
3. we win money by getting blackjacks
4. we win money on proper insurance bets

well 1, 3, & 4 makes perfect logical sense to me.
an i accept that 2 is true and infact i know it tends to be a sginificant part of the money won equation. but i find it difficult to understand how it is so.
i mean yeah you are doubling your bet and if you win doing that more than you lose then your going to come out ahead. and i can see how if you manage to catch a hard double down potential hand in a high true count how it is likely that you would catch a ten. thing is to me it seems so unlikely to catch a hard double down potential hand in the first place. and sure it does happen that you happily get those hard double down hands durring high true counts. and i suppose there is some mathematical frequency that a player can expect to get a hard double down hand at some given positive true count. would to love to know what that frequency is or if anyone even knows.
but i have to wonder about these double down in decent true counts, i mean is just all we can do is cast our nets of optimal bets in the various degrees of high true counts and accept the fact that we get our fair share of those crucial double down opportunities? it would seem there should be some clue by our knowledge of the cards that have been played as to when it would be more likely to catch the cards needed to make up a hard double down hand.
pipe dream perhaps but wow would it be nice!
sorry for the rant but i'd be interested to know any thoughts or knowledge about this issue.
oh and one other thing i wonder about hard double downs is are they still profitable overall when the true count is zero or negative?

 

[k_c]

have a few questions in case anyone knows about these things:

still in referance to the first post http://www.blackjackinfo.com/bb/showpost.php?p=69665&postcount=1
where it just seems so ironic, well even parodoxical to me.
but what i wonder because i don't really know has to do with the thing we all believe and that is as far as winning money off of types of hands over the long run is:
1. we lose money on the normal hands
2. we win money on our double downs
3. we win money by getting blackjacks
4. we win money on proper insurance bets

well 1, 3, & 4 makes perfect logical sense to me.
an i accept that 2 is true and infact i know it tends to be a sginificant part of the money won equation. but i find it difficult to understand how it is so.
i mean yeah you are doubling your bet and if you win doing that more than you lose then your going to come out ahead. and i can see how if you manage to catch a hard double down potential hand in a high true count how it is likely that you would catch a ten. thing is to me it seems so unlikely to catch a hard double down potential hand in the first place. and sure it does happen that you happily get those hard double down hands durring high true counts. and i suppose there is some mathematical frequency that a player can expect to get a hard double down hand at some given positive true count. would to love to know what that frequency is or if anyone even knows.
but i have to wonder about these double down in decent true counts, i mean is just all we can do is cast our nets of optimal bets in the various degrees of high true counts and accept the fact that we get our fair share of those crucial double down opportunities? it would seem there should be some clue by our knowledge of the cards that have been played as to when it would be more likely to catch the cards needed to make up a hard double down hand.
pipe dream perhaps but wow would it be nice!
sorry for the rant but i'd be interested to know any thoughts or knowledge about this issue.
oh and one other thing i wonder about hard double downs is are they still profitable overall when the true count is zero or negative?

I've thought about this. One could ask, "Does the exact state of the spot cards add to or detract from EV?" I think it would be beyond my capabilities to judge in the heat of battle though.

k_c

 

[Ferretnparrot]

I once had this very same thought

later i realized that the probibility of the dealer busting goes up dramatically when the count gets higher, in fact when he shows 4 5 and 6 it can go above (from where it was previously at around 30-42%) 50% meaning it woudl make sense to double and split everything so i think the increased rate at which you get payed on dd negates the rarity of the event itself.

Also you will get more soft hands with more aces at high counts and you often double them.

 

[sagefr0g]

I've thought about this. One could ask, "Does the exact state of the spot cards add to or detract from EV?" I think it would be beyond my capabilities to judge in the heat of battle though.

k_c

not sure what your terminology 'spot' cards means?
but one thing that i think i've noticed in the heat of practice battles on the computer is that say your in the midst of a positive true count. and i don't know say a round, maybe two rounds or so of cards is played out and a unusualy high proportions of those cards just happen to be ten value cards but maybe the true count is still positive then i have these thoughts in my mind that well now this next round comming may have a better chance for that desired combination of a neutral card and a low card or a low card and another low card such that a double down opportunity might be realized sort of thing where you'd still have a good chance of getting a ten should you be so lucky to take a shot at that double down. i don't know if this is just a gambler's fallacy kind of suspicion or one that has any statistical logical basis to it or not.

 

[Kasi]

oh and one other thing i wonder about hard double downs is are they still profitable overall when the true count is zero or negative?

They sure are profitable for a BS player is about all I can tell you for sure. When you are dealt a doubleable hand, you'll probably win 55% of them. You probably have a ~10% +Ev when you are dealt one for a DA2 6-8 deck game.

Of course things change with # of decks, D9 or D10 rules, maybe no DAS rules.

Check out Ken's

http://www.blackjackinfo.com/bjtourn-doublechart.php

and note how many BS doubles have a >50% winning percentage. Not all. But most.

And I think winnning and losing them is taken into account for the overall +EV percent at various TC's for the amount bet and game involved so it's really just part of the game.

 

[k_c]

not sure what your terminology 'spot' cards means?
but one thing that i think i've noticed in the heat of practice battles on the computer is that say your in the midst of a positive true count. and i don't know say a round, maybe two rounds or so of cards is played out and a unusualy high proportions of those cards just happen to be ten value cards but maybe the true count is still positive then i have these thoughts in my mind that well now this next round comming may have a better chance for that desired combination of a neutral card and a low card or a low card and another low card such that a double down opportunity might be realized sort of thing where you'd still have a good chance of getting a ten should you be so lucky to take a shot at that double down. i don't know if this is just a gambler's fallacy kind of suspicion or one that has any statistical logical basis to it or not.

I'm not sure what I mean either lol, but what if you knew there were a lot of 9s,2s, and 6s. You'd have a good chance of drawing 9-2 v 6. Extra 9s would help bust the 6 up and wouldn't be a bad draw to the double down. On the other side of the coin, a 2 wouldn't be a good card to draw to the double down. I can tell you though that from messing around with my CA weirded out compositions can be found that can give an advantage by using weirded out plays, but my approach is to stay away from the weirded out comps or at least not to bet into them if I can help it. I feel if a situation doesn't give +EV using basic strategy, it's not realistically beatable.

k_c

 

[sagefr0g]

They sure are profitable for a BS player is about all I can tell you for sure. When you are dealt a doubleable hand, you'll probably win 55% of them. You probably have a ~10% +Ev when you are dealt one for a DA2 6-8 deck game.

Of course things change with # of decks, D9 or D10 rules, maybe no DAS rules.

so i'm taking it that in general hard double down hands are profitable regardless of true count. but i'm guesing they must be a rather significant degree more profitable (perhaps more likely to win them) when the true count is positive. thats something i don't really know but it seems to make sense.
would be interesting if a simulation could pinpoint this fact or not. i just wonder if kc's program could do that.....

Check out Ken's

http://www.blackjackinfo.com/bjtourn-doublechart.php

and note how many BS doubles have a >50% winning percentage. Not all. But most.

And I think winnning and losing them is taken into account for the overall +EV percent at various TC's for the amount bet and game involved so it's really just part of the game.

that's a great chart and thank you very much both Kasi & Ken lol.
it's to me that these double down's are such a significant contributor to money won and the nature of the hand compositions needed so limited in scope and probably most important in positive true counts where the cards are in essence stacked against them presenting that one just salivates at the idea that it might be predictable as to when they are most likely to present. it seems we already know the score as to when blackjacks and succesful insurance bets are most likely to present but with regards to the double downs presenting it just seems to be a more luck of the draw sort of pre-cognition we can attain. so i guess for now it's just that we know that if we are fortunate enough to get them in a positive true count when the money is out that it is likely to be very profitable lol

 

[k_c]

so i'm taking it that in general hard double down hands are profitable regardless of true count. but i'm guesing they must be a rather significant degree more profitable (perhaps more likely to win them) when the true count is positive. thats something i don't really know but it seems to make sense.
would be interesting if a simulation could pinpoint this fact or not. i just wonder if kc's program could do that.....

Double downs are NOT always profitable regardless of true count. It's a piece of cake for the program to show this. Just remove a bunch of high cards, input a player hand that might be doubled, and click 'Compute'. You will find that a level of high card removals exists for each up card where a previously positive double EV turns negative for each up card. That is what indices are about. The advantage of the program is that you can quickly find the EVs for any specific shoe state. The disadvantage is that this is a piece-meal type of thing, so the best index may not be apparent after just one calculation but you can pretty quickly get an idea of what range the index may be in.

k_c

 

[sagefr0g]

Double downs are NOT always profitable regardless of true count. It's a piece of cake for the program to show this. Just remove a bunch of high cards, input a player hand that might be doubled, and click 'Compute'. You will find that a level of high card removals exists for each up card where a previously positive double EV turns negative for each up card. That is what indices are about. The advantage of the program is that you can quickly find the EVs for any specific shoe state. The disadvantage is that this is a piece-meal type of thing, so the best index may not be apparent after just one calculation but you can pretty quickly get an idea of what range the index may be in.

k_c

thank you kc and what you say pretty much connfirms how i suspected it was. the thing on these hard doubledowns (talkin the normal basic strategy ones) is i think most of them require a relatively low true count before the basic strategy departure kicks in. so i think that woud mean they give one the best result one could expect for a pretty wide spectrum of true counts.
i just wonder does your program indicate if the EV falls perceptively for some given hard double down at some given true count? i mean lets say some double down hand has an ev = 0.58 at some tc and then you check your program at a lower tc and it might indicate an ev = 0.45 ? just curious. thanks again. i suppose it does as you say it will show negative in some instance of card removal where it had been a positive ev for another case.

 

[k_c]

Example

thank you kc and what you say pretty much connfirms how i suspected it was. the thing on these hard doubledowns (talkin the normal basic strategy ones) is i think most of them require a relatively low true count before the basic strategy departure kicks in. so i think that woud mean they give one the best result one could expect for a pretty wide spectrum of true counts.
i just wonder does your program indicate if the EV falls perceptively for some given hard double down at some given true count? i mean lets say some double down hand has an ev = 0.58 at some tc and then you check your program at a lower tc and it might indicate an ev = 0.45 ? just curious. thanks again. i suppose it does as you say it will show negative in some instance of card removal where it had been a positive ev for another case.

If 20 tens, 5 aces, 2 sevens, 2 eights, and 2 nines are removed from a 6 deck shoe and from this shoe dealer is dealt an up card of 10:

Player hand
9-2           Dbl EV=+9.550%, Hit EV=+9.574% using best strategy
8-3           Dbl EV=+9.878%, Hit EV=+9.623%
7-4           Dbl EV=+10.07%, Hit EV=+9.592%
6-5           Dbl EV=+10.11%, Hit EV=+9.720%
The pre-deal overall EV for this shoe using best strategy = -2.714
asssuming S17, split to 4 hands, DAS. Total dependent basic strategy
EV will be a bit worse.

This shoe comp computes to a Hi-Lo true count of about -4.6. If an index of -5 for hard 11 versus 10 is used it's going to be about right or not very far off the mark.

Full shoe EVs are -
9-2 Dbl: +17.39%, Hit: +11.70%
8-3 Dbl: +17.69%, Hit: +11.79%
7-4 Dbl: +17.84%, Hit: +11.79%
6-5 Dbl: +17.85%, Hit: +11.86%
Full shoe overall EV for above conditions = -.4079% using best strategy
For reference, total dependent basic strategy EV = -.4128%

You can see the trend is that the more high cards that are removed the less the double is worth and also at some point hitting becomes better than doubling. For a wonger this isn't a very important index because a wonger would have already left the game long before a -5 TC occurs.

I haven't figured the probability of a player hand of hard 11 for the negative count, but I'm pretty sure it is higher than for a full shoe. In the big picture, this doesn't do any awful lot of good.

Hope this helps.
k_c

 

[blackjack avenger]

You Mean Side Counts

I'm not sure what I mean either lol, but what if you knew there were a lot of 9s,2s, and 6s. You'd have a good chance of drawing 9-2 v 6. Extra 9s would help bust the 6 up and wouldn't be a bad draw to the double down. On the other side of the coin, a 2 wouldn't be a good card to draw to the double down. I can tell you though that from messing around with my CA weirded out compositions can be found that can give an advantage by using weirded out plays, but my approach is to stay away from the weirded out comps or at least not to bet into them if I can help it. I feel if a situation doesn't give +EV using basic strategy, it's not realistically beatable.

k_c

Side counting neutral cards 7,8,9. I believe mostly 7s can have value in single deck but very very very little value in multi deck.

 

[blackjack avenger]

Soft Doubles = Not Much

I once had this very same thought

later i realized that the probibility of the dealer busting goes up dramatically when the count gets higher, in fact when he shows 4 5 and 6 it can go above (from where it was previously at around 30-42%) 50% meaning it woudl make sense to double and split everything so i think the increased rate at which you get payed on dd negates the rarity of the event itself.

Also you will get more soft hands with more aces at high counts and you often double them.

Soft doubles have very little value in multi deck even with indices.

 

[blackjack avenger]

Knowledge of The Cards = Card Counting

have a few questions in case anyone knows about these things:

still in referance to the first post http://www.blackjackinfo.com/bb/showpost.php?p=69665&postcount=1
where it just seems so ironic, well even parodoxical to me.
but what i wonder because i don't really know has to do with the thing we all believe and that is as far as winning money off of types of hands over the long run is:
1. we lose money on the normal hands
2. we win money on our double downs
3. we win money by getting blackjacks
4. we win money on proper insurance bets

well 1, 3, & 4 makes perfect logical sense to me.
an i accept that 2 is true and infact i know it tends to be a sginificant part of the money won equation. but i find it difficult to understand how it is so.
i mean yeah you are doubling your bet and if you win doing that more than you lose then your going to come out ahead. and i can see how if you manage to catch a hard double down potential hand in a high true count how it is likely that you would catch a ten. thing is to me it seems so unlikely to catch a hard double down potential hand in the first place. and sure it does happen that you happily get those hard double down hands durring high true counts. and i suppose there is some mathematical frequency that a player can expect to get a hard double down hand at some given positive true count. would to love to know what that frequency is or if anyone even knows.
but i have to wonder about these double down in decent true counts, i mean is just all we can do is cast our nets of optimal bets in the various degrees of high true counts and accept the fact that we get our fair share of those crucial double down opportunities? it would seem there should be some clue by our knowledge of the cards that have been played as to when it would be more likely to catch the cards needed to make up a hard double down hand.
pipe dream perhaps but wow would it be nice!
sorry for the rant but i'd be interested to know any thoughts or knowledge about this issue.
oh and one other thing i wonder about hard double downs is are they still profitable overall when the true count is zero or negative?

Knowledge of the cards is what card counting is. The hands that have the most value are TT and AT. That is what we are looking for. With the doubling hands of 8 to 11 you are more likely to get a T for a good hand and the dealer with their potenitally stiff 2-6 is more likely to break. There are negative indices that indicate when not to double. They are for the most part rather negative.

 

[sagefr0g]

Knowledge of the cards is what card counting is. The hands that have the most value are TT and AT. That is what we are looking for. With the doubling hands of 8 to 11 you are more likely to get a T for a good hand and the dealer with their potenitally stiff 2-6 is more likely to break. There are negative indices that indicate when not to double. They are for the most part rather negative.

well i can see where AT is certainly valuable and well i know TT is also so as to be able to smooth out those losing normal hands. but i'm thinking these potential double down hands of 8 through 11 must be pretty darned important.

If 20 tens, 5 aces, 2 sevens, 2 eights, and 2 nines are removed from a 6 deck shoe and from this shoe dealer is dealt an up card of 10:

Player hand
9-2           Dbl EV=+9.550%, Hit EV=+9.574% using best strategy
8-3           Dbl EV=+9.878%, Hit EV=+9.623%
7-4           Dbl EV=+10.07%, Hit EV=+9.592%
6-5           Dbl EV=+10.11%, Hit EV=+9.720%
The pre-deal overall EV for this shoe using best strategy = -2.714
asssuming S17, split to 4 hands, DAS. Total dependent basic strategy
EV will be a bit worse.

This shoe comp computes to a Hi-Lo true count of about -4.6. If an index of -5 for hard 11 versus 10 is used it's going to be about right or not very far off the mark.

Full shoe EVs are -
9-2 Dbl: +17.39%, Hit: +11.70%
8-3 Dbl: +17.69%, Hit: +11.79%
7-4 Dbl: +17.84%, Hit: +11.79%
6-5 Dbl: +17.85%, Hit: +11.86%
Full shoe overall EV for above conditions = -.4079% using best strategy
For reference, total dependent basic strategy EV = -.4128%

You can see the trend is that the more high cards that are removed the less the double is worth and also at some point hitting becomes better than doubling. For a wonger this isn't a very important index because a wonger would have already left the game long before a -5 TC occurs.

I haven't figured the probability of a player hand of hard 11 for the negative count, but I'm pretty sure it is higher than for a full shoe. In the big picture, this doesn't do any awful lot of good.

Hope this helps.
k_c

geesh...... wow it's gonna take me a while to digest and understand all that but i get the gist, at least i think i do.

but anyway if you want a laugh. i actually went to a real live casino yesterday and was trying out my fuzzy count. truth be known i was lost as hell the whole time i played which was almost four hours lol. i seem to do pretty well with the FC approach when i practice on the computer but one advantage of the computer is you can set up cvbj to make the rounds fly by really fast. in the casino it's slow as hell so what that does is puts me sort of out of my element to where i lose track of where i think i'm at. lol
and so i was mainly just playing basic strategy and making a few futile attempts at some low multiple unit bets. i was down about half or more of my trip bank so by my guess i was on the sour side of basic strategy standard deviation, not to mention my failed small bet increases. then came a shoe that i was admittedly still pretty much lost in but that i was pretty sure was in some significant positive true count territory. there was about five players at the table and darned near everyone of them except me (of course i had a stiff) pulled a twenty and so did the dealer i think.
so i thought to hell with it i know this is not a proper orthodox procedure to follow but i reasoned that since all those tens had come out that just maybe some of those crucial combination of low cards or low card, neutral card combinations just might come out and i might get lucky enough to get one on two hands for a double down shot. so i had my max bet out 12 units. and dammed if i didn't get a twenty on one hand and an eight on the other hand against a dealer's six.
sucked in my breath and doubled that eight (yeah only good at tc=2) and almost but didn't split the tens lol. got a ten on the double down and the dealer goes on to bust. lol . and then after that i had some lucky winning hands with small bets to the point when i left i was only down about six and a half units. just a story related to the power of double downs i think but i know that it's just anectdotal and who knows what really happened. i sure don't lol.

 

[sagefr0g]

this seems applicable for scoring double downs

if any one is interesting the link below i think is a clue about hopefully scoring those crucial double downs:
http://www.blackjackinfo.com/bb/showpost.php?p=70977&postcount=26

 

[sagefr0g]

They sure are profitable for a BS player is about all I can tell you for sure. When you are dealt a doubleable hand, you'll probably win 55% of them. You probably have a ~10% +Ev when you are dealt one for a DA2 6-8 deck game.

Of course things change with # of decks, D9 or D10 rules, maybe no DAS rules.

Check out Ken's

http://www.blackjackinfo.com/bjtourn-doublechart.php

and note how many BS doubles have a >50% winning percentage. Not all. But most.

And I think winnning and losing them is taken into account for the overall +EV percent at various TC's for the amount bet and game involved so it's really just part of the game.

ok i get your drift. i mean at least part of the point is to double down when basic strategy says to. it's worth it.
thing is look at these charts. especially the second chart where it compares doubling down with indexs and with out.
http://www.blackjackincolor.com/truecount6.htm
dramatic differance and QFIT says basicly in those negative counts the indexs save us from making bad doubles.
for me this is significant because i do a lot of play all six deckers. thing is i just use i18 departures. judging from QFIT's chart there must be a fair number of anti-double down departures that i'm not even aware of.
for one thing i think the i18 is sort of slanted towards the positive true count player only, so i guess i need to dig up a i18 or what ever for the negative counts as well. not that i don't sometimes leave the table on negative counts but i hang and play as well. blackjack avenger was kind enough to throw some depatures out in another post http://www.blackjackinfo.com/bb/showpost.php?p=68630&postcount=16
guess i'm rambling on here. just these double downs right now fascinate me.
but from qfits charts kasi doesn't it look to you like a lot of the money made on double downs comes as a result of ramping the moderate tc's ? thought you might be interested in that consideration as i think you hold back on the ramping some. not being critical of that just something to keep in mind i guess.

 

[sagefr0g]

I once had this very same thought

later i realized that the probibility of the dealer busting goes up dramatically when the count gets higher, in fact when he shows 4 5 and 6 it can go above (from where it was previously at around 30-42%) 50% meaning it woudl make sense to double and split everything so i think the increased rate at which you get payed on dd negates the rarity of the event itself.

Also you will get more soft hands with more aces at high counts and you often double them.

yeah that's one of the interesting things about double downs. how you got sort of two outs if you will. to where you double (free of fear of busting) and get a good hand likely to beat or push the dealer.... or if you don't get such a good hand at least you still don't bust and there is still the possibility the dealer will bust.