Help with incorporating indexes into basic strategy
For example, #1. It makes sense to take insurance when the count is +3 or greater, different from BS. Under BS it's easy to remember: never take insurance. This index play reflects the influence of a positive count when you know that insurance is a totally separate wager: is there a ten under that ace. What I am not sure about is whether to take even money for a blackjack when the count is >= +3. That's the practical effect of insurance when you have a blackjack. Does the "take insurance when count >= +3" apply in this case as well?
There are other instances where the variances from BS in the various indexes are not so clear, at least to me. Is there a chart that has two more columns next to the three that are listed, for example: "Order.....Play.....Index....." and then "Normally.....But instead do this"? These two added columns would remove any uncertainty, at least for me. There are so many different games out there, maybe the "Normally" column won't make sense (I play at Mohegan, at both S17 and H17 tables), but clarity as to "instead do this" would be nice.
As an extremely anal person, I have spent countless hours examining every chart I have ever seen on Basic Strategy. There are parts that make sense to me as the result of thinking about it, and for the parts that don't I accept as a matter of faith what I read here. A lot of the math that you folks present here makes my eyes roll back in my head, but it does convince me that you DO know what you are talking about. There are some index moves for which
I can't say for sure what the alternative is to BS. It would be nice if there was a chart that states clearly what to do, so that I can blindly obey.
That's what I like about this game. There are no emotions or feelings involved. Find the cell in the mental spreadsheet. Do what it says. I am ready to add another set of coulmns and rows in my head. Can you help?
9,9 against 7 at +3. (this isn't an ill 18 play) Even without knowing what the table says to do you know you aren't going to hit or double down and BS says to stand. So you know to split at +3.
Same reasoning for 12 v 4 at 0. BS says to stand here, but the index says to make another play. You can't split (or if you could you already should have) and you never double down on a 12. The only thing left is to hit.
As for insurance yes you take even money when the count is 3+. Even money is just when they call insurance by another name. To put it in math
Place half your wager. If they do have blackjack your original bet pushes and you get paid 2 to 1 on the insurance. So you end up with 2 * half your wager or even money.
If they don't have blackjack then you lose the insurance bet, but win 3:2 on blackjack. Which is the same as even money.
Quote: rhodyBobI am trying to incorporate index plays (Illustrious 18 and Fab 4) into Basic Strategy (BS), but I can't figure out, in all cases, how some of these plays impact on my action. I have the index chart from here, and many of the moves make sense intuitively.
For example, #1. It makes sense to take insurance when the count is +3 or greater, different from BS. Under BS it's easy to remember: never take insurance. This index play reflects the influence of a positive count when you know that insurance is a totally separate wager: is there a ten under that ace. What I am not sure about is whether to take even money for a blackjack when the count is >= +3. That's the practical effect of insurance when you have a blackjack. Does the "take insurance when count >= +3" apply in this case as well?
There are other instances where the variances from BS in the various indexes are not so clear, at least to me. Is there a chart that has two more columns next to the three that are listed, for example: "Order.....Play.....Index....." and then "Normally.....But instead do this"? These two added columns would remove any uncertainty, at least for me. There are so many different games out there, maybe the "Normally" column won't make sense (I play at Mohegan, at both S17 and H17 tables), but clarity as to "instead do this" would be nice.
As an extremely anal person, I have spent countless hours examining every chart I have ever seen on Basic Strategy. There are parts that make sense to me as the result of thinking about it, and for the parts that don't I accept as a matter of faith what I read here. A lot of the math that you folks present here makes my eyes roll back in my head, but it does convince me that you DO know what you are talking about. There are some index moves for which
I can't say for sure what the alternative is to BS. It would be nice if there was a chart that states clearly what to do, so that I can blindly obey.
That's what I like about this game. There are no emotions or feelings involved. Find the cell in the mental spreadsheet. Do what it says. I am ready to add another set of coulmns and rows in my head. Can you help?
You take insurance when the count is +3, and whenever the count exceeds the index. So you also take it at plus 4, 5, 6, 7, and infinity. Having a column that tells you what you normally would do makes no sense. These are deviations from basic strategy. If you don't know what you normally do, you should not be using the chart at all.
The way I learned to use deviation charts, while technically incorrect in reasoning to an insignificant degree, was to assume all hands are played at basic strategy at an index of 0. Say your index is -1. You continue making the deviation when the count is -2; you do not make the play above -1, only below. You do not make the deviation above -1, because you are supposed to play basic strategy at 0, and 0 is above -1. For insurance, the index is +3. You insure at +4. You do not insure below 3, only above. You do not exclusively make the play when the count is at a specific number. The correct play to make differs from basic strategy whenever the ratio of high cards to low cards meets or exceeds a certain threshold. The index number is the answer to the threshold question.
For most plays, it's fine to just remember basic strategy then the deviations on top of that because for the most part, if the count is 0, then play by basic strategy. However, there are some deviations that don't "make sense" in that way. For example, A8v6 (H17), the index is +1. This means you double A8v6 if the count (TC) is +1 or greater. If the count is less than +1, you stay. BUT, Basic Strategy says to double. So if you think "BS is double down, but if count is +1 or greater you double but if it's less than +1 you stay", that'll kinda throw you off.
Insure all hands at TC >= +3, and this means to take even-money on BJ's at TC >= +3.
This is perhaps a little advanced to where you are on the learning curve.....but here's a situation where you may want to insure at a TC less than +3. Say, the TC is +3 and you put out a big max bet and you're dealt a 20, and a bunch of face cards come on the table, and the dealer is showing an Ace up. You recalculate the TC and discover it's less than +3, so you know that insuring is -EV (a losing proposition). However, taking insurance is such a situation may not be the worst play (note: taking insurance is still a -EV play), but, if you want to decrease the variance in the play, you might want to go ahead and take insurance, knowing that your hand of 20 w/ lots of money on it is worthless if the dealer pulls a face from underneath. So, personally, I'd take insurance on that hand, even though it's -EV.
Quote: RSThe way I've incorporated indices* (not indexes) is by erasing BS from your head and just memorizing "hit > X, stay <= X". For example, 12v4. I don't think, "Basic Strategy is to stay, but if the count is negative, then I should hit". Instead, I think "Stay at 0+, hit if < 0". For 13v3, I don't think "Basic Strategy is stay, but if..." whatever. I just think "If count is less than -1, then hit. If it's -1 or greater, stay." If someone were to ask me basic strategy for 16vT (w/o surrender), I wouldn't really know what to say, other than, "hit if negative, stay if 0 or higher". And that play doesn't quite make sense, in a way, since basic strategy is a hit....but if the count is 0 or higher, you stay (which seems counter-intuitive since you'd think index play and basic strategy is the same if count = 0).
For most plays, it's fine to just remember basic strategy then the deviations on top of that because for the most part, if the count is 0, then play by basic strategy. However, there are some deviations that don't "make sense" in that way. For example, A8v6 (H17), the index is +1. This means you double A8v6 if the count (TC) is +1 or greater. If the count is less than +1, you stay. BUT, Basic Strategy says to double. So if you think "BS is double down, but if count is +1 or greater you double but if it's less than +1 you stay", that'll kinda throw you off.
Insure all hands at TC >= +3, and this means to take even-money on BJ's at TC >= +3.
This is perhaps a little advanced to where you are on the learning curve.....but here's a situation where you may want to insure at a TC less than +3. Say, the TC is +3 and you put out a big max bet and you're dealt a 20, and a bunch of face cards come on the table, and the dealer is showing an Ace up. You recalculate the TC and discover it's less than +3, so you know that insuring is -EV (a losing proposition). However, taking insurance is such a situation may not be the worst play (note: taking insurance is still a -EV play), but, if you want to decrease the variance in the play, you might want to go ahead and take insurance, knowing that your hand of 20 w/ lots of money on it is worthless if the dealer pulls a face from underneath. So, personally, I'd take insurance on that hand, even though it's -EV.
You seem to have a better grasp on the subject than you appeared to have in your first post. I'm not sure what your main concern is. To answer your last question first, you would not want to insure. There are some methods that don't strictly follow the index. One is variance control (or cover), one that I follow. For example, if you have a poor hand, you don't insure--I exclude hard 13-17. Another method is taking risk-averse insurance. Each hand has index between 2 and 4. 20 is technically the best hand to insure in this method, but if the count is below 2, you have no justification.
The reason why the index to stand on 16 v. 10 is 0, which differs from the basic strategy play to hit, is because basic strategy always assumes those are the only three cards dealt from the deck--your two cards and the dealers up card. You can't possibly have a positive count with a 16 v. 10, it's always negative. In an infinite deck situation, the count is meaningless, so I can't really tell you why it differs here--other than 0 is not the same thing as infinity.
Quote: RSThe way I've incorporated indices* (not indexes) is by erasing BS from your head and just memorizing "hit > X, stay <= X". For example, 12v4. I don't think, "Basic Strategy is to stay, but if the count is negative, then I should hit". Instead, I think "Stay at 0+, hit if < 0". For 13v3, I don't think "Basic Strategy is stay, but if..." whatever. I just think "If count is less than -1, then hit. If it's -1 or greater, stay." If someone were to ask me basic strategy for 16vT (w/o surrender), I wouldn't really know what to say, other than, "hit if negative, stay if 0 or higher". And that play doesn't quite make sense, in a way, since basic strategy is a hit....but if the count is 0 or higher, you stay (which seems counter-intuitive since you'd think index play and basic strategy is the same if count = 0).
For most plays, it's fine to just remember basic strategy then the deviations on top of that because for the most part, if the count is 0, then play by basic strategy. However, there are some deviations that don't "make sense" in that way. For example, A8v6 (H17), the index is +1. This means you double A8v6 if the count (TC) is +1 or greater. If the count is less than +1, you stay. BUT, Basic Strategy says to double. So if you think "BS is double down, but if count is +1 or greater you double but if it's less than +1 you stay", that'll kinda throw you off.
Insure all hands at TC >= +3, and this means to take even-money on BJ's at TC >= +3.
This is perhaps a little advanced to where you are on the learning curve.....but here's a situation where you may want to insure at a TC less than +3. Say, the TC is +3 and you put out a big max bet and you're dealt a 20, and a bunch of face cards come on the table, and the dealer is showing an Ace up. You recalculate the TC and discover it's less than +3, so you know that insuring is -EV (a losing proposition). However, taking insurance is such a situation may not be the worst play (note: taking insurance is still a -EV play), but, if you want to decrease the variance in the play, you might want to go ahead and take insurance, knowing that your hand of 20 w/ lots of money on it is worthless if the dealer pulls a face from underneath. So, personally, I'd take insurance on that hand, even though it's -EV.
Indexes is also a correct pluralization.
in·dex
noun \ˈin-ˌdeks\
plural in·dex·es or in·di·ces
From merriam-webster.
Many people insure only good hands when they choose to insure at all. That logic is flawed (unless playing for cover, which you shouldn't worry about if just starting out). Insurance is a side bet and to be regarded as completely independent of your hand. If you have a great hand and dealer has BJ, you keep your bet. If you have a shitty hand and dealer has BJ, you also still keep your bet. If you have 16 v A, and you insure, it's still correct to surrender if dealer doesn't have BJ. Think of insurance like one of those side-games casinos offer -- match the dealer, lucky ladies, etc. It is an independent game.
As far as index plays go, there are many, many indices. The most important ones (and, according to Wong, knowing only these will get you the vast majority of the gains of all index plays combined) are
- taking insurance / even money at >=3
- staying hard 16 vs 10 at >0
- surrender 14 vs 10 at >3
- staying hard 15 vs 10 at >4
The gains from other index plays do help, but they help very little in comparison and if you are just starting index plays, just stick with those four.
My problem is that I just need to complete a rows-and-columns chart that contains the index plays so that I can adapt it for my own mind.
Not that it matters in this context, but I have a cheat-sheet that I hold in my hand that is a much-abbreviated version of the BS chart from here. Why? I play at both S17 and H17 tables, and there are three differences in BS in the two games. I need the reminder that is inherent in having such a crutch to keep BS in "real memory". I figure being seen checking it can't hurt as cover, even though so far I don't think I have gotten anyone's attention.
The chart is not complete. It's just the obscure parts of BS, or the parts that vary between H17 and S17. For example, there are no references as to what to do against a dealer 5 or 6 - that's engraved in my mind because it's the favorite part of this whole business. It fits in the palm of my hand, it's of my own making, and it helps.
My intention is to put the more "difficult" parts of these indexes (or "indices", should you be a Merriamist instead of a Websterian), onto the back of this cheat sheet. Hence the need to be sure about what they are. So, just like the BS chart, there ought to be a chart like the one I described in my original post. For many of the index plays the "instead do this" column is obvious. Some so much so that once I confirm them, I can commit them to memory without having to add them to the back of my cheat-sheet. Based on the responses so far, "take insurance at +3 or greater" is one I won't need to write down. Easy-peezy.
What I'll do now is to create such a list, which I hope you all might validate for me, with the ones I am uncertain about left blank.
Can't do it right now. Going out to eat, and for the one or two of you that are from RI who can manage to read this far into this post, you will appreciate that I am going to Twin Oaks tonight for baked stuffed shrimp. If you're from "around here", you'll know what I am talking about.
Quote: arcticfunMathematically, "even money" and "insurance" are exactly the same bet, in the sense that their expected value is the same under given conditions. Snyder makes a nice comment about taking even money when you have a big bet down and the count is not quite 3. He says that even though it is not mathematically optimal to take even money, he will not judge you for doing so because other factors are at play -- how long have you been playing? how emotionally attached are you to that win? how much longer is your session? A big-bet win feels good, so sure - take even money. But if you catch yourself doing it every time, then you aren't emotionally prepared for blackjack.
Many people insure only good hands when they choose to insure at all. That logic is flawed (unless playing for cover, which you shouldn't worry about if just starting out). Insurance is a side bet and to be regarded as completely independent of your hand. If you have a great hand and dealer has BJ, you keep your bet. If you have a shitty hand and dealer has BJ, you also still keep your bet. If you have 16 v A, and you insure, it's still correct to surrender if dealer doesn't have BJ. Think of insurance like one of those side-games casinos offer -- match the dealer, lucky ladies, etc. It is an independent game.
As far as index plays go, there are many, many indices. The most important ones (and, according to Wong, knowing only these will get you the vast majority of the gains of all index plays combined) are
- taking insurance / even money at >=3
- staying hard 16 vs 10 at >0
- surrender 14 vs 10 at >3
- staying hard 15 vs 10 at >4
The gains from other index plays do help, but they help very little in comparison and if you are just starting index plays, just stick with those four.
Insuring poor hands increases variance. Although this should not necessarily deter someone from taking it because it does not change EV, it is a reason to justify not doing so, especially when it elicits cock-eyed looks from the table. Casinos set table limits for variance control, not to maximize EV Also, those are indeed the four most important index plays. However, they hardly account for the vast majority of the gain derived from strategy deviations. Probably just over half. If you examine it, the 16 v. 10 and 14 v. 10 are contrary indices; one has less value in a surrender game, and the other has none in a game without.
Rhodybob--Might I suggest your A8 v. 6 index is incorrect. I use 0-. I double at 0, but do not double at any negative running count. This may not be the exact index, but I have no doubt it is closer than +1.
Quote: arcticfunStanford Wong's book has a 1-page chart that contains all index plays for true counts between -10 and +10.
Which one? I have a copy of Professional Blackjack, which has 100 tables, not counting the appendixes. Maybe tables 9 and 10? Looks to me like there are more than 22 index variations. I'm hoping to incorporate a few of the fab 18, and all of the surrender variations, into what I can bring to the table. My guess is that asking everyone to pause for a few minutes while I thumb through a book entitled "Professional Blackjack" would not work.
Quote: Sonuvabish
Rhodybob--Might I suggest your A8 v. 6 index is incorrect. I use 0-. I double at 0, but do not double at any negative running count. This may not be the exact index, but I have no doubt it is closer than +1.
I'm not sure if I mentioned doubling a soft 19 against a 6, but in fact I WOULD do that in a six-deck, H17 game: https://wizardofodds.com/games/blackjack/strategy/4-decks/ I recall doing it only twice, lifetime, and in both cases I drew a 10 and the dealer busted. I was so overjoyed with those two wins that it's one of the obscure rules imbedded in my memory. No need to refer to the cheat sheet.
RS--Your indices are suspect in H17 shoe games. For S17, the index for A8 v. 6 is +1. I bet you have them mixed. Defeats the purpose to some extent.
Assume 6D, S17, DAS, DA2.
Order Play Index Normal Play Index Play
2 16v10 >= +0 Surrender ??????
3 15v10 >= +4 Surrender ??????
13 16v9 >= +5 Surrender ??????
As the count increases, and a 16 or 15 gets worse, what index play is more advantageous than surrender?
Order Play Index Normal Play Index Play
6 10v10 >= +4 Double ??????
Is the index play to split the 10s? Some places (Mohegan) don't allow split 10s. So double is as good as it gets?
Fab Surr. Play Index Normal Play Index Play
2 15v10 >= +0 Surrender ??????
Just like Indexes #2 #3 and #13 above. What can improve the outcome more than a surrender?
Assume 6D, S17, DAS, DA2.
Order Play Index Normal Play Index Play
2 16v10 >= +0 Surrender ??????
3 15v10 >= +4 Surrender ??????
13 16v9 >= +5 Surrender ??????
As the count increases, and a 16 or 15 gets worse, what index play is more advantageous than surrender?
Order Play Index Normal Play Index Play
6 10v10 >= +4 Double ??????
Is the index play to split the 10s? Some places (Mohegan) don't allow split 10s. So double is as good as it gets?
Fab Surr. Play Index Normal Play Index Play
2 15v10 >= +0 Surrender ??????
Just like Indexes #2 #3 and #13 above. What can improve the outcome more than a surrender?
(Tried to improve the formatting with a fixed-width font. Sorry for the double post if this doesn't work).
If you can't split tens then you just stay.
If the count is >=0 then you surrender. If it's not or you can't surrender then you play by the next index. Hit if the count <4. Stand otherwise.
Basic Strategy is not the same as the decision of a RC=0 index play. You can see this with the 16vT play, where basic strategy says to HIT 16vT, but the index for that is to hit if RC < 0, which means if the RC = 0, you would stand but if you were playing by basic strategy it'd be a hit.
For A8v6, basic strategy says to double, but if you know the TC is less than 1, you don't double. As the count gets higher and higher, the value of doubling A8v6 increases very quickly, while the value of doubling as the count gets lower and lower decreases at a smaller rate. Your +EV in high counts supercedes the -EV in negative counts on a play like A8v6 to the point where it is smarter to always double A8v6 rather than to always stay A8v6. Kind of like how the value of splitting tens vs 6 grows very quickly as the count increases, but a play like doubling 10vT increases at a slow rate as the count increases. (ie: Both indices are at +4. If the TC is +5, doubling 10vT is a small +EV change (from +4), but splitting TTv6 at a +5 has a large +EV change (from +4).) But that's a little off subject.
Quote: RSAccording to Norm's CVBJ, the index for H17 game, A8v6, is a double at +1.
Basic Strategy is not the same as the decision of a RC=0 index play. You can see this with the 16vT play, where basic strategy says to HIT 16vT, but the index for that is to hit if RC < 0, which means if the RC = 0, you would stand but if you were playing by basic strategy it'd be a hit.
For A8v6, basic strategy says to double, but if you know the TC is less than 1, you don't double. As the count gets higher and higher, the value of doubling A8v6 increases very quickly, while the value of doubling as the count gets lower and lower decreases at a smaller rate. Your +EV in high counts supercedes the -EV in negative counts on a play like A8v6 to the point where it is smarter to always double A8v6 rather than to always stay A8v6. Kind of like how the value of splitting tens vs 6 grows very quickly as the count increases, but a play like doubling 10vT increases at a slow rate as the count increases. (ie: Both indices are at +4. If the TC is +5, doubling 10vT is a small +EV change (from +4), but splitting TTv6 at a +5 has a large +EV change (from +4).) But that's a little off subject.
In basic strategy, 16 v. 10 is a hit, but the RC is never 0, it is always negative. The index and basic strategy do not disagree, except when the deck is infinite--where it's always better to follow basic strategy.
Your explanation for A8 v. 6 makes sense. But it differs from every other other index that I have seen. It may be correct making certain assumptions about your play style, and the game you are playing. It also could be rounding up the index to the nearest integer--it could be 0.51; I essentially use 0.2 since I stand on zero and double when positive, I might be closer. .I do not trust your index, and recommend against it as it goes against the grain.
Quote: rhodyBobAfter reviewing the index/surrender plays, there are five of them that I can't figure out. What am I missing?
Assume 6D, S17, DAS, DA2.
Order Play Index Normal Play Index Play
2 16v10 >= +0 Surrender ??????
3 15v10 >= +4 Surrender ??????
13 16v9 >= +5 Surrender ??????
As the count increases, and a 16 or 15 gets worse, what index play is more advantageous than surrender?
Order Play Index Normal Play Index Play
6 10v10 >= +4 Double ??????
Is the index play to split the 10s? Some places (Mohegan) don't allow split 10s. So double is as good as it gets?
Fab Surr. Play Index Normal Play Index Play
2 15v10 >= +0 Surrender ??????
Just like Indexes #2 #3 and #13 above. What can improve the outcome more than a surrender?
Your first three plays are for multi-card situations, or for a game without surrender.
The next play, the normal play is to hit. The index play is to double. A typo? This is not a 10,10 split. This is your 10 vs. the dealer's 10. The 6 is the order of importance for non-surrender index plays. These are all important indices to learn.
The last play, the inequality sign is reversed in the index. The sign should be < 0; there should not be an equal sign.
Whatever site u are using, use a different one.
The use of inequality signs is confusing, and troublesome. I made the assumption that the play had a certain count. "Having 10 v. 10 at a count that s greater than or equal to +4" is how I read all them.
Quote: SonuvabishAlso, those are indeed the four most important index plays. However, they hardly account for the vast majority of the gain derived from strategy deviations. Probably just over half. If you examine it, the 16 v. 10 and 14 v. 10 are contrary indices; one has less value in a surrender game, and the other has none in a game without.
Here is what I used to make the claim about the top 4 index plays providing the "vast majority" of index-play advantage:
http://www.bjrnet.com/faq21_12.htm
It's a good 2/3 of the total advantage offered by the I-18 and F-4.
Quote: arcticfunHere is what I used to make the claim about the top 4 index plays providing the "vast majority" of index-play advantage:
http://www.bjrnet.com/faq21_12.htm
It's a good 2/3 of the total advantage offered by the I-18 and F-4.
You are incorrect. We had been talking about total advantage. Remember, that these 22 indices account for 80% of the value of using all indices.
Assuming these numbers are correct and that surrender does not decrease the value of stand indices, your most important indices account for 47%. Yes, it's over 50% of the advantage offered by the "Illustrious 18". So what? No serious player would advocate learning anything less than 40-50 indices, and do not view the Illustrious 18 as anything more than a ranking system and learning guide. Some less serious players just learn this list. Would you advocate learning only 4 indices?
I advocate the mantra that, as in many other real-world situations, there comes a point where increased effort and complexity provide only decreasing returns. You get pretty far (halfway!) using 4 indices. You need an additional 18 to get to 80%. And an additional 30 to get to 99%. The amount of time to learn the first 4 indices: 1 hour. The following 40? much, much longer.
So yes, I advocate that learning the top 4 and using them correctly 100% of the time will bring you the best return. It would take half a lifetime of continuous play to experience measurable and statistically significant returns on index plays beyond a certain point. It seems reasonable that this "certain point" is when the additional returns are less than 0.01%. Why this number? there is a neat expose somewhere that compares playing craps on the light vs dark side, and how the 0.02% advantage of playing dark would take years to come through. Another justification is Monte Carlo simulations of biased coin-flips. Try it out -- how many flips do you need with a 50.5% vs a 50.52% head-biased coin to see statistically significant results?
Quote: SonuvabishNo serious player would advocate learning anything less than 40-50 indices, and do not view the Illustrious 18 as anything more than a ranking system and learning guide.
LOL
Quote: arcticfunGood point about the 80%.
I advocate the mantra that, as in many other real-world situations, there comes a point where increased effort and complexity provide only decreasing returns. You get pretty far (halfway!) using 4 indices. You need an additional 18 to get to 80%. And an additional 30 to get to 99%. The amount of time to learn the first 4 indices: 1 hour. The following 40? much, much longer.
So yes, I advocate that learning the top 4 and using them correctly 100% of the time will bring you the best return. It would take half a lifetime of continuous play to experience measurable and statistically significant returns on index plays beyond a certain point. It seems reasonable that this "certain point" is when the additional returns are less than 0.01%. Why this number? there is a neat expose somewhere that compares playing craps on the light vs dark side, and how the 0.02% advantage of playing dark would take years to come through. Another justification is Monte Carlo simulations of biased coin-flips. Try it out -- how many flips do you need with a 50.5% vs a 50.52% head-biased coin to see statistically significant results?
You make good points, but....
I used the Illustrious 18 as my start point. You're going to learn them, you should learn them in order of importance to maximize gain. But learning 4 indices doesn't take an hour. It takes 1 minute. There's only so much you can grasp without forgetting before actually putting them into practice. So maybe I'd implement 4 or 5 per session. Didn't take long before I was using over 50. Stopped when nothing new cropped up. When something weird did come up and I didn't know what to do, I made a guess, came home, and looked it up for future reference. No reason not to go farther than 4 indices unless you barely play. If you barely ever play, then I don't see why you'd go beyond insurance.
Some say it would take 10,000 lifetimes to get statistically significant results with anything regarding blackjack. I tend to disagree with this, and generally believe that it tends to take that long to get consistent, reliable, precise results. Otherwise, counting would be pointless, as would operating casino table games. When you play dice on the dark side, your odds are better. Your going to lose 2 cents less on the dollar. Over half a lifetime of continuous play, in other words a major amount of money that is statistically insignificant, that's a lot of money saved. I think you are disregarding variance, or misinterpreting it. Wouldn't you rather be on the better side of variance? Yes, you haven't played long enough to wipe it out (although, in dice, why would you want to?). But your results are hardly insignificant from a practical point of view.
I use many indices. I have surrendered 12 and 17 v. 10, stood on 16 v. 7, hit 17 v. A, doubled 8 v. 3. My count is complex. I use half counts. I neutralize cards, giving them dual-values. I neutralize some half counts. Everything, by itself, is insignificant. But cumulatively, assuming you use hi-lo and only 4 indices, my gain is probably over 33% assuming all other conditions are equal. That's huge. You can use stats to call it insignificant, and simulation experts may agree to an extent. But the likelihood I will win more money is that much greater. From a practical standpoint, it is hardly insignificant. In the end, it truly is better to be lucky than good. But that is not a good reason to forgo being good.
Quote: RSArctic is right. And you're squeezing out a few extra pennies by learning a thousand indices.
LOL
ok--just tryin to get my edge up to 15%
Quote: Sonuvabishok--just tryin to get my edge up to 15%
I was not talking about card counting in that post.
Quote: RSI was not talking about card counting in that post.
Did you mistake the OP for a hole-carder, who simultaneously does ace sequencing, since hole-carding doesn't give that large of an advantage--or did you just mean to hijack the thread? He didn't really seem like one, based on his limited understanding of the game. When you basically say "I am the best player in the universe" without qualification, it's going to be the go-to comment whenever you interject into a conversation to make unqualified contradictions. As for your unqualified contradiction made in this thread--squeezing extra pennies out--playing deviations make up approximately 1/3 of a typical counter's edge. Cutting out half of that is over 16%. 16% is more than pennies. Just so you know.
Quote: SonuvabishDid you mistake the OP for a hole-carder, who simultaneously does ace sequencing, since hole-carding doesn't give that large of an advantage--or did you just mean to hijack the thread? He didn't really seem like one, based on his limited understanding of the game. When you basically say "I am the best player in the universe" without qualification, it's going to be the go-to comment whenever you interject into a conversation to make unqualified contradictions. As for your unqualified contradiction made in this thread--squeezing extra pennies out--playing deviations make up approximately 1/3 of a typical counter's edge. Cutting out half of that is over 16%. 16% is more than pennies. Just so you know.
The post was in a different thread, where the OP asked what the max house edge is that you would play. The point of my post was that you should be looking for a game that offers something so you can get a bigger edge. Instead of playing a 75% pen game with a 0.3% HE, you should play a 85% pen game with a 0.5% edge...because the player's edge (via counting) is going to be greater in the 85% pen game. Don't look at house edge, but what you can get from the game. A game with 1% HE that can be hole-carded is worth a hell of a lot more than a game with a -1% HE (player's +1% edge). Hopefully that makes sense to you. Plus, the games with a tiny house edge tend to be watched a bit more closely than the awful game right beside it. Looking at that thread, it appears I did not hijack it, since no one else responded to my post except you.
One half of the indices does not make up one half of the edge. A huge majority of it comes from the illustrious 18. After that you see diminishing returns. In other words, knowing the illustrious 18 gives you a majority of your playing deviations edge. Learning 10 more indices increases that edge but by a small amount. Learning 10 more increases the edge yet again, but gives you a slightly greater edge.
As for the value of knowing indices...I just a ran a sim. $1m bankroll, spreading $100-$1,600 play-all on a 6D H17 DAS game with 5/6 pen heads up, and full indices. The results were $314/hour with 100 hands per hour.
Ran the exact same sim, where the only difference is the player used the illustrious 18 instead of full indices. $290/hour at 100 hands/hour. The gain from illustrious 18 to full indices is about $24/hour (on a 100-1600 spread), or ~8% increase in return.
I don't feel like setting up a sim for a player using 40-50 of the best indices. Since we know that with every new index learned you gain diminishing returns, your added value of learning the indices after the first 40 really does not mean all that much.
So, although you can gain more using full indices compared to the illustrious 18, I think it's slightly ridiculous to claim something like, "No pro would recommend learning any less than 40-50 indices" [not verbatim].
Instead of memorizing a bunch of indices, and maybe you're the kinda person that can sit down and memorize all of them rather easily without making mistakes in game, you're likely better off by spending your time learning new skills (count two tables at once, go scout for weak dealers [flashers or error-prone], dealers who give better penetration, or another game altogether like video poker or a carnival game).
If you can play for that long with that complicated of a count and that many indices, you are really wasting your time just counting. You could make a lot more money with the same mental exertion. Shuffle tracking is seriously not that hard, and if your memory is that good, sequencing will come naturally.
Quote: AxiomOfChoiceI could swear that I said this exact same thing in another thread (minus running the sims)
If you can play for that long with that complicated of a count and that many indices, you are really wasting your time just counting. You could make a lot more money with the same mental exertion. Shuffle tracking is seriously not that hard, and if your memory is that good, sequencing will come naturally.
Shuffle tracking is dangerous for anyone and impractical for me. My few experiments have not been successful. You may overestimate the mental exertion required for counting, as well as the gain from hacking a strong shuffle. Counting doesn't require a strong memory...learning indices is like riding a bike. True, most APs don't straight count--which is why my count is on steroids.
Quote: RSThe post was in a different thread, where the OP asked what the max house edge is that you would play. The point of my post was that you should be looking for a game that offers something so you can get a bigger edge. Instead of playing a 75% pen game with a 0.3% HE, you should play a 85% pen game with a 0.5% edge...because the player's edge (via counting) is going to be greater in the 85% pen game. Don't look at house edge, but what you can get from the game. A game with 1% HE that can be hole-carded is worth a hell of a lot more than a game with a -1% HE (player's +1% edge). Hopefully that makes sense to you. Plus, the games with a tiny house edge tend to be watched a bit more closely than the awful game right beside it. Looking at that thread, it appears I did not hijack it, since no one else responded to my post except you.
One half of the indices does not make up one half of the edge. A huge majority of it comes from the illustrious 18. After that you see diminishing returns. In other words, knowing the illustrious 18 gives you a majority of your playing deviations edge. Learning 10 more indices increases that edge but by a small amount. Learning 10 more increases the edge yet again, but gives you a slightly greater edge.
As for the value of knowing indices...I just a ran a sim. $1m bankroll, spreading $100-$1,600 play-all on a 6D H17 DAS game with 5/6 pen heads up, and full indices. The results were $314/hour with 100 hands per hour.
Ran the exact same sim, where the only difference is the player used the illustrious 18 instead of full indices. $290/hour at 100 hands/hour. The gain from illustrious 18 to full indices is about $24/hour (on a 100-1600 spread), or ~8% increase in return.
I don't feel like setting up a sim for a player using 40-50 of the best indices. Since we know that with every new index learned you gain diminishing returns, your added value of learning the indices after the first 40 really does not mean all that much.
So, although you can gain more using full indices compared to the illustrious 18, I think it's slightly ridiculous to claim something like, "No pro would recommend learning any less than 40-50 indices" [not verbatim].
Instead of memorizing a bunch of indices, and maybe you're the kinda person that can sit down and memorize all of them rather easily without making mistakes in game, you're likely better off by spending your time learning new skills (count two tables at once, go scout for weak dealers [flashers or error-prone], dealers who give better penetration, or another game altogether like video poker or a carnival game).
I didn't say half the indices make up half the edge. I said the top 4 indices make up half of the playing deviation edge, and that not learning more, you are giving up about 1/6 of your total edge.
Your sim demonstrates that what I said is essentially true. I estimated that 16% of your edge is giving up by not learning more, 10% of which was in the rest of the I18, and that you would lose 7% from the rest! I did your sim in my head, apparently.
I personally would be fine by telling a player learn the Catch 22 (those extra 4 on top of the Illustrious 18) plus the Fab 4, and you are fine. That's 26, and that should be about 95% of your attainable total edge. Changing counts would do as much as learning more indices at that point, I assume. I should have been more clear--I have never heard of a pro using less than 40-60 indices. I don't really know what they would or would not recommend. Schlesinger is a pro--he obviously recommended the I18 since he wrote them. My bad. But stopping at the top 4 indices? Highly inadvisable.
Your last part is the type of advice I'd hear as a newbie. It makes sense to me now, but as a newbie, I would have no idea why you are telling me this. I want to learn stuff about blackjack, so it is suggested I scout 3-card poker? Some things are impractical, others hold no interest, others are extremely difficult. As for me now, some of this stuff still falls in the same category, but also completely contradicts my style and goals.
I appreciate the reply, and apologize for my abruptness. Hopefully the OP has benefited.
