Fuzzy counting, my ass

[EasyRhino]

(note that this is not specifically a jab at sagefrog)

Was playing a doubledeck game with two other players also at the table. We're a few rounds in, and in one round a ton of small cards come out on the felt, the dealer draws to a ridiculous many-carded 20, and wipes out the table. He then says "okay, bump the bets up, the good cards are coming"

One of the players increases her bet, gets a good hand, and congratulates the dealer on his awesomeness. To which he responds that he can usually do a pretty good job of tracking the cards.

The only problem? We were still in a negative count when that hand started. The spate of low cards had merely shifted the count from very negative to sligtly negative.

I thought it illustrated the difference pretty well between keeping a count and merely "tracking the cards". Too bad I couldn't share it with anyone at the table.

 

[geneticfreak]

I've been guilty of using a fuzzy count now and then. I use KO and lets say the count on a 6 deck shoe goes from -16 to -8 in one round, I might bump my bet an extra unit or two depending how deep into the show we are. Even though it isn't at the key count yet, more times than not, a ton of big cards come out and produce a winning hand. Not mathematically correct, but could be used as a cover play.

 

[SecurityRisk]

(note that this is not specifically a jab at sagefrog)

Was playing a doubledeck game with two other players also at the table. We're a few rounds in, and in one round a ton of small cards come out on the felt, the dealer draws to a ridiculous many-carded 20, and wipes out the table. He then says "okay, bump the bets up, the good cards are coming"

One of the players increases her bet, gets a good hand, and congratulates the dealer on his awesomeness. To which he responds that he can usually do a pretty good job of tracking the cards.

The only problem? We were still in a negative count when that hand started. The spate of low cards had merely shifted the count from very negative to sligtly negative.

I thought it illustrated the difference pretty well between keeping a count and merely "tracking the cards". Too bad I couldn't share it with anyone at the table.

You probably already know this, but that dealer wasn't counting. He simply saw a lot of small cards, and hey, a lot of small cards, time to bump up your bet.

 

[moo321]

A fuzzy count could still affect EV. Not powerfully, but better than nothing.

 

[EasyRhino]

I use KO and lets say the count on a 6 deck shoe goes from -16 to -8 in one round, I might bump my bet an extra unit or two depending how deep into the show we are.

Sure, and with an unbalanced count like KO, there's a mathematical basis for something like this. The idea being that your "bet more" point is at a lower threshold early in the shoe, and a higher threshold later. There's a mess of discussion buried in old posts here.

 

[sagefr0g]

A fuzzy count could still affect EV. Not powerfully, but better than nothing.

exactly, the idea is on the order of actions described in Blackjack BluebookII ( ie. watching for babies comming out ect.)that are not conventional counting but trying to glean some of the advantage that counting yields.
the trick is to model the fuzzy approach as closely as possible to orthodox counting while avoiding the more tedious aspects of counting. thats what i've been working on.
the problem i'm having with it currently is with the betting levels. i believe i can accurately nail the first bet up from my minimun bet. problem is when it is likely a good time to raise the bet further (just having in mind the bet level one is at and not an accurate RC) then one doesn't know if the RC has raised enough to justify further raising the bet. it becomes a matter of a judgement call. so still working on that aspect.

 

[sagefr0g]

(note that this is not specifically a jab at sagefrog)
.

lol, thanks dude. i want my spreadsheet back :flame:
just joking

Was playing a doubledeck game with two other players also at the table. We're a few rounds in, and in one round a ton of small cards come out on the felt, the dealer draws to a ridiculous many-carded 20, and wipes out the table. He then says "okay, bump the bets up, the good cards are coming"

One of the players increases her bet, gets a good hand, and congratulates the dealer on his awesomeness. To which he responds that he can usually do a pretty good job of tracking the cards.

The only problem? We were still in a negative count when that hand started. The spate of low cards had merely shifted the count from very negative to sligtly negative.

I thought it illustrated the difference pretty well between keeping a count and merely "tracking the cards". Too bad I couldn't share it with anyone at the table.

yup, this can be a definate problem with respect to the fuzzy approach. truly you need to be sure that you are not in negative territory when raising that bet. and the fuzzy approach can leave you open to that danger if you aren't wary as hell of it.

 

[SilentBob420BMFJ]

exactly, the idea is on the order of actions described in Blackjack BluebookII ( ie. watching for babies comming out ect.)that are not conventional counting but trying to glean some of the advantage that counting yields.
the trick is to model the fuzzy approach as closely as possible to orthodox counting while avoiding the more tedious aspects of counting. thats what i've been working on.
the problem i'm having with it currently is with the betting levels. i believe i can accurately nail the first bet up from my minimun bet. problem is when it is likely a good time to raise the bet further (just having in mind the bet level one is at and not an accurate RC) then one doesn't know if the RC has raised enough to justify further raising the bet. it becomes a matter of a judgement call. so still working on that aspect.

you dont count cards sage? i thought u did

 

[Bojack1]

You know I'm not too keen on the whole fuzzy counting method. No offense to anyone using it but its just a lazy approach to trying to get something you want. Those paths usually lead to dead ends. Basically you still want the advantage, but don't want the hassle that goes along with it.

Look at at it this way. When we were all very young we learned how to read. We weren't born knowing how, it was taught to us. It takes a few years, but even young children can read with no problem with no real concious thought on how. Its the same with counting cards. Most people start to learn but never get to the point where its practically no longer a concious thought process. Maybe its because its not a necessity like reading is, or maybe because we are no longer children being forced to do whats expected of us. As adults its up to us now to make the decisions on what we want. Counting can be become as easy as reading a stop sign, and when it does fuzzy counting will seem quite a silly and even harder way to play. Not to mention there really is more luck than advantage to the process.

 

[sagefr0g]

you dont count cards sage? i thought u did

did is correct. for about two years. but now, well i'm working on a new approach and don't play so much as i used to. probably wont get to play blackjack reguarly for i dunno anywhere from two weeks to maybe a year depending on current personal business. so i only play every so often now. but anyway i was never much of a hot shot at the game just a lot of luck on top of playing with an advantage for a couple of years. i just used hi/lo and a hacked together betting scheme based on advice of those who obviously know their stuff, my understanding of ROR, Kelly and the use of some simulations ect. ..
but in the mean time i'm trying and working on an approach that is based on card counting and advantage play. an approach which is for now very much 'up in the air' or a work in progress that might never turn out to be a tenable approach but so far is fun for me. and i'm just calling it the 'fuzzy count' , kind of a take off on fuzzy logic so to speak but really i'm not even knowledgeable as to what fuzzy logic is. but i think fuzzy logic is a lot like what we all do when it comes to intelligent guessing, application of estimation skills, intuition, analogies, wisdom (maybe), human thought, judgement, stuff like that and then trying to base those skill sets on card counting with out actually counting cards.... lol. what the fuzzy count for me so far has evolved into is really concentrating on what cards have come out with respect to are they big or little cards and then relating that to how deep into the deck one might be at any given time so as to estimate the TC. then i'm trying to hold this all together with what bet level i'm currently at as a result of my judgement of the current 'value' of the game to me bet level wise. the stumbling block with all this for now has been that the decision as to whether to raise or lower the bet is dependent on what constitutes some very fuzzy information realtive to orthodox counting.
but i think i can get away with this approach for now since i'm not playing very often and for the most part the game is fairly even if one plays perfect basic strategy. in other words i'm hoping that with a little luck i might just come out with an advantage and have more fun in the process. orthodox card counting became for me a bit more of a grind than i cared to deal with.
for more see:
http://www.blackjackinfo.com/bb/showpost.php?p=54250&postcount=1

 

[sagefr0g]

You know I'm not too keen on the whole fuzzy counting method. No offense to anyone using it but its just a lazy approach to trying to get something you want. Those paths usually lead to dead ends. Basically you still want the advantage, but don't want the hassle that goes along with it.

Look at at it this way. When we were all very young we learned how to read. We weren't born knowing how, it was taught to us. It takes a few years, but even young children can read with no problem with no real concious thought on how. Its the same with counting cards. Most people start to learn but never get to the point where its practically no longer a concious thought process. Maybe its because its not a necessity like reading is, or maybe because we are no longer children being forced to do whats expected of us. As adults its up to us now to make the decisions on what we want. Counting can be become as easy as reading a stop sign, and when it does fuzzy counting will seem quite a silly and even harder way to play. Not to mention there really is more luck than advantage to the process.

this is all very true sir for the most part. i like your analogy of reading, same goes for typing, speaking, language comprehension and maybe even bike riding. with regard to math it's been my experience that a lazy approach to trying to get something you want a path which usually leads to dead ends. where basically you still want the advantage, but don't want the hassle that goes along with it can be a very key approach to a rewarding successful way to approach understanding and the application of mathematics. just as how physics ideas are applied to the problem of work, where in one uses say a lever to lift a heavy object or a pulley, well it's the idea of not straining so hard to get what you want. this type of approach can likewise be applied to mathematical problems. as an example consider how in mathematics one can take a formula for which one really hasn't a clue as to how it was derived or any real manner of understanding and yet still apply it successfully to a mathematical problem that some mathematical genius spent years of solving in the past.
but i think your right that counting cards can become second nature. it's like kids and well some grownups that have to count beans on their fingers while others are able to engage a more mental process. lol
but something i'm intrigued with about the fuzzy count aspect has to do with an experience i had way back in elementary school. it was in my first 'official' math class. the teacher was having us work out some simple math problems. and what we were supposed to be doing was writing it out on paper and showing how we got the answer to the problem. but what i was doing was just writing the answer on the paper. i really had no idea exactly how i came up with the answers but it was some almost subconcious mental gymnasitics that was going on and i was getting correct answers. well the teacher, rightly so i should imagine (bless her soul) came around to me and explains that she wants the steps taken to solve the problem written down on the paper. lol... but even though she was right about all this i've always sensed that i had to bury something rather magical and wonderful in the process even though i did learn to add, subtract, multiply and divide in an orthodox manner in the process.
just some thoughts on all this Bojack, but i truly respect your approach and i think i understand your reasons for it as after all you are in the professionl league.

 

[SilentBob420BMFJ]

did is correct. for about two years. but now, well i'm working on a new approach and don't play so much as i used to. probably wont get to play blackjack reguarly for i dunno anywhere from two weeks to maybe a year depending on current personal business. so i only play every so often now. but anyway i was never much of a hot shot at the game just a lot of luck on top of playing with an advantage for a couple of years. i just used hi/lo and a hacked together betting scheme based on advice of those who obviously know their stuff, my understanding of ROR, Kelly and the use of some simulations ect. ..
but in the mean time i'm trying and working on an approach that is based on card counting and advantage play. an approach which is for now very much 'up in the air' or a work in progress that might never turn out to be a tenable approach but so far is fun for me. and i'm just calling it the 'fuzzy count' , kind of a take off on fuzzy logic so to speak but really i'm not even knowledgeable as to what fuzzy logic is. but i think fuzzy logic is a lot like what we all do when it comes to intelligent guessing, application of estimation skills, intuition, analogies, wisdom (maybe), human thought, judgement, stuff like that and then trying to base those skill sets on card counting with out actually counting cards.... lol. what the fuzzy count for me so far has evolved into is really concentrating on what cards have come out with respect to are they big or little cards and then relating that to how deep into the deck one might be at any given time so as to estimate the TC. then i'm trying to hold this all together with what bet level i'm currently at as a result of my judgement of the current 'value' of the game to me bet level wise. the stumbling block with all this for now has been that the decision as to whether to raise or lower the bet is dependent on what constitutes some very fuzzy information realtive to orthodox counting.
but i think i can get away with this approach for now since i'm not playing very often and for the most part the game is fairly even if one plays perfect basic strategy. in other words i'm hoping that with a little luck i might just come out with an advantage and have more fun in the process. orthodox card counting became for me a bit more of a grind than i cared to deal with.
for more see:
http://www.blackjackinfo.com/bb/showpost.php?p=54250&postcount=1

"i just used hi/lo and a hacked together betting scheme based on advice of those who obviously know their stuff, my understanding of ROR, Kelly and the use of some simulations ect.." that sounds to me like you knew what you were doing, but then why did you give that up? i believe people who say, "ya i used to count but blah blah i dont anymore", could never really count well enuf to begin with, because if you could, why wouldnt you take a few hours to get back into it? why would you take a step backwards? that is similar to, "yes i know basic strategy, and i used to play by it exactly, but now i use my instincts on some of my hands"..

this is all very true sir for the most part. i like your analogy of reading, same goes for typing, speaking, language comprehension and maybe even bike riding. with regard to math it's been my experience that a lazy approach to trying to get something you want a path which usually leads to dead ends. where basically you still want the advantage, but don't want the hassle that goes along with it can be a very key approach to a rewarding successful way to approach understanding and the application of mathematics. just as how physics ideas are applied to the problem of work, where in one uses say a lever to lift a heavy object or a pulley, well it's the idea of not straining so hard to get what you want. this type of approach can likewise be applied to mathematical problems. as an example consider how in mathematics one can take a formula for which one really hasn't a clue as to how it was derived or any real manner of understanding and yet still apply it successfully to a mathematical problem that some mathematical genius spent years of solving in the past.
but i think your right that counting cards can become second nature. it's like kids and well some grownups that have to count beans on their fingers while others are able to engage a more mental process. lol
but something i'm intrigued with about the fuzzy count aspect has to do with an experience i had way back in elementary school. it was in my first 'official' math class. the teacher was having us work out some simple math problems. and what we were supposed to be doing was writing it out on paper and showing how we got the answer to the problem. but what i was doing was just writing the answer on the paper. i really had no idea exactly how i came up with the answers but it was some almost subconcious mental gymnasitics that was going on and i was getting correct answers. well the teacher, rightly so i should imagine (bless her soul) came around to me and explains that she wants the steps taken to solve the problem written down on the paper. lol... but even though she was right about all this i've always sensed that i had to bury something rather magical and wonderful in the process even though i did learn to add, subtract, multiply and divide in an orthodox manner in the process.
just some thoughts on all this Bojack, but i truly respect your approach and i think i understand your reasons for it as after all you are in the professionl league.

i did that in school also, i hated showing my work.. i would be like, "you want me to show you that i divided both sides by x? but thats obvious".. or in english class they would say "rewrite the question into your answer"

 

[sagefr0g]

..... why wouldnt you take a few hours to get back into it? why would you take a step backwards? that is similar to, "yes i know basic strategy, and i used to play by it exactly, but now i use my instincts on some of my hands"..

i found the effort extended for the reward realized to not be worth while. it was especially discouraging for me when after an extended period of intensive counting to finally realize an advantageous count and then experience some of the worst negative fluctuation one could expect. while i realize that it is true that as long as you continue to play with an advantage that over time the negative and positive fluctuation should even out and one should be left with the proceeds of the advantage, none the less i guess i'm just the type that needs to see a more immeadiate reward for ones efforts.
the fuzzy count is relatively effortless for me, natural and comfortable, even fun as opposed to what for me became relatively stressful effort, not fun too me, ie. orthodox counting. the advantage of orthodox counting is greater i'm sure. the advantage of fuzzy counting is questionable but the potential is there. so i'm willing to try it. win some money or lose some money (of course hopefully winning some money) i'll now at least not experience the dissapointment of those negative fluctuations as measured against the effort i have extended because well there truly is very little effort at all.

i did that in school also, i hated showing my work.. i would be like, "you want me to show you that i divided both sides by x? but thats obvious".. or in english class they would say "rewrite the question into your answer"

there you go. the point being is it possible to cut through all the bull sh!t and still get it right? can one see a situation for what it is and discern the correct action to take with out going through all the rituals?

 

[eps6724]

there you go. the point being is it possible to cut through all the bull sh!t and still get it right? can one see a situation for what it is and discern the correct action to take with out going through all the rituals?

I think this would fall under the category of 'different people learn in different ways'. If I tried "fuzzy math" all I would get is "fuzzy answers".

BUT I am not you, and if you say you can do it than I accept that. I use 'fuzzy counting" with respect to lots of babies only as to thinking that I might gain a one-shot short advantage and maybe throw out another chip, but keep counting and wait for the longer run that gives.

I would LOVE to be able to "fuzzy think" (tho I feel that it speaks more to your subconscious memory and how your personal mind works) but counting just makes more sense to my way of thinking.

 

[beat320]

I would LOVE to be able to "fuzzy think" (tho I feel that it speaks more to your subconscious memory and how your personal mind works) but counting just makes more sense to my way of thinking.

haha me too, the only problem I have with the conventional "fuzzy counting" is the whole "oh man! did you see all of those 2-7's come out last hand? there were hardly any high cards! I think I'll raise my bet by 10 units" I really don't think that is what sage is describing though.

 

[sagefr0g]

I think this would fall under the category of 'different people learn in different ways'. If I tried "fuzzy math" all I would get is "fuzzy answers".

right, part of what i'm trying to do is really just estimation instead of trying to attain the accuracy that orthodox counting can yield. just as in the case where an accurate RC becomes tainted by the unavoidable estimation involved in obtaining a TC (since estimation of the number of decks remaining to be dealt is required in the calculation). i'm estimating both, the RC and the number of decks left to be dealt. then i'm trying to use my mind asking my self after each round is this situation as i percieve it justification enough to raise the bet and by how much.

BUT I am not you, and if you say you can do it than I accept that. I use 'fuzzy counting" with respect to lots of babies only as to thinking that I might gain a one-shot short advantage and maybe throw out another chip, but keep counting and wait for the longer run that gives.

pretty much that's what i'm doing but i'm trying to do that in a cumlative way over a number of rounds.

I would LOVE to be able to "fuzzy think" (tho I feel that it speaks more to your subconscious memory and how your personal mind works) but counting just makes more sense to my way of thinking.

counting is more accurate and requires more work. fuzzy counting is less accurate and requires less work.
imagine a robot that can shoot 100% accurate free throws in basketball. then compare that to what ever skill level a given basketball player can achieve. so by analogy let the counter be the robot and the fuzzy counter some skill level basketball player.
thats what in essence i'm trying to achieve only on the felt rather the court lol. and of course thats all clouded by negative and positive fluctuation and errors in estimation.

 

[SilentBob420BMFJ]

i found the effort extended for the reward realized to not be worth while. it was especially discouraging for me when after an extended period of intensive counting to finally realize an advantageous count and then experience some of the worst negative fluctuation one could expect. while i realize that it is true that as long as you continue to play with an advantage that over time the negative and positive fluctuation should even out and one should be left with the proceeds of the advantage, none the less i guess i'm just the type that needs to see a more immeadiate reward for ones efforts.
the fuzzy count is relatively effortless for me, natural and comfortable, even fun as opposed to what for me became relatively stressful effort, not fun too me, ie. orthodox counting. the advantage of orthodox counting is greater i'm sure. the advantage of fuzzy counting is questionable but the potential is there. so i'm willing to try it. win some money or lose some money (of course hopefully winning some money) i'll now at least not experience the dissapointment of those negative fluctuations as measured against the effort i have extended because well there truly is very little effort at all.

there you go. the point being is it possible to cut through all the bull sh!t and still get it right? can one see a situation for what it is and discern the correct action to take with out going through all the rituals?

the math class example has nothing to do with counting tho, because with the math, i actually was doing all the work in my head, whereas you are not doing any work in your head, your just fuzzy counting, which wont really work in 6 decks, and also, selective memory will take over.. if you are betting more only because you think you have an advantage, then this is a very bad way to play, but if you would have bet that much anyways, even if you were flat betting, then i guess its fine.. for instance, i would never bet more than the table minimum if i wasnt counting.. actually, i still only bet the table minimum! thats cuz i dont have the bankroll to spread, but i think im going to start doing $10-15 soon

 

[aslan]

I remember one time recently that I was hitting so well that I lost the count in all the excitement of winning big. I tried to remember a picture of the last round that was swooped up before I had time to refocus from winning to counting. I used my impression of the balance of high and low cards to increase or decrease the count and just continued this for a couple of more rounds. It all worked out. It was fuzzy counting for sure. Now I'm back to pinpoint counting; I don't trust my fuzzy counting yet, but maybe I should practice doing it in a simulation where I can go back and see what the real count was. Sounds like work tho, and KO is really not hard to do anyway.

 

[sagefr0g]

the math class example has nothing to do with counting tho, because with the math, i actually was doing all the work in my head, whereas you are not doing any work in your head, your just fuzzy counting, which wont really work in 6 decks, and also, selective memory will take over.. if you are betting more only because you think you have an advantage, then this is a very bad way to play, but if you would have bet that much anyways, even if you were flat betting, then i guess its fine.. for instance, i would never bet more than the table minimum if i wasnt counting.. actually, i still only bet the table minimum! thats cuz i dont have the bankroll to spread, but i think im going to start doing $10-15 soon

ideally, the game plan is to direct ones attention to the cards that hit the felt with the ambition of realizing the value of the game betwise. this does require work in your head but is differant than doing all the steps in your head as one does when counting the RC and calculating the TC. your spot on about selective memory holding the potential for disaster. additionally when playing this way because it is so simple there creeps in a tendancy for thinking that you should make bet decisions for reasons other than the game plan outlined above in the first sentance. again i'd agree that could be a recipe for disaster. the point being is that fuzzy counting is so simple that one needs to take caution that one sticks to it and doesn't stray from that simplicity in all the nebulous acts of gambling superstition that are prettty much summed up by the idea of the gambler's fallacy.
i'm pretty much forced to play at mostly full six deck & eight deck tables (if i want to play at all) since thats how it is at the joints i frequent. not the best conditions for orthodox counting. a lot of wonging play is difficult because i'm there so frequently but i can get away with some wonging. but anyway for me trying to use orthodox counting under such conditions became very tedious. you get slow dealers, slow table mates and such games creep along in general anyway. under such conditions one can easily experience multiple sessions where in one never see's a really decent count. that constitutes (for me) a considerable amount of work if your counting cards that just doesn't pay off. so the fuzzy count approach for me relieves the tedium while allowing for a decent potential to take some risk in a positive advantage situation. one still does have to keep ones betting levels within ones tolerance level for ROR.
for me it's a work in progress. i'm still trying to find a way to better judge the actuall RC with out actually maintaining the running count. for now i'm just using mental concentration on the cards as they come out and a rather poor memory. sounds silly i know, but whats the harm in trying to find a way. it would be nice if it were possible.

 

[sagefr0g]

I remember one time recently that I was hitting so well that I lost the count in all the excitement of winning big. I tried to remember a picture of the last round that was swooped up before I had time to refocus from winning to counting.

lol, yep i used to experience the same thing.

I used my impression of the balance of high and low cards to increase or decrease the count and just continued this for a couple of more rounds. It all worked out. It was fuzzy counting for sure. Now I'm back to pinpoint counting; I don't trust my fuzzy counting yet, but maybe I should practice doing it in a simulation where I can go back and see what the real count was. Sounds like work tho, and KO is really not hard to do anyway.

i practice it in cvbj..... where you can look and see what the actual TC is at anytime durring play.