Shuffle Tracking Question for the experts...

#1
Hello everyone! I'm back again pleading for help!!!

I am trying to get to grips with shuffle tracking, and decided it was much easier to "cut bad cards out of play" than to "cut good cards into play"..

I will try to explain what I want to do, and wondered if someone could just confirm that my thinking and strategy is correct..

At the moment I am playing a 6 deck shoe. The casino knows I am a counter and they are cutting the shoe in half (50% pen). Lets say I am counting the cards on a new shoe, and after the first deck has been dealt the RC is +5. Obviously during the shuffle these cards will get riffled with another random deck. My question is, if after the shuffle I cut the shoe in the such a way as to move the original first deck with a RC of +5 (which is now randomly shuffled with another deck) to somewhere behind the cut card, is it safe to start my RC of the new shoe at +5? Am I right in saying that if the low card slug of 52 cards plus the 52 random cards that were shuffled with it are cut SOMEWHERE behind the cut card, then it really does not have to be a precise cut as long as the cards are somewhere in the first three decks of the shoe?

I seems to me, if this were correct, that in the situation of a casino half-cutting a 6 deck shoe (50% penetration), the shuffle tracker does not need need to be so accurate when cutting low cards out of play, as long as they are placed behind the cut card and they will not be dealt in the game...

One other question... If I know for sure that with a particular shuffle routine I can cut the first deck played behind the cut card, is it correct to remember the PEAK count (the highest count) of the first deck played, then cut this deck to the back of the shoe after the shuffle (obviously mixed throughly with other random cards), and to start the RC at the PEAK (the highest RC in the first deck dealt of the shoe)...

Sorry if I explained things a bit badly... it is not easy to put some of my thoughts on paper!!

Thanks in advance guys,,

Spin.
 

iCountNTrack

Well-Known Member
#2
sp1n-d1zzy said:
Hello everyone! I'm back again pleading for help!!!

I am trying to get to grips with shuffle tracking, and decided it was much easier to "cut bad cards out of play" than to "cut good cards into play"..

I will try to explain what I want to do, and wondered if someone could just confirm that my thinking and strategy is correct..

At the moment I am playing a 6 deck shoe. The casino knows I am a counter and they are cutting the shoe in half (50% pen). Lets say I am counting the cards on a new shoe, and after the first deck has been dealt the RC is +5. Obviously during the shuffle these cards will get riffled with another random deck. My question is, if after the shuffle I cut the shoe in the such a way as to move the original first deck with a RC of +5 (which is now randomly shuffled with another deck) to somewhere behind the cut card, is it safe to start my RC of the new shoe at +5? Am I right in saying that if the low card slug of 52 cards plus the 52 random cards that were shuffled with it are cut SOMEWHERE behind the cut card, then it really does not have to be a precise cut as long as the cards are somewhere in the first three decks of the shoe?

I seems to me, if this were correct, that in the situation of a casino half-cutting a 6 deck shoe (50% penetration), the shuffle tracker does not need need to be so accurate when cutting low cards out of play, as long as they are placed behind the cut card and they will not be dealt in the game...

One other question... If I know for sure that with a particular shuffle routine I can cut the first deck played behind the cut card, is it correct to remember the PEAK count (the highest count) of the first deck played, then cut this deck to the back of the shoe after the shuffle (obviously mixed throughly with other random cards), and to start the RC at the PEAK (the highest RC in the first deck dealt of the shoe)...

Sorry if I explained things a bit badly... it is not easy to put some of my thoughts on paper!!

Thanks in advance guys,,

Spin.
Cutting a poor slug out of play is basically like starting a fresh shoe and counting that slug and the dealer placing that slug in discard tray. The running count will be the count of the slug you tracked and cut out of play, and your TC divisor is shoe size-tracked slug size.
However i recommend you also track rich slugs and cut them into play, shuffles are still pretty simple even in 2011 in the dear old motherland ;)
 

moo321

Well-Known Member
#3
Why are you playing someplace where they know you are a counter and you will be watched? Do you want to burn this shuffle out too? Why don't you let a teammate play the shuffle?
 

RJT

Well-Known Member
#4
Before i start i'll warn you that my opinion is likely to differ from that of most of the other posters here with regards to shuffle tracking in that i think the best advice you can be given about tracking high cards is "don't". The degree of accuracy that is required when tracking packets of high cards - or more specifically short packets - is far higher than any part-time player is going to achieve and errors are far too costly. Any part time player attempting this - and many of the pros too - risk doing far more damage to their bankroll than the potential gain would justify.

As to cutting low cards out of play, it's a weak technique. Ultimately it's not strong enough in itself to replace counting, only augment it and it doesn't have nearly the degree of built in cover that high card packets do as your play is still roughly corrolated with the count.

In the situation you describe, yes your accuracy doesn't need to be nearly as high as high cards would require. As long as the full 2 decks you're 1 deck's been mixed through ends up behind the cut card, you're ok. As to peak count - no you cannot simply take the highest point that the count reaches during your packet, it must be the packet for the full deck you're tracking.

Moo makes a very good point however - if the casino have already ID'd you as a counter, this strategy won't work long and is more likely to burn out a playable game for someone else than extend your lifespan any great deal. I get the impression that there aren't many casinos around you and that now you've been pegged in your home casino you're desperately trying to find a way to continue playing. My advice would be travel to other casinos or accept that you're done.

RJT.

RJT.
 
#5
Thanks for your comments everyone!

RJT, can you explain why I must count a complete deck? I was getting good results just counting the first round of cards. The reason is they often plug very low. Is it ok just to count the first round as long as those cards are cut to the back?

Also, why would it be wrong to start my RC at the peak, if I know for sure those cards are cut to the back? Surely the packet size is irrelevant. What's important is that the cards are not in play?
 

RJT

Well-Known Member
#6
sp1n-d1zzy said:
Thanks for your comments everyone!

RJT, can you explain why I must count a complete deck? I was getting good results just counting the first round of cards. The reason is they often plug very low. Is it ok just to count the first round as long as those cards are cut to the back?

Also, why would it be wrong to start my RC at the peak, if I know for sure those cards are cut to the back? Surely the packet size is irrelevant. What's important is that the cards are not in play?
If the RC rises to +12 in the first half deck but then drops back to +4 in the next half deck and the whole first deck ends up being cut out of play, then you've only cut out 4 extra low cards. The peak value of +12 has been offset by the extra high cards that came out in the second half deck.

If you know that the whole first deck is going to be cut out of play and you only use the count from the first round of play you're making a weak system even weaker by discarding relevant information.

RJT.
 

Sucker

Well-Known Member
#7
sp1n-d1zzy said:
I am trying to get to grips with shuffle tracking, and decided it was much easier to "cut bad cards out of play" than to "cut good cards into play"..
:confused: If you have the skill to cut bad cards out of play, then you have the skill to cut good cards into play. You're now 99% of the way there - why be too lazy to do the job perfectly?

sp1n-d1zzy said:
The casino knows I am a counter and they are cutting the shoe in half (50% pen).
:eek: This is WHY the casino knows you're a counter. Cutting good cards INTO play is the whole BEAUTY of shuffle tracking in the FIRST place! When you're making all of your max bets right off the top of the deck, it's going to take one helluva smart pit boss to EVER figure THAT out!
 

RJT

Well-Known Member
#8
Sucker said:
:confused: If you have the skill to cut bad cards out of play, then you have the skill to cut good cards into play. You're now 99% of the way there - why be too lazy to do the job perfectly?
I'm sorry Sucker, but that's simply not true. Errors cutting bad cards out of play are no where near as costly as those cutting good cards in. When cutting low cards out - especially in this circumstance where so much of the shoe ends up behind the cut card - all you have to know at the end is that the low cards ended somewhere behind the cut. With high cards you need to know the precise location of your money cards so you know when to place the big bets, otherwise you end up placing big bets into garbage massively increasing your variance.

RJT.
 
#9
sp1n-d1zzy said:
Hello everyone! I'm back again pleading for help!!!

I am trying to get to grips with shuffle tracking, and decided it was much easier to "cut bad cards out of play" than to "cut good cards into play"..

I will try to explain what I want to do, and wondered if someone could just confirm that my thinking and strategy is correct..

At the moment I am playing a 6 deck shoe. The casino knows I am a counter and they are cutting the shoe in half (50% pen). Lets say I am counting the cards on a new shoe, and after the first deck has been dealt the RC is +5. Obviously during the shuffle these cards will get riffled with another random deck. My question is, if after the shuffle I cut the shoe in the such a way as to move the original first deck with a RC of +5 (which is now randomly shuffled with another deck) to somewhere behind the cut card, is it safe to start my RC of the new shoe at +5? Am I right in saying that if the low card slug of 52 cards plus the 52 random cards that were shuffled with it are cut SOMEWHERE behind the cut card, then it really does not have to be a precise cut as long as the cards are somewhere in the first three decks of the shoe?

I seems to me, if this were correct, that in the situation of a casino half-cutting a 6 deck shoe (50% penetration), the shuffle tracker does not need need to be so accurate when cutting low cards out of play, as long as they are placed behind the cut card and they will not be dealt in the game...

One other question... If I know for sure that with a particular shuffle routine I can cut the first deck played behind the cut card, is it correct to remember the PEAK count (the highest count) of the first deck played, then cut this deck to the back of the shoe after the shuffle (obviously mixed throughly with other random cards), and to start the RC at the PEAK (the highest RC in the first deck dealt of the shoe)...

Sorry if I explained things a bit badly... it is not easy to put some of my thoughts on paper!!

Thanks in advance guys,,

Spin.
If you are getting 50% pen, leave. Unless it is an extremely weak shuffle you are not getting enough information from the mother shoe to effectively shuffle track.
 

Sucker

Well-Known Member
#10
RJT said:
Errors cutting bad cards out of play are no where near as costly as those cutting good cards in.
This is a very common misconception among shuffle trackers. The fact is; errors cutting bad cards out of play are EXACTLY as costly as those cutting good cards in. The added benefit to cutting the good cards into play is the fact that your max bets will be off the top, whereby massively reducing the heat level and the waiting time. And I must also mention the fact that the player who cuts the game to the front will also be able to play all of the NMSE games.
 

Sucker

Well-Known Member
#11
Automatic Monkey said:
If you are getting 50% pen, leave. Unless it is an extremely weak shuffle you are not getting enough information from the mother shoe to effectively shuffle track.
Absolutely! For 50% pen, the shuffle MUST be weak enough so that you can match all of the slugs to each other, rather than just tracking one or two slugs; and THEN you MUST be able to cut your master slug to the front.
 

RJT

Well-Known Member
#12
Sucker said:
This is a very common misconception among shuffle trackers. The fact is; errors cutting bad cards out of play are EXACTLY as costly as those cutting good cards in. The added benefit to cutting the good cards into play is the fact that your max bets will be off the top, whereby massively reducing the heat level and the waiting time. And I must also mention the fact that the player who cuts the game to the front will also be able to play all of the NMSE games.
Sucker you're wrong.

If i have a 1 deck packet with 12 extra high cards, when cutting to it if i cut one card out i have a 12/52 (3/13) chance of it being one of my extra high cards. If the opposite is true and i have 12 extra low cards in a deck, what i really have is a 5 deck playzone with 12 extra high cards. So if i mistakenly cut one card out of my playzone i have a 12/260 (3/65) chance that it's one of my extra high cards. Differentiating between low and high cards is valueless, however the difference between big and small packets - and correspondingly knowing in which direction the errors are going to cost you more - is crucial.

Cutting low cards out is far more tollerant to errors because - as long as you cut in the right direction, ie cutting cards into your manovered packet rather than out - you have far less chance of removing the extra high cards you are hoping to play through.

Alongside this small packets cause you to bet more. With a one deck packet with 12 extra high cards, assuming a house edge of 0.5% and 0.5% advantage gain per TC point, you've got an edge of 5.5%. Assuming a unit of $100 and using a TC-1 system (this is all just to make a point, far from optimal) i'd bet $1100/hand. Assuming 9 rounds/deck i wager a total of $9900. The same 12 high cards spread over 4 decks (or cutting 12 low cards out of play give me an average advantage of 1% and a bet of $200/hand. This causes me to wager a total of $7200. Despite the same number of cards being cut in or cut out, you end up wagering more when playing shorter packet.

On top of all of that, assuming that you always cut a 1/4 deck off in the better direction (13 cards ahead of your high card deck or 13 cards behind the end of your low card deck), when playing a high card packet you end up with big bets out on the first round that is actually being placed into a disadvantage. If playing a low card packet, the high probability is that more than the deck will end behind the cut card anyway, meaning you suffer no penalty.

While academically you're correct in your assertation that there is no difference, practically you could not be further from the truth. One of the most widely overlooked aspects of tracking by those who write about it or attempt it is the difference the direction of the cutting error makes. If you don't understand the difference between cutting cards in and out of your packet in different situations, you're going to struggle.

RJT.
 

Southpaw

Well-Known Member
#13
I really enjoyed RJT's comments in this thread, and I would say that he covered much of the bases here.

If they are dealing 3/6, it is still not going to be all that great of a game, even if you are able to cut low cards out of play every single time, which in fact you won't be able to. See, when you're cutting the low cards out of play, you're getting two benefits: (1) A positive RC--and thus usually a positive expectation--off the top of the shoe; and (2) Your TC divisors will be smaller because you ought to think of the low cards that you have cut out of play as sitting in the discard tray as iC&T has already pointed out. (Being able to use smaller TC divisors is equivalent to better penetration and thus leads to higher TC freq., which is highly advantageous to the AP).

However, if you are cutting high cards to the top, the 3/6 penetration should not concern you; it simply does not matter. In fact, some will consider this superior. You get to play the good stuff and then it is time for a shuffle; there are less -EV hands to be avoided.

As many others have pointed out, I don't like the idea of continuing to play at a store you've already burned out.

So, the upshot of all this is that (1) restricting yourself to only cutting out lows will not yield a very good profit when the penetration is so shallow because at heart, it is still a straight-counting game, albeit with nuances; and (2) it should not be you that is the shuffle tracker in this casino because you've already been burned.

So, the only viable option seems to be is to have someone else play for you that is cutting highs to the top. As RJT has said, being able to consistently cut to your target (I like Snyder's rule of +/- 3 cards) is not an easy skill to learn and thus it is probably not something that a "friend" should be trusted with.

Spaw
 

RJT

Well-Known Member
#14
Southpaw said:
I really enjoyed RJT's comments in this thread, and I would say that he covered much of the bases here.
Thanks!

Southpaw said:
(I like Snyder's rule of +/- 3 cards)
I'm not sure i do. It is true that it applies to more than just the cutting which is a huge positive, it applies to locating your packet in the final stack before the shuffle, locating it after the shuffle and then cutting to the finished packed - you have to remember with tracking, that unlike counting errors are very unlikely to cancel out and hence cumulate.

The reason i'm not fond of it is as i pointed to before, when playing high card packets (or more accurately short packets) cutting a card out is normally far more damaging than cutting one in. With a 2 deck packet (assuming a 6d shoe) it's twice as damaging to cut one card out as it is one in. With a one deck packet, 5 times as damaging. The smaller the packet the more dramatic the consequences. Knowing that should affect the way you cut.

All this of course assumes a consistent distribution through the rest of the shoe. If you were to have information on the packet just above the one you're trying to cut to, you may find occasions where it's better to cut cards out than in.....

RJT.
 
#15
RJT said:
Sucker you're wrong.

If i have a 1 deck packet with 12 extra high cards, when cutting to it if i cut one card out i have a 12/52 (3/13) chance of it being one of my extra high cards. If the opposite is true and i have 12 extra low cards in a deck, what i really have is a 5 deck playzone with 12 extra high cards. So if i mistakenly cut one card out of my playzone i have a 12/260 (3/65) chance that it's one of my extra high cards. Differentiating between low and high cards is valueless, however the difference between big and small packets - and correspondingly knowing in which direction the errors are going to cost you more - is crucial....
A meaningless distinction. If you miss one of your high cards, you're talking about what, a difference between a 4% and a 5% advantage? If you miss one of your low cards it's a difference between a 0.3% and a 0.2% advantage.

If the purpose of your cutting is to generate a play zone that you can play all by itself, it has to be an affirmative thing, you can't give yourself a significant edge just by removing a few negative cards from a shoe. If you're using the cut to steer low cards away from the play zones you determine by counting (either to the front of the shoe or behind the cut card, any place other than where you will ever have a big bet out) your method is appropriate.
 

Renzey

Well-Known Member
#16
iCountNTrack said:
shuffles are still pretty simple even in 2011 in the dear old motherland ;)
I find this claim to be very troubling. Most of the games I see use a two pass shuffle with a step ladder on the first pass. The final dilution of rich slugs seems to average about 80% with a WILD variance.
This, I say from practicing at home with marked cards and other "cheater" props as visual aids.
Can a step ladder be usefully tracked, other than for the occasional very heavy first half deck??
 

iCountNTrack

Well-Known Member
#17
Renzey said:
I find this claim to be very troubling. Most of the games I see use a two pass shuffle with a step ladder on the first pass. The final dilution of rich slugs seems to average about 80% with a WILD variance.
This, I say from practicing at home with marked cards and other "cheater" props as visual aids.
Can a step ladder be usefully tracked, other than for the occasional very heavy first half deck??
Who said anything about the US being my old dear Motherland. I am in the process of writing an article publishing results about the effect of tracked slug dilution so stay tuned :)
 
#18
Renzey said:
...Can a step ladder be usefully tracked, other than for the occasional very heavy first half deck??
Yes, very effectively, with a computer. I wrote the program and tested it.

Tell me, is it a felony to use a computer to track a shuffle, if the computer is 1000 miles away from the casino?
 

RJT

Well-Known Member
#19
Automatic Monkey said:
A meaningless distinction. If you miss one of your high cards, you're talking about what, a difference between a 4% and a 5% advantage? If you miss one of your low cards it's a difference between a 0.3% and a 0.2% advantage.

If the purpose of your cutting is to generate a play zone that you can play all by itself, it has to be an affirmative thing, you can't give yourself a significant edge just by removing a few negative cards from a shoe. If you're using the cut to steer low cards away from the play zones you determine by counting (either to the front of the shoe or behind the cut card, any place other than where you will ever have a big bet out) your method is appropriate.
I don't consider it meaningless in the slightest and i actually practice what i preach. Perhaps you should consider the mistake rate when tracking? How many cards were you out when you held that location, how much more when you located your packet in the final stack and what about the cut? The vast majority of players who ever attempt tracking don't end up cutting 1 card out, often just from the cut - and this comes from my own training - they are out by 13+ cards. Suddenly that "meaningless" statistic isn't so meaningless. Rather than get into any further worthless debate with you where we go round and round about our differing opinions on playing style as we have in years gone by i'll leave it to other readers to decide who they'd rather take their information from.



RJT.
 
#20
RJT said:
I don't consider it meaningless in the slightest and i actually practice what i preach. Perhaps you should consider the mistake rate when tracking? How many cards were you out when you held that location, how much more when you located your packet in the final stack and what about the cut? The vast majority of players who ever attempt tracking don't end up cutting 1 card out, often just from the cut - and this comes from my own training - they are out by 13+ cards. Suddenly that "meaningless" statistic isn't so meaningless.
It makes little to no sense to use a tracking technique where you need precision on the daughter shoe of less than one round of cards. You don't know how many cards a round will consume nor at what point in the play zone you will get your cards.


RJT said:
Rather than get into any further worthless debate with you where we go round and round about our differing opinions on playing style as we have in years gone by i'll leave it to other readers to decide who they'd rather take their information from.
RJT.
OK. :laugh:
 
Top