rookiego21: Shuffling Tracking calculations

[rookiego21]

Hi...I saw the 'NRS' formula from bjmaths.com for tracking. I am using the formula TC= (rc+A)/(N-L/52).

However, there are 2 questions that striked me:

1. If halfway through the segment with slugs; the running count denoted by A in the formula is positive. Do i simply use this number into the formula to calculate the TC? I felt that i will be overestimating the TC (or maybe not because i am dividing by the number of Pseudo deck denoted by N).

2. Instead of cutting the segment with slugs right to the top; i can only cut it to play after 1 deck is dealt.This is due to casino's requirement of cutting at least a certain amount of cards. After 1 deck, the running count is negative. Must i make adjustment to the starting TC of my segment? If yes, how?

I am assuming that the segment consist of 50% known data from previous shoe & merged with 50% of unknowns after the cut card.

Hope someone can enlighten me. thanks.

[Norm Wattenberger]

> 1. If halfway through the segment with slugs; the
> running count denoted by A in the formula is positive.
> Do i simply use this number into the formula to
> calculate the TC? I felt that i will be overestimating
> the TC (or maybe not because i am dividing by the
> number of Pseudo deck denoted by N).

That's accounted for in the formula. Incidentally, N can be off by a fair amount without too much cost. r can also be rounded.

> 2. Instead of cutting the segment with slugs right to
> the top; i can only cut it to play after 1 deck is
> dealt.This is due to casino's requirement of cutting
> at least a certain amount of cards. After 1 deck, the
> running count is negative. Must i make adjustment to
> the starting TC of my segment? If yes, how?

Well you don't have to. But you can to make use of the additional info and the SCORE will increase the more cards you cut in front of the play zone if you do. To do so you would need to use the Extended NRS forumla. There was an excellent post by Ted Forrester made in 2004 that describes the process at games.groups.yahoo.com/group/blackjackcardcounterscafe/message/22118. Also, CVData contains an extended NRS calculator that works in the demo without a license.

[rookiego21]

Thanks!!! i have read the post bt Ted Forrester & understand the extended NRS formula.

However, i have 1 more question.

Example:

4/8D game. Tracked slug size = 0.5 deck.

No. of slugs in tracked segment = 3

However, end of 4D the running count is -24.

By doing an averaging on the unplayed 4 decks, the average count of each half deck will be +24/8 = +3. ie. each half deck will have on average 3 low cards.

Assuming, that the merged segment will be 1 deck. If we adopt the average true count method, the count is exactly zero. That means we will not bet big at all.

However, if we use the NRS fomula; we will assume that the segment will have 3 high cards; without factoring the average count of the unknowns.. Is this the right way; ignoring the unknowns? If yes; isn't it very risky?

Thanks

[Norm Wattenberger]

Extended NRS doesn't ignore unknowns. That's why it's so messy. We always have unknowns in BJ. The point of NRS is to create a pseudo-shoe taking into account as much info as we can. An obvious case is cutting bad cards out. Then the PZ is all unknowns. But, we can start with a positive RC.

[rookiego21]

> Extended NRS doesn't ignore unknowns. That's why it's
> so messy. We always have unknowns in BJ. The point of
> NRS is to create a pseudo-shoe taking into account as
> much info as we can. An obvious case is cutting bad
> cards out. Then the PZ is all unknowns. But, we can
> start with a positive RC.

However, the example i mentioned in my previous post is different..
We know that on average, the so call 'unknowns' have 3 low cards for each half deck.Our tracked half deck segement has 3 high cards. By using the averaging method, the tracked segment with 3 high cards are 'neutralised' by the average 3 low cards of the unknowns.

If we use the NRS formula, what is the right way to cater for the average low cards in the unknowns?

[Norm Wattenberger]

The count of the first half of the shoe doesn't matter. What is the count of the slugs? (The TZ) If it is positive and you cut it into play, then the count of the PZ will be positive on average; albeit diluted. It will NOT be neutralized because you don't mix all of the unknowns in.

In NRS the 1.5 decks in the slugs is the known and the remaining 6.5 decks is the total unknowns and has the opposite count. But the count is diluted by only the part of the unknowns that are mixed into the TZ to form the PZ.

In extended NRS, this is further adjusted by the fact that we also know the count of the cards (y) that we cut in front of the PZ by the time we reach the PZ. That alters the count of the total unknowns.

d'=d-b (where b=the one deck you cut in front of the PZ) is the adjusted total number of unknowns. That is used throughout the extended NRS to give us a new N (pseudo-decks) and ultimately TC. All that messiness in extended NRS gives us a TC that takes into account the count of the knowns in the TZ, the count of the knowns in front of the PZ, the dilution of the PZ by the unknowns that are mixed into the PZ and all their relative sizes.

Fortunately most of the messy calculations are done before we sit down at the table. Assuming you control the cut card.

[rookiego21]

> The count of the first half of the shoe doesn't
> matter. What is the count of the slugs? (The TZ) If it
> is positive and you cut it into play, then the count
> of the PZ will be positive on average; albeit diluted.
> It will NOT be neutralized because you don't mix all
> of the unknowns in.

> In NRS the 1.5 decks in the slugs is the known and the
> remaining 6.5 decks is the total unknowns and has the
> opposite count. But the count is diluted by only the
> part of the unknowns that are mixed into the TZ to
> form the PZ.

> In extended NRS, this is further adjusted by the fact
> that we also know the count of the cards (y) that we
> cut in front of the PZ by the time we reach the PZ.
> That alters the count of the total unknowns.

> d'=d-b (where b=the one deck you cut in front of the
> PZ) is the adjusted total number of unknowns. That is
> used throughout the extended NRS to give us a new N
> (pseudo-decks) and ultimately TC. All that messiness
> in extended NRS gives us a TC that takes into account
> the count of the knowns in the TZ, the count of the
> knowns in front of the PZ, the dilution of the PZ by
> the unknowns that are mixed into the PZ and all their
> relative sizes.

> Fortunately most of the messy calculations are done
> before we sit down at the table. Assuming you control
> the cut card.

Thank you very much....All the calculations are solved already....