Index Deviation

I was wondering if anybody deviated from their index plays based on (rc) count flux. Or if you see a certain amount of facecards or little cards prior to your turn.

For example, sometimes when theres a super high count and its followed by the deal, the TC is now dropped by half. In addition to this you see to face-cards fall right before you. You now have a border line situation wether or not to double 9vs7. Would you still double??? I wouldnt!(dont)
Or how about if the situation was reversed.

These inutuitive descions based on how, and how much the RC or TC fluctuates helps me determine those borded line descions. This have been a secret weapon in my arsenal, that seem to help me with those close call plays.

Anybody else use such a technique?

jack said:

For example, sometimes when theres a super high count and its followed by the deal, the TC is now dropped by half. In addition to this you see to face-cards fall right before you. You now have a border line situation wether or not to double 9vs7. Would you still double??? I wouldnt!(dont)

Click to expand...

Of course I would! Even though the count dropped, the index tells you that there are still enough high cards in the shoe to make the double down more profitable than a hit. There is no reason to think that another high card will not come out just because a few just did. In fact, with a high TC you are expecting them to keep coming out!

Avoiding index plays based on the “count flux” is exactly the same as avoiding a BS play based on the flow of the cards. You've worked hard to become an Advantage Player. Don’t let these superstitious ideas turn you back into a ploppy.

-Sonny-

they say in a borderline decision you play basic strategy (round your tc down), but i dont agree with this, because i think whatever you have calculated/estimated it is, you should go with that.. if its close follow bs? if thats true then the index plays are underestimated, and wrong.. taking insurance at 3.1 should be the right play and taking it at 2.9 should be the wrong play..

i would like to know how estimated these indexes are (they obvious could be off as much as .5 tc, which is a lot), and i would love for somebody to link me to a place where they list indexes to the nearest tenth, not the nearest whole, so it would look something like this.. insurance +3.2, 16 vs 10 +.4, 12 vs 3 +1.8, etc.. because i think that rounding to the nearest whole number is highly inaccurate for indexes, especially for those dividing by half decks.. think about it, if there was almost 3 decks left, and one of the +3 indexes were actually like 2.6, you would know to make the index play

would like to know how estimated these indexes are (they obvious could be off as much as .5 tc, which is a lot), and i would love for somebody to link me to a place where they list indexes to the nearest tenth

Click to expand...

The sams case index generator on snyders website does this. Have you checked it out yet?

SilentBob420BMFJ said:

i think that rounding to the nearest whole number is highly inaccurate for indexes, especially for those dividing by half decks..

Click to expand...

Precise indices? There’s no such thing. There is so much estimation and assumption that goes into index generation that many people just use an approximation formula. Here’s an old discussion about it:

(Dead link: http://www.advantageplayer.com/blackjack/forums/bj-main/webbbs.cgi?read=18623)

Here are some more articles:

http://www.bjmath.com/bjmath/tcindex/pmrt.htm (Archive copy)
http://www.bjmath.com/bjmath/tcindex/igen.htm (Archive copy)

In any case, even if you could calculate precise index numbers, would you be calculating the TC to that many decimal places too? Would you be dividing by the exact number of cards that are left in the shoe? If your indices are completely accurate and your TC isn’t then you haven’t gained anything.

Also, see my relp here:

http://www.blackjackinfo.com/bb/showthread.php?t=7603

-Sonny-

jack said:

The sams case index generator on snyders website does this.

Click to expand...

I'm afraid not:

Blackjack Strategy Table Generator

Version 1.0

Programmed by Sam Case

Based on Arnold Snyder's Algebraic Approximation Method

This generator (like many programs and spreadsheets) uses an approximation formula. Those decimal places are not necessarily accurate. If a formula is only an approximation, it doesn’t matter how many decimal places it returns if they aren’t accurate.

-Sonny-

Sonny said:

Precise indices? There’s no such thing. There is so much estimation and assumption that goes into index generation that many people just use an approximation formula. Here’s an old discussion about it:

(Dead link: http://www.advantageplayer.com/blackjack/forums/bj-main/webbbs.cgi?read=18623)

Here are some more articles:

http://www.bjmath.com/bjmath/tcindex/pmrt.htm (Archive copy)
http://www.bjmath.com/bjmath/tcindex/igen.htm (Archive copy)

In any case, even if you could calculate precise index numbers, would you be calculating the TC to that many decimal places too? Would you be dividing by the exact number of cards that are left in the shoe? If your indices are completely accurate and your TC isn’t then you haven’t gained anything.

Also, see my relp here:

http://www.blackjackinfo.com/bb/showthread.php?t=7603

-Sonny-

Click to expand...

also Zengrifter has some thoughts on these issues in his interview:
http://www.blackjackinfo.com/interviews.php

Sonny said:

I'm afraid not:

Blackjack Strategy Table Generator

Version 1.0

Programmed by Sam Case

Based on Arnold Snyder's Algebraic Approximation Method

This generator (like many programs and spreadsheets) uses an approximation formula. Those decimal places are not necessarily accurate. If a formula is only an approximation, it doesn’t matter how many decimal places it returns if they aren’t accurate.

-Sonny-

Click to expand...

I completely agree. Believe it or not, after using index plays, especially in a 2D game i can tell just about how accurate they really are. To say the least when i first started using the A02 a few years back, i was extremely disappointed when i kept hitting stiffs against dealers bust card and busting, vs when i would of won.

The same thing is happening with my new count. Which you may already know, im using the sam case index generator to generate my indices. Like the A02 they are flawed.

How! and do mean how would i go about getting precise indexs. Is there a precise program(software) that you would recommend?

jack said:

How! and do mean how would i go about getting precise indexs. Is there a precise program(software) that you would recommend?

Click to expand...

You’ve got to do it yourself. Only you can decide how accurate your deck estimations are, whether you want to round up/round down/ truncate/floor the TC, select a specific bet spread and bet ramp, a specific game with specific penetration and a specific number of players, round up/round down/ Hit to the Left/truncate/floor the indices, etc.

Then comes the fun part: You have to find a way to estimate the discards and convert the RC to TC just as precisely as you calculated your indices. This means calculating fractions/decimals on the fly at the tables with perfect precision and perfect deck estimation.

Then comes the really fun part: You have to play the exact game you used above with the same rules, same penetration, same number of players, same bet ramp and same accuracy that your calculations are based on.

Then comes the really really fun part: Once all the smoke has cleared and all the numbers are completely accurate and your skills are finely crafted – it probably won’t add up to more than a penny or two in many cases. You could probably earn more by using a simpler (faster!) system and playing an extra few hands per hour.

-Sonny-

If the TC is exactly zero, and I have a 16v10, then I might look at the relative composition of 6s to 5s to determine if I hit or stand. (if there are more 6s than 5s, then I'd hit). It's more of a tie-breaker than a deviation though.

Sonny said:

Of course I would! Even though the count dropped, the index tells you that there are still enough high cards in the shoe to make the double down more profitable than a hit. There is no reason to think that another high card will not come out just because a few just did. In fact, with a high TC you are expecting them to keep coming out!

Avoiding index plays based on the “count flux” is exactly the same as avoiding a BS play based on the flow of the cards. You've worked hard to become an Advantage Player. Don’t let these superstitious ideas turn you back into a ploppy.

-Sonny-

Click to expand...

i was thinking the same thing, when i was reading what he wrote i was like wtf is he talking about? flux? fluctuation? that cant be predicted, so what is going on here..

jack said:

The sams case index generator on snyders website does this. Have you checked it out yet?

Click to expand...

nope

Sonny said:

Precise indices? There’s no such thing. There is so much estimation and assumption that goes into index generation that many people just use an approximation formula. Here’s an old discussion about it:

(Dead link: http://www.advantageplayer.com/blackjack/forums/bj-main/webbbs.cgi?read=18623)

Here are some more articles:

http://www.bjmath.com/bjmath/tcindex/pmrt.htm (Archive copy)
http://www.bjmath.com/bjmath/tcindex/igen.htm (Archive copy)

In any case, even if you could calculate precise index numbers, would you be calculating the TC to that many decimal places too? Would you be dividing by the exact number of cards that are left in the shoe? If your indices are completely accurate and your TC isn’t then you haven’t gained anything.

Also, see my relp here:

http://www.blackjackinfo.com/bb/showthread.php?t=7603

-Sonny-

Click to expand...

if the true count is rounded, and the indexes are rounded, that is worse than having only one of those two rounded.. you cant tell me that if you had an index of +2.8 and you were pretty sure there are exactly 3 decks left, that you wouldnt use basic strategy instead of the index, whereas you would have probably done the index otherwise, thus you just saved money

this is exactly why numbers vary from person to person, because of rounding, the more you round, the more off your numbers could/will be.. example: .9*.9 = .81, but lets say you rounded one of those numbers 1*.9=.90, now lets say you rounded both, 1*1=1.. rounding both instead of rounding none is about a 20% difference..

Sonny said:

You’ve got to do it yourself. Only you can decide how accurate your deck estimations are, whether you want to round up/round down/ truncate/floor the TC, select a specific bet spread and bet ramp, a specific game with specific penetration and a specific number of players, round up/round down/ Hit to the Left/truncate/floor the indices, etc.

Then comes the fun part: You have to find a way to estimate the discards and convert the RC to TC just as precisely as you calculated your indices. This means calculating fractions/decimals on the fly at the tables with perfect precision and perfect deck estimation.

Then comes the really fun part: You have to play the exact game you used above with the same rules, same penetration, same number of players, same bet ramp and same accuracy that your calculations are based on.

Then comes the really really fun part: Once all the smoke has cleared and all the numbers are completely accurate and your skills are finely crafted – it probably won’t add up to more than a penny or two in many cases. You could probably earn more by using a simpler (faster!) system and playing an extra few hands per hour.

-Sonny-

Click to expand...

what does bet ramp have to do with indexes? please dont tell me that indexes are based on how much money you have out and how many players are at the table (i dont see why it would matter anyway), because if that is true, then ive been playing wrong all this time, because i flat bet

SilentBob420BMFJ said:

because of rounding, the more you round, the more off your numbers could/will be.. example: .9*.9 = .81, but lets say you rounded one of those numbers 1*.9=.90, now lets say you rounded both, 1*1=1.. rounding both instead of rounding none is about a 20% difference..

Click to expand...

Yes, but that 20% difference will add up to about a 0% difference in your EV. Even being off by 2-3 index points in either direction (plus or minus) will not hurt you that much. Check out some of the links I gave earlier in this thread.

-Sonny-