Standard Deviation question and request for the SCORE of TKO

This is a question dealing with the standard deviation numbers listed specifically in Blackjack Attack.

I back-count mostly 8 deck shoe games, the SD/Hand for 6/8 S17 DAS for a 1-8 spread is given as 3.74. The SD/100 is given as 18.54. I don't really understand what these numbers mean. Other people have quoted the Standard Deviation formula to be 1.1*(square root of N). Wouldn't this formula suggest that the standard deviation wouldn't change as the number of decks, rules change. I am confused about the subject and an explanation of how these numbers came about and what they exactly mean would be helpful.

The book also compares the SCORE for different games playing Hi-Lo and Preferred KO. There are no SCORE numbers, however, for TKO. This is the system i use and i am intrested in how effective it is, seeing that Preferred KO actually outperformed Hi-Lo in some instances (mainly back-counting multiple deck games). If the system were to be true counted, i would think it's effectiveness would increase even more. Perhaps it is a supior system when true counted.
Has anyone ever run simulations on TKO (with the I-18, using the correct true count values, not the rounded)? I guess that's all for now.

Impala

"I back-count mostly 8 deck shoe games, the SD/Hand for 6/8 S17 DAS for a 1-8 spread is given as 3.74. The SD/100 is given as 18.54. I don't really understand what these numbers mean. Other people have quoted the Standard Deviation formula to be 1.1*(square root of N). Wouldn't this formula suggest that the standard deviation
wouldn't change as the number of decks, rules change. I am confused about the subject and an explanation of how these numbers came about and what they exactly mean would be helpful."

The 1.1 figure is an approximation of the SD/Hand for a BS player who is flat-betting. As such, it has no bearing on the sims of Chapter 10, where a) we are counting and using the I18, and, much more importantly, we are varying our bet sizes, sometimes as high as a 1-12 spread.

The result is that the SD/Hand for a counter is considerably higher than it is for a flat-bettor.

As for the SD/100, for the person "playing all," it represents 10 times the SD/Hand, since SD is a square root function, and the square root of 100 is 10. For the back-counter, the figure represents the SD/Hand multiplied by the square root of the number of hands actually played, per 100 hands seen. Since this is usually of the order of 25 played hands, and the square root of 25 is 5, the back-counting SD/100 entries are normally about 5 times the SD/Hand values.

As for TKO, I think there have been many sims that show it to be a very fine system. I don't have the precise comparisons.

Don

i still have no idea what these numbers mean. Does it mean that after 100 hands, your winnings could vary 18.54 units in either direction from your EV? Still don't know how to use these numbers to determine what my fluctuations will be in my bankroll in a given amout of hands/sessions. This is probably a simple question, but i am still confused.

impala