Simulations of counting types
Steve F 88
Abdul Jalib <abd...@earthlink.net> has written Re: Spread question
>
> Downtown double deck is a horrible game (assuming H17 and no huge rules
> to make up for that) by Vegas standards. You would need a big spread to
> get any sort of edge on the game, a big enough spread to get booted,
> likely.
>
> If you play a strip double deck (S17, DOA, DAS) with the end-of-play
> card at 70% or better and spread 1-5, you should be able to clear 1
> unit per hour. So, if you want to make $20 per hour, you'll need to
I have run my Blackjack simulator on this scenario with the following
results:
==========================================================================
Output from Fry's BlackJack (FBJ) V 1.4 Simulation
NDEK = 2: Number of decks is two
NCSH = 30: Number of cards remaining to shuffle deck
IDBL = 0: Double Down only on any first two cards
ISPL = 1: Split pairs unlimited
ISPA = 0: Split aces get only one card
ISPT = 0: Splitting of 10 and face card allowed
ISPD = 1: Double Down after Splitting allowed
INSU = 0: Insurance allowed
ISUR = 0: Surrender not allowed
JHSS = 1: Dealer stands on all Seventeens
JBJP = 0: BlackJack pays 3 to 2
JTIE = 0: Ties are a push
JPEK = 1: Dealer peeks only on 11
KSYS = -6: Playing system is Uston count - - Efficiencies
Card value: A 2 3 4 5 6 7 8 9 TEN Bet Play
---------------- --- --- --- --- --- --- --- --- --- --- ------ ------
Count value: -1 0 1 1 1 1 1 0 0 -1 .949 .781
KBET = -1: Amount to bet is the True Count plus KBET
Number of hands per visit = 1000
Number of units in bankroll each visit = 1000
Number of units to bet Minimum, Normal, Maximum = 1, 1, 5
Average per visit Average per visit
Units Won (sigma) Max. Bank. (sigma)
--------- --------- ---------- ---------
12.9 63.3 43. 35.
==========================================================================
Now this simulation is for 1000 hands, which I estimate to be 5 hours. So
instead of one unit per hour I think it is closer to 2.6 units per hour.
Running the simulation with a 1/2/16 spread (bet 2 units with a zero count
and one unit with a negative count), which can be interpreted as a 1-8
spread, the results are:
==========================================================================
Number of units to bet Minimum, Normal, Maximum = 1, 2, 16
Average per visit Average per visit
Units Won (sigma) Max. Bank. (sigma)
--------- --------- ---------- ---------
16.3 87.7 60. 49.
==========================================================================
This gives you about 3.3 units per hour on average.
jims...@aol.com (Jimsa207) has written Re: 7,8,9 vs A,2,8 counted as 0
>
> I have been counter many years. Always counted 7,8,9 as 0; A,10 as -1;
> & 2-6 as +1. Recently read book (Black Jack Your Way to Riches) that
> suggests A,2,8 as 0; 9 & 10 as -1; & 2-7 as +1.
>
> Does anyone have any experiance using both or have an opinion which
> may be better. Only close place to play is an 8 deck game, double on
> anything, double after split, split anything, dealer stands on soft 17.
That book is the only BJ book I own, and it was my introduction to modern
counting. I got it at a book sale for 50 cents 6 years ago.
To run Jim's count on my simulator, I first looked at the list of counts:
+/- 1: Basic Strategy A 2 3 4 5 6 7 8 9 T
2: Fry count . . . . 0 1 1 1 1 1 1 1 1 -2 .617
3: Canfield count . 0 0 1 1 1 1 1 0 -1 -1 .623
(Hi-Opt I) 4: Einstein count . 0 0 1 1 1 1 0 0 0 -1 .615
5: Revere-1 count . 0 1 1 1 1 1 0 0 -1 -1 .592
6: Uston count . . . -1 0 1 1 1 1 1 0 0 -1 .547
7: Braun count . . . -1 1 1 1 1 1 0 0 0 -1 .510
8: Gordon count . . 0 1 1 1 1 0 0 0 0 -1 .574
(Hi-Opt II) 9: Stepine count . . 0 1 1 2 2 1 1 0 0 -2 .671
10: Revere-2 count . -2 1 2 2 2 2 1 0 0 -2 .527
11: Archer count . . 1 1 1 1 1 1 1 1 1 -2 .584
12: Ita count . . . . -1 1 1 1 1 1 1 0 -1 -1 .532
13: Aux. 2 count . . 0 0 0 1 1 1 1 0 0 -1 .637
14: Aux. 3 count . . 0 1 1 2 2 2 1 0 -1 -2 .672
15: Aux. 4 count . . 0 1 1 1 2 1 1 1 0 -2 .670
and got the following results for count type 7:
==========================================================================
Output from Fry's BlackJack (FBJ) V 1.4 Simulation
NDEK = 8: Number of decks is eight
NCSH =125: Number of cards remaining to shuffle deck
IDBL = 1: Double Down on any number of cards
ISPL = 1: Split pairs unlimited
ISPA = 0: Split aces get only one card
ISPT = 0: Splitting of 10 and face card allowed
ISPD = 1: Double Down after Splitting allowed
INSU = 0: Insurance allowed
ISUR = 0: Surrender not allowed
JHSS = 1: Dealer stands on all Seventeens
JBJP = 0: BlackJack pays 3 to 2
JTIE = 0: Ties are a push
JPEK = 0: Dealer peeks on 11 and 10 for Natural
KSYS = -7: Playing system is Braun count - - Efficiencies
Card value: A 2 3 4 5 6 7 8 9 TEN Bet Play
---------------- --- --- --- --- --- --- --- --- --- --- ------ ------
Count value: -1 1 1 1 1 1 0 0 0 -1 .968 .729
KBET = -1: Amount to bet is the True Count plus KBET
Number of hands per visit = 1000
Number of units in bankroll each visit = 1000
Number of units to bet Minimum, Normal, Maximum = 1, 1, 5
Average per visit Average per visit
Units Won (sigma) Max. Bank. (sigma)
--------- --------- ---------- ---------
1.7 44.8 34. 26.
==========================================================================
With the count introduced by Canfield (does anyone know his real name?)
the simulation produces the following:
==========================================================================
KSYS = -3: Playing system is Canfield count Efficiencies
Card value: A 2 3 4 5 6 7 8 9 TEN Bet Play
---------------- --- --- --- --- --- --- --- --- --- --- ------ ------
Count value: 0 0 1 1 1 1 1 0 -1 -1 .869 .890
Average per visit Average per visit
Units Won (sigma) Max. Bank. (sigma)
--------- --------- ---------- ---------
0.2 43.8 34. 27.
==========================================================================
This shows with 8 decks the Canfield count is useless, while the Braun
count can show a small gain (only a .14% player advantage), due to the
inclusion of the Ace. But notice the variance in relation to the gain
and see it is not very profitable.
My blackjack program is free (I don't expect any Shareware money any more)
at http://members.aol.com/SteveF88/blackjac.htm
This program is a very simplistic (runs in DOS with the driver ANSI.SYS
loaded) character-based (not graphic-based) program that will play
blackjack with you like a game or run a simulation given the list of
simulation parameters as shown above.
-- Steve Fry