Double Deck Exit Count QuestionAs I am playing the 10 and five dollar games, my bedspread, obviously looks a bit ridiculous as such I’d like to know at the 26 and 52 penetration levels, are there Hilow or Keio eggs accounts where I should exit until the shuffle.

I ran 300 million hands and got $56.70 with the settings. 

One thing that could be causing the difference is the way true count is calculated. I divide the running count by the number of decks remaining. I assume the player can recognize and divide with 1/2 decks so I round to the nearest 1/2 deck and depending on where in the shoe, the divisor is either 6 decks remaining, 5.5, 5, 4.5, 4, 3.5, 3, 2.5, 2, 1.5, 1, and .5 deck although with 75% depth we never get that far. 1/2 deck is likely a pretty highly accurate divisor to recognize and divide with  in terms of what can be done at the tables.

I tried rounding to the nearest full deck, so using 6, 5, 4, 3, 2, 1 and ran 300 million hands and I get $54.31.

I may try doing a ceiling round to the full deck and see how the result compares. 

Another difference could be if insurance and insurance results are included in the total $ bet, and % EV that you have. I am assuming they would be.

I do include insurance in total $ bet and % won or lost.

Thanks,

Phil

I just run 2.5 million hands at a time as well, meant to note that in my previous post. Thanks.

Phil

I added truncate to the rounding choices.

Thanks,

Phil

I am still getting numbers that are higher per 100 rounds than expected. Am getting $57 to $58, so likely significantly higher than $53. Can you check my settings compared to the settings on your simulations when you get $53?

Settings I am using:

6 decks deal to 75%, cut card is after card 234 (.75 x 312), stand on soft 17, double after first 2 cards, double after split allowed, no resplit of Aces, no surrender, and blackjack pays 3:2.

Hi-Lo count with betting levels: $25 at < 1, $50 at >= 1, $100 at >= 2, $200 at >= 3, and $400 at >= 4. My average initial bet has been at around $59.50

Basic strategy is wizard of odds 4-8 decks stand on soft 17.

https://wizardofodds.com/games/blackjack/strategy/4-decks/

For sweet 16 strategy indexes I am using:

Stand:

16 v. 10  >=  0

15 v. 10  >=  4

16 v. 9    >=  5

13 v. 2    >= -2

13 v. 3.   >= -3

12 v. 2    >=  3

12 v. 3    >=  2

12 v. 4    >=  0

12 v. 5    >= -2

12.v. 6    >= -1

double:

11 v. A.   >=  1

10 v. A    >=  4

10 v. 10 >=   4

 9 v.  2  >=    1

 9 v.  7  >=    3

Insurance at >= 3

I am dividing running count by the decks remaining rounded to 1/2 deck and then flooring this result to get true count. The bet is based on the true count calculated before the hand is dealt.. Strategy decisions are made assuming I have counted my 2 cards, the dealers up card, and if hitting counting the hit card received.

cards out.       divisor

1 -      14 divide by  6

15    - 40               5.5   

41    - 66                  5

67    - 92               4.5

93   - 118                 4

119 - 144              3.5

145 - 170                3

171 - 196             2.5

197 - 222                2

223 - 248             1.5

249 - 275                1

276 - 312             0.5

Thank you,

Phil

Wanted to add - average percent advantage is coming out to around 0.86%.

Phil

Looks like my average is closer to $56.50 per 100 rounds not $58, and looks like 0.83% advantage. 

Still looks significantly higher than $53.

Thanks,

Phil

pjsatz, the $53 was coming from the sim done with your simulator and probably just not enough big of a sample.

With the settings you describe and CVCX included sim, it shows an EV of $57.12, SD $1224.84/100 rounds playing head on with the dealer. The average bet is $59.54 and the %EV is 0.959%

The %EV we use is the IBA (Initial Bet Advantage) and if your %EV is lower it's probably because you are using the TBA (Total Bet Advantage) which takes into account all the money wagered instead of just the initial bet. This is not the way blackjack serious players or researchers work. The IBA is the norm.

Can't help much with doing sims on CVData for you, I am an active player and doesn't spend much time on my computer...

Thank you for running the sims and for the data. I am an active player as well.

I am confused about whether normal practice is to show % advantage as % of Initial Bet or normal practice is to show % advantage of Total Bet including splits, double, and insurance. I interpreted the earlier comments as saying I would be out of step if I presented % of Initial Bet Advantage and that I should be presenting % of Total Bet. I can show both. 

It sounds like my $ EV lines up with pretty well with CVCX. CVCX is getting a little higher $ EV, I am getting $56.50 and CVCX is getting $57.12. There may be a difference in how true counts are calculated.

I should be able to calculate Standard Deviation per 100 rounds. I would like to show SCORE, adjusted SCORE, or DI, not sure.

Thank you,

Phil

I am a little bit confused here. Before we make any meaningful comparison, let us clarify this definition first. 

I would go by this following definition: 

cards out.       Divisor

0-13     divide by  6

14 -39                5.5   

40-65                   5

66-91                 4.5

92-117                  4

118 - 143             3.5

144 - 169                3

170 - 195             2.5

196 - 221                2

222 - 247             1.5

248 - 273               1

274 - 299            0.5

300-312.             0.5

Is this better?

Sorry, yes, I think my other table was a little off.. I have been running with the following:

Cards out  Divisor

0 - 13           6
14 - 39         5.5
40 - 65         5
66 - 91         4.5
92 - 117       4
118 - 143     3.5
144 - 169     3
170 - 195     2.5
196 - 221     2
222 - 247     1.5
248 - 273     1
274 - 312     0.5

Thanks,

Phil

I believe these published HiLo TC deviation indices were calculated by using exact deck numbers as the divisors. That is why I asked you to add an exact deck option there.  The current simulator is good enough though.   

I can look at adding an exact option, and eventually a way to select the cards out and divisors. It shouldn’t really slow it down.

Thanks, 

Phil

Waste of time. No one on earth plays that way. Make the simulator PRACTICAL.

Don

"Make the simulator practical"

No need for simulations...just open BLACKJACK  ATTACK and truth Don's work.

Stop wasting time procrastinating and get to work.