Why Split 9's?

[Sonny]

Permit me to chose an arbitary figuire of pair of 9's occuring 1 in 200. ( Smarter guys than me like Sonny would no doubt have the actual correct number).

Let's use 1 in 2,439.

If you don't split, what is the penalty in terms of the increase in House Advantage in % terms over BS.

From the EV tables I linked to in my last post we can see that splitting has an EV of 22.0803% (for the 2D S17 DAS game). If we stand then the EV is only 8.5625%, so we have lost about 13.5178%. That is the conditional cost of making that play. Multiplying that by the frequency of the hand we get a total loss of 13.5178%*0.00041=0.005542% (I can feel Qfit wincing at my 8-decimal place inaccuracy ). That is the absolute cost of always making that play, which is the amount that the house edge will increase. It's a very big mistake to make but it is rare enough that it doesn't cause that much damage overall. A $10 player will lose a few extra pennies per hour even though he will be giving up $1.35 every time he makes that mistake.

-Sonny-

 

[London Colin]

If you never split 9s under 2D S17 DAS ....

From Sonny's bjmath table -

	   [B]Stand      Split     Diff      Freq.     Product[/B]
9,9 vs 9  -0.189530  -0.088330  0.101200  0.000308  0.000031
9,9 vs 8   0.085625   0.220803  0.135178  0.000410  0.000055
9,9 vs 6   0.275070   0.448757  0.173687  0.000410  0.000071
9,9 vs 5   0.201095   0.398538  0.197443  0.000410  0.000081
9,9 vs 4   0.170851   0.315373  0.144522  0.000410  0.000059
9,9 vs 3   0.136895   0.239640  0.102745  0.000410  0.000042
9,9 vs 2   0.129292   0.202500  0.073208  0.000410  0.000030
					
                                            Total:  0.000370  (0.0370%)

 

[Mr. T]

Thanks guys.
If you don't mind me saying , Sonny, it just proves that you are smarter than me.

 

[Mr. T]

Op! I just notice Sonny's figure is 0.005542% .

While London Colin figure is 0.0370%

How come this huge difference.

 

[London Colin]

Sonny just worked it out for 9,9 vs 8, the second line in my list.

 

[Mr. T]

If you don't mind me saying, Sonny, London Colin may be smarter than you.

 

[k_c]

Confirmation

My tdca program computes EV using total dependent basic strategy for a player hand of 9-9 versus an empty dealer to be +7.0250% for 2 decks, S17, DAS, split to 4 hands. This means that if player was offered a hand of 9-9 from a freshly shuffled 2 deck shoe with the above rules he should accept the offer. He would proceed to play his 9-9 according to basic strategy, depending on dealer's up card for an overall EV of +7.0250%.

Using the same scenario except changing the rules so that no splits are allowed changes player's EV to -0.0498%
Player's loss in EV due to not being able to split 9-9 = 7.0250% - (-0.0498%) = 7.0748%

From a freshly shuffled 2 deck shoe the probability of being dealt 9-9 = 8/104 * 7/103
Loss of total dependent basic strategy EV adopting a strategy of not splitting = 7.0748% * 8/104 * 7/103 = ~.037%

So if it is given player has a hand of 9-9 and does not split as called for according to basic strategy this will cost 7.0748% for that particular hand and if he makes a habit of doing this it will decrease his overall EV by about .037%.

 

[Dopple]

I would split two 9s against the 7 if the count is +5

Harry
Go Jets

Looking my trusty Uston APC tables I get +10 for 99 v 7 and blank for 8 on up. Were does the difference come from?

 

[bjsim]

Blackjack Simulation - Bjsim

I vote that Colin is Smarter and Nicer!!!

I will deal later on with upgrading bjsim to deal with hundreds of millions hands Simulations and still keep it as free web application.(Vs. Desktop).

I fixed the EV to be based on the Orig bets for all Splits and Doubles and to included Pushes.
Sample can be viewed at
(Dead link: http://www.bjsim.com/sample/BJSIM141.5N3NMD1.htm)

I also fixed the Total EV to be based on the Orig bets.
Does anyone (Colin or Sonny) know if this is the right calculation or the Total EV should be based on total bets per hand.

Another issue that i had is if to count Double after Split in the Grand Total Ev.

I counted those hands in the Doubles 11 on 8 Separately but not in the Grand Total EV . Grand Total always based on Orig bets

Any idea how if this is correct will be appreciated.

 

[London Colin]

I vote that Colin is Smarter and Nicer!!!

I very much doubt that either of those assertions is true.

I will deal later on with upgrading bjsim to deal with hundreds of millions hands Simulations and still keep it as free web application.(Vs. Desktop).

I fixed the EV to be based on the Orig bets for all Splits and Doubles and to included Pushes.
Sample can be viewed at
(Dead link: http://www.bjsim.com/sample/BJSIM141.5N3NMD1.htm)

I also fixed the Total EV to be based on the Orig bets.
Does anyone (Colin or Sonny) know if this is the right calculation or the Total EV should be based on total bets per hand.

Another issue that i had is if to count Double after Split in the Grand Total Ev.

I counted those hands in the Doubles 11 on 8 Separately but not in the Grand Total EV . Grand Total always based on Orig bets

Any idea how if this is correct will be appreciated.

My previous post dealt with this issue as fully as I know how. All EV calculations should be in terms of the inital bets only, and all outcomes should be included; you can't pick and choose. The grand total EV is simply the total amount won divided by the total of all the initial bets.

I'm not 100% sure what changes you have made, but your tables still seem to imply some confusion. (In part, possibly because you haven't changed the headings.)

E.g.,

[B]Split           Hands     Win   Lose  Push  Bet Amt    Win/LoseAmt  EV(%Win) 
               Played                       Include DAS [/B]
 
Split 9 On 8    2,878    1,436  1,155 287  $309,300    $30,400      22.65%

What does the above mean?

If the rules allow splitting to 4 hands and DAS, then for every round (i.e. initial hand) of 9,9 vs 8, there are 17 possible outcomes - win/lose between 0 and 8 bets.

If you want to calculate the EV for splitting the initial hand, and compare it with that for standing, then you need to track the total amount won/lost for each entire round (whether that round consist of a single hand on which you stand, or up to split 4 hands, with each one possibly doubled).

 

[zengrifter]

I was going to let this one go since Qfit already responded to it, but there are just too many issues to ignore.

That’s a HUGE range! An answer that vague is not helpful at all. The advantage for standing on 9,9 vs. 8 (2D S17 DAS) is about 8.56%. So is splitting really better than standing if we only know that it is somewhere between 4% and 10%? It is impossible to make the right decision with that information.

The results on your website show an advantage of 11.6% so I’m not sure where the 4-10% came from. Maybe that was the result of the individual sims before you averaged them together?

As numerous people have said before, 1M rounds is not nearly enough to get accurate numbers. According to your website you only played 2,726 hands of 9,9 vs. 8 to get your EV estimate. Play another 2k hands and see how different the results are. Then try it again. Do you understand why those results are completely unreliable?

The numbers you give are not correct. If we look at the true EV (Archive copy) of that play we can see that it is around 22.1%, not 4-10% (or 11.6%).

So there are obviously problems with your program and your methodology. You really need to address these issues if you want to make your software practical.

-Sonny-

He meant DEPENDING on whether one of the 9s becomes a double down. zg

 

[zengrifter]

(Sorry no link as I get criticized for links to the correct tables.)

There is no criticism of QFIT linking to correct tables, is there guys?!?! zg

 

[zengrifter]

Another issue that i had is if to count Double after Split in the Grand Total Ev.

Grand Total Ev hence forth shall be called GTEV!
OFIT and Ken add GTEV to your website glossaries! zg

Ps - While we are at it, does anyone else have any
other new terms or acronyms for BJ?

 

[daddybo]

Grand Total Ev hence forth shall be called GTEV!
OFIT and Ken add GTEV to your website glossaries! zg

Ps - While we are at it, does anyone else have any
other new terms or acronyms for BJ?

DAP: Disadvantaged Player. One who plays at at a disadvantage regardless of what he/she thinks. Also a kinder name for ploppy.

 

[daddybo]

Grand Total Ev hence forth shall be called GTEV!
OFIT and Ken add GTEV to your website glossaries! zg

Ps - While we are at it, does anyone else have any
other new terms or acronyms for BJ?

Then there would be...

PAD or Padding: Play(ing) at a disadvantage, whether intentionally or not.

 

[daddybo]

Grand Total Ev hence forth shall be called GTEV!
OFIT and Ken add GTEV to your website glossaries! zg

Ps - While we are at it, does anyone else have any
other new terms or acronyms for BJ?

then there would be..

KAMKS: Killing All My Kittens ... with these assanine/useless posts.

and of course TB and PB, TB: temporary ban vs PB Permanent ban. :laugh::laugh::laugh:

OK, I'm through.

 

[Frankie]

Suppose a casino had a promo offering you 18 every hand you played. You'd lose. 18 is a net loser.

I think this is wrong. According to Wizard of Odds, I get a player edge of 3.4% for all hands of 18, in an infinite deck situation.

http://wizardofodds.com/blackjack/appendix1.html

I used these numbers and assumed that there were 4 chances of 10 vs. 1 of every other card (obviously). Average expected return was +.0336.

 

[Harry1941]

Basic strategy was invented by organized crime to make their casinos more profitable!


Harry