davethebuilder
Joined: Jan 11, 2015
• Posts: 61
January 14th, 2021 at 6:11:18 AM permalink
I am writing to ask if anyone can explain in plain English the meaning of the two Distribution of Ranks tables which can be found at https://wizardofodds.com/games/blackjack/cut-card-effect The first table showing the reduction in house edge for CSM games using different decks is fairly easy to understand but the next two tables have me totally confused. For example:

What is being measured and what is the practical meaning of the results?

To run a CSM simulation requires precise knowledge of the algorithm the machine uses, however, in this case was an actual machine used?

Mike, if you're reading this you're input is appreciated in helping me interpret the data.

Thanks,

Dave
Casino Enemy No.1
heatmap
Joined: Feb 12, 2018
• Posts: 1205
January 14th, 2021 at 1:02:11 PM permalink
Quote: davethebuilder

To run a CSM simulation requires precise knowledge of the algorithm the machine uses, however, in this case was an actual machine used?

the algorithm IS the rules you are playing blackjack with - you are testing the output of the algorithm in the machine not the algforithm itself

but think of it as if you are the dealer

"deal one card face up to player 1"
"deal one card face up to dealer"
"deal one card face up to player 1"
"deal one card face down to dealer"
"is the face up card a ten?" > "check hole card" > "is hole card an ace?" > "annouce blackjack take any losing bets that arent another blackjack"

obviously this isnt the entire thing but - whatever rules you are trying to simulate must be programmed into the simulation kind of like that

so if the dealer hits on soft 17 that would have to be programmed into the simulation and then compared statistically
davethebuilder
Joined: Jan 11, 2015
• Posts: 61
January 21st, 2021 at 3:16:49 AM permalink
Heatmap,

Thanks for your reply but I don't think it answers the question. That is, if you input the number of decks and the rule set along with other parameters in a commercially available simulator and run a large enough sample size you will obtain a house edge. You can then vary the position of the cut card and rerun the sims and obtain slightly different results. That's the cut card effect.

As far as I can see the Distribution of Rank tables could only be measured by knowing the exact algorithm the CSM uses as this determines the output of the machine and that can then be statistically analysed. At least that's how I read it but I would appreciate some confirmation. Also, why after so many trials are the results so different, why is there a preference towards the large cards, what is the meaning of the exercise and what is the practical effect on the player?
Casino Enemy No.1
heatmap
Joined: Feb 12, 2018
• Posts: 1205
January 21st, 2021 at 6:08:33 AM permalink
fair warning of which i should have said before - im not a math guy per say but a programmer who knows a lot of little things about a lot of random casino things. im always on the internet reading documents such as patents and brochures for gaming machines. I know a bit of computer security and what not. but by no means an expert.

i know a bit about the process of creating a casino game - but never have i ever - yet.

the only time ive ever heard of anyone getting source code to any machines is gaming control boards.

but because of lack of money that is usually outsourced to a more reputable company, of which usually has a source code review process usually.

a person invents a game - they send out the source code to be analyzed and tested for fairness - they get a certification if it is.

Ive also heard that sometimes because of intellectual property they are not allowed to look at source code in some instances - but that could be bull because i cant find a source as of this moment.

now assuming you cant look at the source code from my understanding games are usually required to have some kind of "API" input mechanism that allows a computer to connect to the program and "prod" for an output that can be statistically measured.

assuming you can see the source code - yes everything you said is most likely true but like i said im not a math guy,

and i personally am also assuming that you normally cant see the source code.
DRich
Joined: Jul 6, 2012
• Posts: 6877
Thanks for this post from:
January 21st, 2021 at 7:30:57 AM permalink
Quote: heatmap

now assuming you cant look at the source code from my understanding games are usually required to have some kind of "API" input mechanism that allows a computer to connect to the program and "prod" for an output that can be statistically measured.

assuming you can see the source code - yes everything you said is most likely true but like i said im not a math guy,

and i personally am also assuming that you normally cant see the source code.

I have submitted many games to Nevada Gaming Control for approval and to the best of my knowledge they have never really looked at source code. The source code is just submitted with the game in case any accusations of cheating come up. They will then do a forensic anaysis of the code to try to determine if there was intentional cheating. I did have a bug in one game that caused the machine after free games no longer debited the credit meter so in a certain circumstance players could be playing for free. In that case Gaming Control came to the office and we showed them the source code and what the bug was.

Fortunately it wasn't my bug.
Living longer does not always infer +EV