MathExtremist
Joined: Aug 31, 2010
• Posts: 6526
April 1st, 2013 at 10:35:48 AM permalink
Quote: FrankScoblete

thecesspit, that is a great way to look at it --- by changing the probabilities with a controlled throw, you set up a new set of random results. (Yes, you do have to have the ability to do this.)

Now, one step further. Since the new probabilities are being paid off by the casino as if they were the old probabilities, the player edge is established because of that. For example, if the six is being hit more than 5 times out of 36 (just for an example, say 6 times per 36 rolls), but is still being paid 7 to 6 on a place bet (the house mistakenly thinking it has manipulated the payout to give itself the edge) then the player now has the edge.

New random probabilities. Old payouts. Possible player edge. This is establishing a new set of probabilities; the casino lagging with an old set of payments.

I hope this isn't news to you. It's obvious that changing the die face distribution away from uniform (p=1/6 for each) leads to new EVs for each craps bet. If that wasn't the basis of your prior books, what was?

Note that increasing the frequency of the six isn't sufficient to establish a player edge on the place 6 bet. Increasing the frequency of six vs. seven is what's required. If you increase both the frequency of six *and* seven at the same ratio, the edge on the place 6 bet doesn't change. Similarly, if you decrease the frequency of seven (seemingly the goal of the SRR measurement) but also decrease the frequency of six by the same amount, you haven't done yourself any favors vis-a-vis the place 6 bet.

That's why SRR is a useless measurement. SRR doesn't provide sufficient information to compute the altered edge on any bets other than the Any 7 bet (which you shouldn't really care about anyway) -- and therefore doesn't provide enough detail on how to bet under the assumption that you have the ability to alter the probabilities. I'm not opining on whether that's actually possible in a casino setting, but that's unrelated to the discussion of how to bet under the assumption you can.
"In my own case, when it seemed to me after a long illness that death was close at hand, I found no little solace in playing constantly at dice." -- Girolamo Cardano, 1563
Ahigh
Joined: May 19, 2010
• Posts: 5152
April 1st, 2013 at 11:12:59 AM permalink
Quote: MathExtremist

I hope this isn't news to you. It's obvious that changing the die face distribution away from uniform (p=1/6 for each) leads to new EVs for each craps bet. If that wasn't the basis of your prior books, what was?

Note that increasing the frequency of the six isn't sufficient to establish a player edge on the place 6 bet. Increasing the frequency of six vs. seven is what's required. If you increase both the frequency of six *and* seven at the same ratio, the edge on the place 6 bet doesn't change. Similarly, if you decrease the frequency of seven (seemingly the goal of the SRR measurement) but also decrease the frequency of six by the same amount, you haven't done yourself any favors vis-a-vis the place 6 bet.

That's why SRR is a useless measurement. SRR doesn't provide sufficient information to compute the altered edge on any bets other than the Any 7 bet (which you shouldn't really care about anyway) -- and therefore doesn't provide enough detail on how to bet under the assumption that you have the ability to alter the probabilities. I'm not opining on whether that's actually possible in a casino setting, but that's unrelated to the discussion of how to bet under the assumption you can.

When we were talking about the dice being biased, I made an excel spreadsheet to just illuminate the modified edges for various unequal distributions of face outcomes from one through sixth.

That being said, I have no evidence that the faces that appear can be controlled at all from my own throws.

I would love a demonstration of this, though.

Yesterday at Red Rock, I heard a disgusted player leaving after losing a grand saying "yeah and all the sevens were 6-1 seven outs."

No doubt word has traveled about the theory of biased dice.
FrankScoblete
Joined: Mar 27, 2013
• Posts: 436
April 1st, 2013 at 11:17:16 AM permalink
It's how he said it and the coupling of old payoffs for new probabilties.
7craps
Joined: Jan 23, 2010
• Posts: 1977
April 1st, 2013 at 11:26:21 AM permalink
Quote: FrankScoblete

thecesspit, that is a great way to look at it ---
by changing the probabilities with a controlled throw, you set up a new set of random results.

But is it not also true that
every set of 36 rolls by anyone at any time has a new set of random results.

And over many sets of 36 the probabilities of each outcome is converging to a certain value, with or without bias.
if there are less 7s what are there less of and more of in the other #s?

Variance in a multinomial distribution.

You and I may not understand this concept but ME sure does.
winsome johnny (not Win some johnny)
Ahigh
Joined: May 19, 2010
• Posts: 5152
April 1st, 2013 at 11:30:05 AM permalink
My software gives me sets to use that raise the expected # to seven ratio for all 10 outcomes besides seven.

I then have charts when aiming for two sets of numbers with fewer sevens.

The only time I have ever been banned from shooting was while I was with Teddy consulting with this chart to change my set on each throw.

I had a long roll, but it wasn't anything super duper special.

But the ratio to sevens is what my software looks at .. fewer sevens and MORE of what you want to hit .. both.
AlanMendelson
Joined: Oct 5, 2011
• Posts: 3866
April 1st, 2013 at 12:03:56 PM permalink
Quote: MathExtremist

That's why SRR is a useless measurement. SRR doesn't provide sufficient information to compute the altered edge on any bets other than the Any 7 bet....

I agree with MathExtremist about the value of the SRR statistic. My added concern is that someone might be avoiding the seven but all of their other throws are numbers you are not betting on, for example horn numbers, or 4s and 10s while you are betting the inside.

If you are really going to test for dice influencing it has to be a test of the success rate for passes (or don't pass for a darkside player) and for hitting the numbers that are bet.
wrighj03
Joined: Mar 25, 2013
• Posts: 7
April 1st, 2013 at 7:57:27 PM permalink
Quote: AlanMendelson

I agree with MathExtremist about the value of the SRR statistic. My added concern is that someone might be avoiding the seven but all of their other throws are numbers you are not betting on, for example horn numbers, or 4s and 10s while you are betting the inside.

If you are really going to test for dice influencing it has to be a test of the success rate for passes (or don't pass for a darkside player) and for hitting the numbers that are bet.

May I point out that if we collectively spent more time throwing dice than complaining about odds we could all be gtc players. Srr isn't as important it is a metrics to base decisions and examine performance. Like pitch and axis control.
Ahigh has the right idea.
Also when I changed my grip I had better results but lost axis control in one die?
The three biggest things I noticed release angle, grip and speed changed my overall results.
AlanMendelson
Joined: Oct 5, 2011
• Posts: 3866
April 1st, 2013 at 9:15:31 PM permalink
But Ahigh has no pitch or axis control.
thecesspit
Joined: Apr 19, 2010
• Posts: 5936
April 1st, 2013 at 11:47:56 PM permalink
Quote: FrankScoblete

thecesspit, that is a great way to look at it --- by changing the probabilities with a controlled throw, you set up a new set of random results. (Yes, you do have to have the ability to do this.)

You set up a new probability distribution. And as MathExtremist says, you need to know all the numbers to work on it.

Quote:

If I am right in moving further on the assumption you created, I'd like to use this in an article or in a book. I'd give you full credit either handle or real name --- your choice.

Just let me know.

Sure, but I'm not sure I've said anything of great insight. I've edited it so I took out my ham-fisted tablet-typing.

I slightly disagree with MathExtremist of Rolls to Seven Ratios : It is useful as a testable claim. And if a player does have a 6.3 RSR, over X thousand rolls to give a high level of confidence, it does indicate they should look at what other numbers could be profitable. And of course, you can use software to do that, or use all the rolls to calculate your own personal probability function for chucking dice. And of course, RSR of 6, could still mean an advantage on one number (if I roll 6 6/36 times and 8 4/36 times, for example).
"Then you can admire the real gambler, who has neither eaten, slept, thought nor lived, he has so smarted under the scourge of his martingale, so suffered on the rack of his desire for a coup at trente-et-quarante" - Honore de Balzac, 1829
7craps
Joined: Jan 23, 2010
• Posts: 1977
April 2nd, 2013 at 12:10:42 AM permalink
Quote: thecesspit

You set up a new probability distribution. And as MathExtremist says, you need to know all the numbers to work on it.

Well, this is one thing that Smart Craps program does automatically. Even in the shareware version.
From the Pro Test Player Edge section

While not immediately obvious, it is possible to convert Pro Test results into game expectations. To do so, you need to know:
· The rules of the craps game you are playing in, such as the odds and pay schedule.
· The specific bets you are going to make.
· The dice sets you will use at each point in the throwing cycle.
Smart Craps has two ways for you to determine your edge in the game:
1) Exact mathematical analysis for given simple bets (such as pass, or don't pass, with or without odds).

Converting Pro Test results to exact player expectation
Normally for a random shooter, each of the 36 dice outcomes has exactly 1 in 36 chance of occurring.
For dice setters, the odds will vary for each potential outcome in a predictable manner.
Once we have a combination of Pro Test results (a shooter may only pass one or two tests, or possibly all three),
these can be converted into specific probabilities for each of the 36 outcomes.

I did not pass one of the 3 test the first few times I did try
but the program did say I had a small edge on a few bets with the results I entered.
I have not tried any new rolls to see different results.
winsome johnny (not Win some johnny)