falcon - 29 January 2012 01:16 PM

[q=guido1] Stating that computer simulations done in seconds with tens of thousands or millions of virtual rolls of the dice are infallible, is folly because:

1.  Computers cannot do random.

Sure they can and they do it as well as any human can.
It was proven way back last century with very primitive computers with the atomic bomb project.

As easy as a human can flip a fair coin so can a computer.
Neither one can do a better job than the other.
A computer can just do many faster than we can.

falcon - 29 January 2012 01:16 PM

2.  Computers programmed with the parameters of the so-called HA for any bets will, in fact, produce the responses based on those mathematical projects during the course of their simulations because the program requires it to do so in order to meet the program specifications.

Now I see your misunderstanding again as was shown in your “hoax” thread.

HA is NOT programmed into computers.

HA is a function (a product) of two simple variables.
1) The probability of an event happening (say the winning probability) and
2) the payoff for that win. 

The pass or come bet in craps HA is -7/495.
That comes from the probability of winning at 244/495 (you love that number)
and a payout of $1 per $1 wagered.

Now if it would pay $1.0287 for a win…we could then see a 0% house edge.
Right?

If the pass line paid $3 for a win instead of just $1 the player would have a very large 97.17% advantage.
But that advantage does not guarantee always winning. That is a function of bet size to bankroll to variance to the number of wagers (action)

falcon - 29 January 2012 01:16 PM

3.  Computers will not make adjustments of simulation problems based on actual table trends or modifications a player might wish to make during sessions played.

Computers WILL make adjustments of simulation problems based on actual table trends or modifications a player might wish to make during sessions played.

yes they do if they are programmed to do. I do it ALL the time in WinCraps and Excel.
It is very easy once you learn how to do it properly.

Now, a computer can not be programmed to bet for a hunch, like a hot blonde walks up to the table and her low cut top is barley hanging on so now the player should bet make a $75 3 way 7 bet and a $25 hard ten.

falcon - 29 January 2012 01:16 PM

4.  Computers do not buy in to a game with $300/$500/$1,000/$5,000 or some finite amount and play toward a goal $$ amount for winning or losing. 
While I know such simulations exist and can be accomplished, such short term results would probably show extremely high loss ratios or busted bankrolls proving only the obvious that a long or “hot” shooter is the only way to make out.

It is nice to program a betting method in a computer, different size bets and different size bankrolls, set if so 1 million players or more play it independently of one another and see the results.

Too many players play the short term, see that they can not lose and think everyone should play this way.

Only by seeing the results of many can we have a good idea of what to expect from one.

falcon - 29 January 2012 01:16 PM

And by the way, isn’t that how it really works, i.e., a limited buy-in and praying to the craps gods for a 45 minute shooter(s) that converts lots of points and the “math” and HA be damned?

falcon/tuttigym/DB+W

You got it.
But Craps players NEVER play just one session in their lifetime. They play many.
Now the math and computer simulations can show us what we can expect to “see” for more than one session.

You are stuck also with the belief that since “perfect distributions” never happen, so the math is really not useful.

Your “hoax” thread at the Wizard of Vegas was in part about 244 wins out of 495 pass line bets has never happened.
Looks like one posted in that thread did an Excel count and showed that it happened very close to expectation in just 35,000 ACTUAL documented dice rolls from the Zumma Craps System Tester (plug)...not computer simulations!

kcadillac - 17 January 2012 04:33 PM

OK….......$27 Pass Line bet.  $3 three-way craps.  $12 six and eight.    All working ont the CO roll.  Examine this closely…...hmmm….not bad.

This looked VERY familiar to me.  It took a while but then I realized that it is similar to one of the hedging bets that John Patrick documented in Advanced Craps. 

$25 PL + $4 three way craps + $22 inside ( place 5,6,8,9)

Winners and losers:
2 + $2
3 + $4
4   - $4
5 + $3
6 + $3
7   - $1
8 + $3
9 + $3
10 - $4
11 + $21
12 + $2

Now this is the GRIND! And very close to kcadillac’s bet.  Set your dice to double sixes and double fives and go to town!

I think a lot of players are missing the point with this type of come out roll set-up.   
    It is not for a player who is just looking to make a profit on the CO roll…...it is for the player who wants to LOSE LESS on the pass line bet and CO roll.  This type of play gets u set to begin your betting series…....with less possibilities of losing money on the CO roll. 
that’s about it man….....as it was shown to me by a heavy right bettor.
good luck.

So you are a two to one favorite to win the pass line, so you hedge the crap out of it, with the purpose of having ” less possibilities of losing money on the CO roll”. You are right, I am missing the point.

guido1:  Thanks for taking the time and effort to respond on my post regarding computers and the simulations.

I have talked with a number of computer professionals, and they have confirmed to me that regarding a game like craps random is not possible.  I paraphrase now - computers are limited to how they are programmed and cannot solve problems that might require information that is not available.  There are diseases that cannot be cured by submitting all the known information we have to a computer to spit out the chemistry needed for a cure.  Math problems are the same.

I stated that the parameters for the HA must be programmed not that the HA itself be programmed.  The distinction for me would be the 495 possibilities of the PL sets along with the 244 ways to win in the pursuit of the perfect set.  I doubt that the program included that the dice must be different colors (one red and one green perhaps) so that the computer would make sure that the a 6 point would include a 4-2 and a 2-4 with in the perfect set. 

Somehow the folks that embrace the above do not agree that the “perfect” set has to be one of each of the 495 PL win/lose possibilities.  They are of the opinion that if two 6 points are converted it makes no difference if one is a 4-2 converted twice, and that is not how the 495 set is made up.  That is why I am incredibly confident that it has NEVER been accomplished in reality.

[q=guido1]:  HA is a function (a product) of two simple variables.
1) The probability of an event happening (say the winning probability) and
2) the payoff for that win.  [/q]

I definitely agree with the above and that is what I was referring to.  Leave those two parameters or requirements out and the computer will not provide the data associated with the set of 495.  In fact, if one does not tell the computer about the set of 495, it might not provide accurate responses to programmed simulations regardless of the system used.

[=guido1]:  Now, a computer can not be programmed to bet for a hunch, like a hot blonde walks up to the table and her low cut top is barley hanging on so now the player should bet make a $75 3 way 7 bet and a $25 hard ten. [/q]

Man I love your sense of humor.

[q=guido1]:  Computers WILL make adjustments of simulation problems based on actual table trends or modifications a player might wish to make during sessions played.

yes they do if they are programmed to do. I do it ALL the time in WinCraps and Excel.
It is very easy once you learn how to do it properly [/q]

I believe I did state that one can program changes or variations of play during simulated play, but again those must be pre-programmed, such as:  Place bet $125 across the board after the point is established; after three rolls and no 7 out; bring down all bets and wait for the point to end and start again.  The problem is that in reality, if a shooter had displayed a propensity of losing or having rolls of five or more tosses, I might play according to his past performance.  The computer cannot do that.

[q=guido1]:  Too many players play the short term, see that they can not lose and think everyone should play this way.

Only by seeing the results of many can we have a good idea of what to expect from one.[/q]

The problem with that theory is it all comes from a simulation which might raise “expectations” of a player where the reality and the bankroll just will not hold out.

I have not seen or read the “Zumma Craps Tester.”  I remain skeptical simply because I am confident that two different color dice were not used and therefore the differentiation of the points were probably not verified.

Again, thanks for taking the time.  I realize my thinking is well beyond the mainstream of craps thought but unless someone can show me how the so-called HA has anything to do with winning or losing a given bet beyond the actual simple mathematical odds, i.e., 36 ways to throw the dice and after the point is established, there are now 30 ways to win and six ways to lose.

Take care,

falcon

falcon - 30 January 2012 12:32 PM

guido1:  Thanks for taking the time and effort to respond on my post regarding computers and the simulations.

I have talked with a number of computer professionals, and they have confirmed to me that regarding a game like craps random is not possible.

I too have talked with a few computer professionals and one of my best friends, who is a computer expert, has even demonstrated how a computer can be as random as we can, even rolling 2 dice.
It is true that some older RNG programs did a horrible job rolling dice, or choosing random numbers. For computers today, it is no problem.

falcon - 30 January 2012 12:32 PM

I paraphrase now - computers are limited to how they are programmed and cannot solve problems that might require information that is not available.  There are diseases that cannot be cured by submitting all the known information we have to a computer to spit out the chemistry needed for a cure.  Math problems are the same.

We are not talking about solving problems, just rolling 2 dice.
My friend would have his computer, using WinCraps, roll the dice for a few shooters.
He would print out the results. I can also do this.
Then we would take 2 new casino dice and play Craps on his actual casino table. He used to work in casinos.
After a dozen rolls or so, we would stop and take a look at the dice in a magnifying glass. WOW!
The edges, that still seemed sharp and pointed showed numerous nicks, scratches, roundness, finger oil prints and he asked if I now trust these dice to produce random rolls better than what the computer did.
Now, what method do you trust. computer or actual dice? One can play Craps both ways.
The point was, a computer can produce better random dice rolls and know which die produced what number way better than real dice on a real craps table because the real dice are subject to wear and contamination much more than the computer’s RNG.
How many Craps players even question the randomness ability of actual casino dice in play?

falcon - 30 January 2012 12:32 PM

Again, thanks for taking the time.  I realize my thinking is well beyond the mainstream of craps thought but unless someone can show me how the so-called HA has anything to do with winning or losing a given bet beyond the actual simple mathematical odds, i.e., 36 ways to throw the dice and after the point is established, there are now 30 ways to win and six ways to lose.

Take care,

falcon

You are absolutely correct. The HA has NOTHING to do with winning or losing one bet. And it never can.
Think of HA as the “short change” policy of every casino.
Bet $1 on the Yo 11 and most casinos pay you 15 to 1. You net $15 and get your $1 back. (16 and down)You are happy!
But, You should have been paid 17 to 1. They ripped you off for $2!
Now, we know that a casino will not pay you exactly what the odds of winning say they should, so it is now up to us to find the best payouts.
The best payouts with variance gives us the best chance to be ahead after many bets, not just one bet.

The longer and longer we all play Craps, because of being “short changed”, we now need more wins to over come the losses to show a profit. And the odds of that keeps getting smaller and smaller as time goes on.

Players Gamble
Casinos are in the business to make money because they can “short change” all players.

guido1 - 30 January 2012 02:09 PM

falcon - 30 January 2012 12:32 PM

guido1:  Thanks for taking the time and effort to respond on my post regarding computers and the simulations.

I have talked with a number of computer professionals, and they have confirmed to me that regarding a game like craps random is not possible.

I too have talked with a few computer professionals and one of my best friends, who is a computer expert, has even demonstrated how a computer can be as random as we can, even rolling 2 dice.
It is true that some older RNG programs did a horrible job rolling dice, or choosing random numbers. For computers today, it is no problem.

A little enlightenment.  A computer requires input in order to make any calculation.  Random functions are generated by input.. often times including clock data takem at the time of the computation.  If we can assume that the clock data is random (as in its a random chance what millisecond will be hitting when the data was input), then a definitely random non-random number has been generated.  Beyond that its just how much nitpicking you want to do.

Southern-Comfort - 30 January 2012 02:52 PM

A little enlightenment.  A computer requires input in order to make any calculation. 
Random functions are generated by input.. often times including clock data takem at the time of the computation.  If we can assume that the clock data is random (as in its a random chance what millisecond will be hitting when the data was input), then a definitely random non-random number has been generated.  Beyond that its just how much nitpicking you want to do.

I agree to this point and I think most will also.

When I code my software programs to have a RNG module, it only requires a one time input once the program starts.
No need to keep “seeding” the RNG. It is now off and running and needs no further input.

The value of knowing the seed value is so we humans can ask the computer to run the same sequence again.
I find that feature more valuable than an ounce of gold.

guido1 - 30 January 2012 02:09 PM

falcon - 30 January 2012 12:32 PM

guido1:  Thanks for taking the time and effort to respond on my post regarding computers and the simulations.

I have talked with a number of computer professionals, and they have confirmed to me that regarding a game like craps random is not possible.

I too have talked with a few computer professionals and one of my best friends, who is a computer expert, has even demonstrated how a computer can be as random as we can, even rolling 2 dice.
It is true that some older RNG programs did a horrible job rolling dice, or choosing random numbers. For computers today, it is no problem.

falcon - 30 January 2012 12:32 PM

I paraphrase now - computers are limited to how they are programmed and cannot solve problems that might require information that is not available.  There are diseases that cannot be cured by submitting all the known information we have to a computer to spit out the chemistry needed for a cure.  Math problems are the same.

We are not talking about solving problems, just rolling 2 dice.
My friend would have his computer, using WinCraps, roll the dice for a few shooters.
He would print out the results. I can also do this.
Then we would take 2 new casino dice and play Craps on his actual casino table. He used to work in casinos.
After a dozen rolls or so, we would stop and take a look at the dice in a magnifying glass. WOW!
The edges, that still seemed sharp and pointed showed numerous nicks, scratches, roundness, finger oil prints and he asked if I now trust these dice to produce random rolls better than what the computer did.
Now, what method do you trust. computer or actual dice? One can play Craps both ways.
The point was, a computer can produce better random dice rolls and know which die produced what number way better than real dice on a real craps table because the real dice are subject to wear and contamination much more than the computer’s RNG.
How many Craps players even question the randomness ability of actual casino dice in play?

falcon - 30 January 2012 12:32 PM

Again, thanks for taking the time.  I realize my thinking is well beyond the mainstream of craps thought but unless someone can show me how the so-called HA has anything to do with winning or losing a given bet beyond the actual simple mathematical odds, i.e., 36 ways to throw the dice and after the point is established, there are now 30 ways to win and six ways to lose.

Take care,

falcon

You are absolutely correct. The HA has NOTHING to do with winning or losing one bet. And it never can.
Think of HA as the “short change” policy of every casino.
Bet $1 on the Yo 11 and most casinos pay you 15 to 1. You net $15 and get your $1 back. (16 and down)You are happy!
But, You should have been paid 17 to 1. They ripped you off for $2!
Now, we know that a casino will not pay you exactly what the odds of winning say they should, so it is now up to us to find the best payouts.
The best payouts with variance gives us the best chance to be ahead after many bets, not just one bet.

The longer and longer we all play Craps, because of being “short changed”, we now need more wins to over come the losses to show a profit. And the odds of that keeps getting smaller and smaller as time goes on.

Players Gamble
Casinos are in the business to make money because they can “short change” all players.

[q=guido1]:  You are absolutely correct. The HA has NOTHING to do with winning or losing one bet. And it never can.
Think of HA as the “short change” policy of every casino.
Bet $1 on the Yo 11 and most casinos pay you 15 to 1. You net $15 and get your $1 back. (16 and down)You are happy!
But, You should have been paid 17 to 1. They ripped you off for $2!
Now, we know that a casino will not pay you exactly what the odds of winning say they should, so it is now up to us to find the best payouts.
The best payouts with variance gives us the best chance to be ahead after many bets, not just one bet.

The longer and longer we all play Craps, because of being “short changed”, we now need more wins to over come the losses to show a profit. And the odds of that keeps getting smaller and smaller as time goes on.[/q]

Finally, I truly understand that the HA is a result of the “short change” on payouts.  Now if the rest of the craps establishment could recognize this fact and then realize that winning and/or losing a given bet has nothing to do with the HA, progress will be made.

I am going to disagree with your last two sentences above simply because walking a way from the table a net winner has cut into the profit of the house, and it is not my mission to “catch up” to the HA you so eloquently defined above.  The HA is the cherry on top of the basic profit made from any player that walks away from the table with less money then he came with.  If a player breaks even, then the HA becomes the basic profit. 

Virtually all agree that craps is a negative expectation game, and the house, in general, wins about 75%-80% of the time in various amounts.  Those gross house wins will ofset the 20%-25% patron wins by about 45% to 55%.  The HA on top will then solidify a table’s profitability.  Of course, there are rare times when a table is unprofitable, but after a time it will catch up simply because the odds are clearly on the side of the house.

falcon

guido1 - 30 January 2012 02:09 PM

Now, what method do you trust. computer or actual dice?

I’ll never be 100 percent certain on that. WinCraps can now figure which strategy I’m using in less than a dozen rolls and deal me the losing combinations. The distribution charts show basically normal results. It’s just that darned timing thing.

Sancho Panza - 02 February 2012 07:49 AM

guido1 - 30 January 2012 02:09 PM

Now, what method do you trust. computer or actual dice?

I’ll never be 100 percent certain on that. WinCraps can now figure which strategy I’m using in less than a dozen rolls and deal me the losing combinations. The distribution charts show basically normal results. It’s just that darned timing thing.

BS!
WinCraps up to version 5.1b Never has had that ability.
(Now he has been working on a multi-player Pro version, not yet ready for public yet)
A few years back he added a few different RNGs and the Mersenne twister. They are all fair.
Steen is no stupid coder.

Lets bet.
1 million dollars.
I will email Steen, he is good at quick replies, as long as he is not on a long flight, and ask him.
For 1 million $$$ he will even show the code if he gets a big cut of the wager!

I doubt you fleas out there could even understand the code if it was in front of your eye balls.

kcadillac - 17 January 2012 04:33 PM

OK….......$27 Pass Line bet.  $3 three-way craps.  $12 six and eight.    All working ont the CO roll.  Examine this closely…...hmmm….not bad.

What about this method is ADVANCED?

$27 Pass Line bet.  Nothing advanced here.

Oh, I get it. All 4 bets at the same time.
You must be proud of yourself.
Next!