I have said this for over a year, and it's this fact that led me to create the balance I have.

But at the same time, there is some support to the claims that Harley makes. That's about all as far as I am concerned.

I think Harley and a lot of the stuff he publishes actually does hurt the casino's image; and I think that this is unfortunate.

But I also think that casinos could do a better job assuring people that the dice are fair, too.

So that's about it.

99.9% already assume the dice are fair, so why should a casino waste their time and money ?

Quote:Harley

Chicken Feed Random Roll from Straight out on 12 foot table:

Face ... Total ... Less 5 ... Squared

.. 1 ........ 9 ........ 4 ........... 16

.. 2 ........ 2 ........ -3 ........... 9

.. 3 ........ 3 ........ -2 ........... 4

.. 4 ........ 4 ........ -1 ........... 1

.. 5 ........ 10 ....... 5 ........... 25

.. 6 ........ 2 ........ -3 ........... 9

Totals .... 30 ........ ........... 64

I am nothing close to a math person or statistics expert, but 30 rolls? I can't imagine 30 rolls is proof of anything. In card simulations we're usually talking about 1,000,000 hands or more. I think I could roll a perfectly crafted, world championship balanced die 30 times and come up with the numbers you show above. That's a horrible sample size.

ZCore13

Quote:FatGeezusI was at a table where the shooter was throwing the dice very hard. They would bounce all the way back to where he was shooting from.

The stickman asked "WHAT THE HELL ARE YOU TRYING TO DO, KNOCK THE SPOTS OFF!!!"

No, he's trying to win the bonus award. Don't you know about the bonus award?

Quote:New Jersey Casino Control Commission Regulations, Chapter 19:47-1.8 Throw of the diceUpon selection of the dice, the shooter shall make a Pass or Don't Pass Bet after which he shall throw the two selected dice so that they leave his hand simultaneously and in a manner calculated to cause them to strike the end of the table farthest from him. If the dice, having been thrown in such a manner, return to the shooter's hand without first touching the layout, the shooter's Pass or Don't Pass Bet shall be paid at odds of 25 to 1.

http://www.state.nj.us/casinos/actreg/reg/docs_chapter47/c47s01.pdf

... Yes, I'm kidding about that last part. But imagine what the dice pit would be like if that were an actual rule...

Only 2.5% of the time would your squared differences be the same or worse than those shown.Quote:Zcore13Quote:Harley

Chicken Feed Random Roll from Straight out on 12 foot table:

Face ... Total ... Less 5 ... Squared

.. 1 ........ 9 ........ 4 ........... 16

.. 2 ........ 2 ........ -3 ........... 9

.. 3 ........ 3 ........ -2 ........... 4

.. 4 ........ 4 ........ -1 ........... 1

.. 5 ........ 10 ....... 5 ........... 25

.. 6 ........ 2 ........ -3 ........... 9

Totals .... 30 ........ ........... 64

I am nothing close to a math person or statistics expert, but 30 rolls?

I can't imagine 30 rolls is proof of anything.

In card simulations we're usually talking about 1,000,000 hands or more.

I think I could roll a perfectly crafted, world championship balanced die 30 times

and come up with the numbers you show above. That's a horrible sample size.

ZCore13

30 rolls is fine (the minimum) for the chi-squared test with the minimum expected values all at 5

60 rolls would be finer, but that takes way more work to accomplish.

His 64/5 = a 12.8 critical value and it does FAIL at the 5% level. 1 in 20

FAILS at 1 in 20!

CNN Headline news

Not the 2.5% level.

The p-value = 2.53% or about 1 in 40

30 roll samples of a perfectly fair die would return

the same results or worse from observed values to expected values.

Harley is convinced that the data he has shown here,

and more data kept secret so not to reveal the combo to the bank vault,

IS the proof needed to at least a 99% significance level. (he stated that in a different post)

Excellent.

Beyond a reasonable doubt

at 1 in 40

What does it look like on a chart?

To the right of the red line is all it is.

Not that rare of an event.

1 in 40 is the same as 25,000 out of 1 million.

Those that want to see something NOT from Excel

The casinos that are knowingly using biased dice at their Craps tables better run and hide those dice.

Their day of reckoning is at hand

Dammit! That was a secret!!!Quote:SOOPOOSo Giorgio has found the secret to beating craps! Throw the dice really hard and bet the don't for a sure win! Thanks!

Quote:MathExtremist

... Yes, I'm kidding about that last part. But imagine what the dice pit would be like if that were an actual rule...

That was well and truly hilarious!

Not by the standards of Harley.Quote:KeyserThe Chi Square (Chance of Randomness) testing is a waste of time on such small trials.

Harley is just one of the experts from a team of experts.

he says throwing the dice hard produces more random rolls.

"throw it hard and you will not fall victim to the imbalance so often"

His provided an excellent sample of 30 dice rolls and his chi-squared test.

I can also point out he did not follow all the normal 4 conditions that others do (he did not follow #2)

The chi-square goodness of fit test is appropriate when the following conditions are met:

- The sampling method is simple random sampling.
- The population is at least 10 times as large as the sample.
- The variable under study is categorical.
- The expected value of the number of sample observations in each level of the variable is at least 5.

Back to throwing those dice hard and fast

(not slow and wimpy)