CrystalMath
CrystalMath
Joined: May 10, 2011
  • Threads: 8
  • Posts: 1881
October 31st, 2011 at 10:34:36 AM permalink
I just reviewed video 7 - "Why internet gambling should be regulated May 20th #7"

There were 25 winners. Three wagers were accidentally placed as don't pass, and incidentally, they all three were winners. Clem says in the video that he won 22, but I think it's due to the pass/don't pass discrepancy. Also, his credit balance drops 50 credits from start to end, so 25 winners it is.


bet type outcome
pass w 7
pass l 3
pass l 12
pass l 3
pass l 12
pass l 6 11 4 4 9 3 7
donít w 4 3 8 7
pass l 3
pass w 7
pass l 8 7
pass w 10 3 10
pass l 9 6 7
pass l 6 7
pass w 11
pass l 2
pass l 4 7
pass l 9 4 8 6 5 2 7
pass l 12
pass l 9 8 6 6 7
pass w 8 10 6 10 10 3 5 6 6 10 8
donít w 5 9 7
pass l 10 7
pass l 2
pass l 5 9 7
pass l 4 2 7
pass l 5 7
pass l 2
pass l 2
pass w 4 10 3 4
pass l 8 12 3 3 7
pass w 9 9
pass l 12
pass w 7
pass l 5 7
pass l 6 3 9 7
pass l 3
pass l 3
pass l 10 9 8 5 5 4 6 7
pass l 8 7
pass w 8 8
pass l 5 8 4 7
pass l 8 9 10 6 6 6 6 7
pass l 10 7
pass l 8 7
pass l 6 5 7
pass l 8 9 3 9 6 7
pass l 4 6 7
pass l 9 6 7
pass l 8 5 7
pass w 9 11 9
pass w 5 5
pass l 9 3 10 8 11 7
pass w 4 9 9 5 4
pass l 5 4 12 7
pass l 2
pass l 9 5 7
pass w 11
pass l 4 7
pass w 4 6 4
pass w 7
pass l 6 7
pass w 11
pass l 3
pass l 6 5 7
pass l 6 8 10 7
pass l 5 4 7
pass l 2
pass l 4 9 10 7
pass l 8 7
pass l 10 7
pass l 8 3 11 6 6 10 9 4 4 7
pass l 5 6 7
pass l 8 7
pass l 8 7
pass l 9 4 2 6 8 4 3 10 6 7
pass w 11
pass l 8 11 7
pass l 6 8 3 7
pass w 11
pass l 6 8 11 3 3 7
pass l 3
pass l 8 6 2 6 2 7
donít w 9 8 10 7
pass l 8 12 6 10 11 9 7
pass l 2
pass l 3
pass l 5 8 9 2 10 7
pass l 12
pass w 6 6
pass l 4 7
pass l 9 11 7
pass l 6 8 8 7
pass l 5 11 11 10 8 8 3 9 9 7
pass w 7
pass w 7
pass l 8 7
pass l 9 5 3 6 3 11 3 6 6 7
pass l 5 8 7
pass w 8 8
pass l 4 6 6 7
I heart Crystal Math.
DorothyGale
DorothyGale
Joined: Nov 23, 2009
  • Threads: 40
  • Posts: 639
October 31st, 2011 at 10:39:56 AM permalink
Quote: CrystalMath

I just reviewed video 7 - "Why internet gambling should be regulated May 20th #7"


At what point do we feel comfortable declaring this software rogue? I am comfortable with it ... it's not huge, I mean, it's nothing like the UB/AP scandal, and the malware is so obvious ... but it should be enough to get this vendor (and all the casinos that use their software) put on a rogue list, like that at Casinomeister, don't you think?
"Who would have thought a good little girl like you could destroy my beautiful wickedness!"
MathExtremist
MathExtremist
Joined: Aug 31, 2010
  • Threads: 88
  • Posts: 6526
October 31st, 2011 at 10:44:48 AM permalink
Quote: boymimbo

My theory is that: come out roll is rigged to replace 7s and 3s with 12s (from time to time) and to leave point numbers the same.
Once the point is established, 7s are rigged to appear much more frequently.


Clempops4 is absolutely right in that it is possible to rig gaming software to make a player lose when the site wants that to happen. Several of my earliest clients used a "risk mitigation" algorithm to do exactly this.

However, it's very, very, very dangerous to do with craps because you can bet on both sides. If you can discover the pattern in how the game is tweaked to be unfair (and boymimbo may be right above), then it should be possible to place small bets to trigger that unfair distribution and then place large bets to take advantage of it after the comeout roll.

Here's one, just to get started: make a standard-amount bet on the passline. Then, after a point is established don't take any odds but instead make a lay bet on the point for 5x or more of the line bet amount. If the game stays rigged in the same way, you'll absolutely clean up.
"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
boymimbo
boymimbo
Joined: Nov 12, 2009
  • Threads: 17
  • Posts: 5994
October 31st, 2011 at 10:47:23 AM permalink
I'd feel quite comfortable with rogueing this casino and all those that use their software, as it's obvious that the casino is cheating given the non-random results, especially when the bet turns from don't pass to pass.

I'd like to hear the wizard's thoughts on all of this. Is there any recourse for Clem to recoup his money ($1,500)?
----- You want the truth! You can't handle the truth!
boymimbo
boymimbo
Joined: Nov 12, 2009
  • Threads: 17
  • Posts: 5994
October 31st, 2011 at 10:50:47 AM permalink
I'd love to see the distribution of the dice with a doey-don't (what would the frequency of 12s increase to)? I'd love to see as well what would happen with the solution as described above by Math. If the software is sophisticated enough, it would calculate the rolls needed to make you lose and then come up with a much higher frequency of that roll.
----- You want the truth! You can't handle the truth!
MathExtremist
MathExtremist
Joined: Aug 31, 2010
  • Threads: 88
  • Posts: 6526
October 31st, 2011 at 10:58:07 AM permalink
Quote: boymimbo

If the software is sophisticated enough, it would calculate the rolls needed to make you lose and then come up with a much higher frequency of that roll.


I highly doubt the software is that sophisticated. Here's what I'd do to test it:
1) Play $5 each pass/don't pass without a bet on midnight for, say, 200-300 decisions. Check to see if the distribution changed. There are a few ways it could:
2a) If it favors 12 greater than p=1/30 but keeps everything else even, try another 200 decisions with 5/5 + $1 on midnight.
2b) If it favors one side over the other, try another 200 decisions with 5/5 and just take odds on the favored side.
If 2a or 2b show different results than 1, stop playing - the software is adaptive. If not, play enough to make back your money, cash out, and then file the report. But try to get your money out first...
"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
DorothyGale
DorothyGale
Joined: Nov 23, 2009
  • Threads: 40
  • Posts: 639
October 31st, 2011 at 11:02:04 AM permalink
Quote: MathExtremist

I highly doubt the software is that sophisticated. Here's what I'd do to test it:
1) Play $5 each pass/don't pass without a bet on midnight for, say, 200-300 decisions. Check to see if the distribution changed. There are a few ways it could:
2a) If it favors 12 greater than p=1/30 but keeps everything else even, try another 200 decisions with 5/5 + $1 on midnight.
2b) If it favors one side over the other, try another 200 decisions with 5/5 and just take odds on the favored side.
If 2a or 2b show different results than 1, stop playing - the software is adaptive. If not, play enough to make back your money, cash out, and then file the report. But try to get your money out first...

It could have as simple a trigger as "if the player plays only the pass or don't, then turn on the bias, any other bets, turn the bias off" ... that's not sophisticated ... as the OP noted, it played fair in practice mode and it played fair when he first started ... the on/off switch can be pretty dumb and still make a fortune for the house ...

--Ms. D.
"Who would have thought a good little girl like you could destroy my beautiful wickedness!"
MathExtremist
MathExtremist
Joined: Aug 31, 2010
  • Threads: 88
  • Posts: 6526
October 31st, 2011 at 11:05:40 AM permalink
Quote: DorothyGale

It could have as simple a trigger as "if the player plays only the pass or don't, then turn on the bias, any other bets, turn the bias off" ... that's not sophisticated ... as the OP noted, it played fair in practice mode and it played fair when he first started ... the on/off switch can be pretty dumb and still make a fortune for the house ... .


Only if nobody figures out what it is...
"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
DorothyGale
DorothyGale
Joined: Nov 23, 2009
  • Threads: 40
  • Posts: 639
October 31st, 2011 at 11:12:09 AM permalink
I created a poll about this issue here ... math by democracy ... if only ...
"Who would have thought a good little girl like you could destroy my beautiful wickedness!"
Wizard
Administrator
Wizard
Joined: Oct 14, 2009
  • Threads: 1367
  • Posts: 22724
October 31st, 2011 at 11:15:00 AM permalink
Let's suppose my goal is to independently prove the software is cheating. I don't need to figure out exactly just how it is doing it, I just want to prove the existence of it. Does the alleged cheating seem to be more egregious for the pass or don't pass bettor? Suppose I'm testing just pass results after the COR, which were said to look normal. Would it be a robust test to compare results against the following expectations?


Event Prob
Point 4 win 4.17%
Point 5 win 6.67%
Point 6 win 9.47%
Point 8 win 9.47%
Point 9 win 6.67%
Point 10 win 4.17%
Point 4 loss 8.33%
Point 5 loss 10.00%
Point 6 loss 11.36%
Point 8 loss 11.36%
Point 9 loss 10.00%
Point 10 loss 8.33%
Total 100.00%


Or, do you think some other test would make the case better?
It's not whether you win or lose; it's whether or not you had a good bet.

  • Jump to: