Impossible 843 of 3200 wins
Quote: boymimboMath: Given that the experiment went MORE extreme (7 on 41 rolls of 74) when Eliot placed the numbers, there must be software in place to detect this and act accordingly.
Granted, but if they haven't been updating their software in several years, the question is really whether the algorithm they have in there is beatable. That still hasn't been tested by, say, playing doey-don't. Imagine you were playing $5 pass, $5 don't on every roll, you established 4, 5, 8 and 9, and then took odds on 4 and 5 and laid on 8 and 9. Do you really think that the game would roll 8, 9, 7 in order?
Quote: MathExtremistGranted, but if they haven't been updating their software in several years, the question is really whether the algorithm they have in there is beatable. That still hasn't been tested by, say, playing doey-don't. Imagine you were playing $5 pass, $5 don't on every roll, you established 4, 5, 8 and 9, and then took odds on 4 and 5 and laid on 8 and 9. Do you really think that the game would roll 8, 9, 7 in order?
They would roll only 12s and the odds and lay bets would never resolve. I know that's extreme and I'm somewhat joking. I suspect that there might be some ways to beat the casino. Still, I admire the Wizard and the OP for wanting to warn the world rather than exploit the cheating casino.
Quote: MathExtremistGranted, but if they haven't been updating their software in several years, the question is really whether the algorithm they have in there is beatable. That still hasn't been tested by, say, playing doey-don't. Imagine you were playing $5 pass, $5 don't on every roll, you established 4, 5, 8 and 9, and then took odds on 4 and 5 and laid on 8 and 9. Do you really think that the game would roll 8, 9, 7 in order?
If I was coding a rogue system, I'd do a quick calculation of what numbers win which amount for the house, and then tweak the dice probabilities to roll so the roll is in my favour. So there would be no trigger based of the initial pass line bet, but each individual roll. Making it unexploitable.
Quote: thecesspitIf I was coding a rogue system, I'd do a quick calculation of what numbers win which amount for the house, and then tweak the dice probabilities to roll so the roll is in my favour. So there would be no trigger based of the initial pass line bet, but each individual roll. Making it unexploitable.
Me too. I think I would have it determine the winning outcome of every possible roll, including expected value of any bets still on the table. Then, in "cheat mode" it would not accept any outcome with an outcome greater than expected value. If it was, then choose another outcome for the dice, until a "bad" outcome (for the player) is found. Of course, I wouldn't have it in "cheat mode" all the time, just with a certain low probability. Even if I was inclined to cheat, my policy would be to fleece the sheep, not slaughter it.
Quote: MathExtremistGranted, but if they haven't been updating their software in several years, the question is really whether the algorithm they have in there is beatable.
"Where there's smoke, there's fire." So it's usually best to just clear out. And it doesn't take all this long to see the "writing on the wall" if you're at all trying to read it.
What's the best you can hope for by trying to hang around? - More "fire". They're certainly going to only deny, deny, deny; and plead no-contest (and no knowledge of the fact) if, and when, otherwise.
Quote: CrystalMathStill, I admire the Wizard and the OP for wanting to warn the world rather than exploit the cheating casino.
Don't most on-line casinos have strict and legal player- terms and conditions, including no such subterfuge(s)? Even in a casino, i would be wary of inadvertently falling complicit in others' possible crimes.
Quote: WizardEven if I was inclined to cheat, my policy would be to fleece the sheep, not slaughter it.
And of were most reasonable persons (and casinos) not to cheat.
Can a video crap game even be served up in a Class 11 version? Or is this a case of out right crooked software manipulation?
Quote: WizardMe too. I think I would have it determine the winning outcome of every possible roll, including expected value of any bets still on the table. Then, in "cheat mode" it would not accept any outcome with an outcome greater than expected value. If it was, then choose another outcome for the dice, until a "bad" outcome (for the player) is found. Of course, I wouldn't have it in "cheat mode" all the time, just with a certain low probability. Even if I was inclined to cheat, my policy would be to fleece the sheep, not slaughter it.
100% agree. Just keep slowly bleeding the fools who play, rather than destroy them quickly. Just have every 100th roll be non random and that is enough to more than double the house edge, and it is very unlikely anyone would notice. Cheating on half the rolls becomes obvious quickly.
Quote: WizardOf course, I wouldn't have it in "cheat mode" all the time, just with a certain low probability. Even if I was inclined to cheat, my policy would be to fleece the sheep, not slaughter it.
When you have served as a consultant to online software providers, I truly hope this is not something you have suggested for them to implement. Because this is exactly what always happens to me with some major online casino softwares, that is, always running about 2% - 5% below expected return percentage but at the same time being unable to prove it because of the vast amount play data required for 2% to become significant.
Quote: boymimboThe other table game that BLR runs is American Roulette. Does anyone want to try blowing their money there and look for a biased game? Maybe bet on all the numbers except 0/00? Or bet 2 of the 3 streets and see how often 3rd street appears?
They question is would they really bother to cheat on a game which already has >5% house edge and prints money for the casino as it is?
Given the length of time this software has been out there and the lack of complaints out there, it is either not noticed or is little used.
Quote: boymimboNo, it's clear that the BLR Tech software is biased to take a game with a 1.41%/1.36% and turn it into a 50% HA by cheating. I wouldn't be surprised at all if they took a 5.26% game and turn it into a 50% HA the same way. If they were smart they would have turned the 1.41% game into a 5% game where the bias would be impossible to detect, even in 3,200 rolls.
Given the length of time this software has been out there and the lack of complaints out there, it is either not noticed or is little used.
Yeah it is quite shocking actually how long the game has been around. It just goes to show how easy it is to misconsider being cheated as simply bad luck. One player at CM contacted the casino in question as he too had lost and the casino's response was hilarous:
"Unfortunately the people who starts this threads are not people like yourself that accept losing and spread their guilt for their irresponsible gaming in the most irresponsible way."
The answer to that is YES. Its the nature of the person, not the nature of the game. I'm sure that crazy wheel of fortune where you just plop your money down on a representation of a 1,5, 10, 20, etc. dollar bill with a 30 percent house edge is going to be gaffed if they can gaff it.Quote: Jufo81They question is would they really bother to cheat on a game which already has >5% house edge and prints money for the casino as it is?
I too would be a great believer in "fleecing the sheep" for a sustained income over a greater period of time but most thieves have different values and would probably gaff everything in sight. As well as cheating the programmer who did it for them!
On-Edit: Please don't get me wrong. I realize that most of the people who would do such things are smart and shrewd and certainly know that it would be better to fleece than slaughter. Its just that it is not their nature to do it. American roulette already providing a 5.26 percent house edge would indeed be gaffed. It would never occur to them to not gaff it.
I tried the game to check this out. I didn't have the bankroll or time to conduct a lengthy experiment. I tried two experiments. First, I bet "First 12" and "Second 12" to see if "Third 12" came out abnormally high. It didn't. Then I bet "Red" to see if "Black" came out abnormally high. It didn't.Quote: boymimboThe other table game that BLR runs is American Roulette. Does anyone want to try blowing their money there and look for a biased game? Maybe bet on all the numbers except 0/00? Or bet 2 of the 3 streets and see how often 3rd street appears?
Again, I didn't play enough to get a sample that allows me to say the game is likely fair, but I also didn't see any obvious bias.
The way I understand the math, it will always take more time and effort to pronounce something as biased because there is always the possibility that some very negative and even slightly suspicious numbers are really just part of "random".Quote: teliotAgain, I didn't play enough to get a sample that allows me to say the game is likely fair, but I also didn't see any obvious bias.
Perhaps I am being overly negative in saying that a thief would instruct a programmer to "gaff everything in sight" but that is usually the situation. Any investigator or auditor knows that once you start finding "x" as being wrong... you can bet "y" and "z" are wrong also. Its often a bargaining tool. If an attorney settles with one of his complaining clients, then the bar won't send anyone in to look at the attorneys other files as well. Rarely is it that only Robert is getting robbed to pay Paul.
The tests I designed could produce a probable bias in as few as 100 rolls. For example, this morning I wagered simultaneously "Any Red" AND "Any Black" to see if 0 and 00 came up in a statistically abnormal amount. But, nothing funny happened.Quote: FleaStiffThe way I understand the math, it will always take more time and effort to pronounce something as biased because there is always the possibility that some very negative and even slightly suspicious numbers are really just part of "random".
Also, I played a little roulette today, betting only on red. Here are my results before the game crashed:
Wins = 11
Losses = 19
Clearly not enough data to make any conclusions.
Possibly, or your winnings simply hit some secret "trigger point" at which the pre-existing routines started to be activated.Quote: clempops4Everything was normal until all of a sudden I started losing bets. So I think that this program was redesigned sometime this year.
I would fee free to add that the initial post was unclear as to his types of bets and at first unresponsive to questions regarding his data set.
I would say "purporting to show his play".
"What was even more striking is that in every way I could have won, my results were less than expected, and in every way I could have lost, they were more. Following are my overall win/loss results:"
.... I agree. This is indeed even more striking so I would wonder why you are not making this appear more promptly because so few people are likely to wade through the jargon about chi square and [/]degrees of freedom. I'd reverse the order and make your "more striking" point be your first punch ... and then follow up as a second punch with all that temperature stuff or whatever a square degree is.
Quote: FleaStiffI would not claim to have tested the software. You had no access to the software and do not know for certain just what software is actually being used.
Come again? I happen to have a sports betting account at a placing using this software for their casino and played it for hours yesterday. It looked exactly like the game in the videos clempops4 posted. I've got videos to prove it.
Now, I suppose that we're all in error calling this software BLR Technologies. Would you care to give me a little more to go on that we're putting the blame on the wrong company? As evidence it is BLR, check this post by clempops4.
I'm referring to the actual software... which to me implies access to the actual computer code.
I would simply avoid saying that you tested software when you yourself did not have access to the underlying code. Perhaps "tested the operation of the site". To me, testing software means that you yourself had access to the computer code.
Quote: FleaStiffI would simply avoid saying that you tested software when you yourself did not have access to the underlying code. Perhaps "tested the operation of the software". To me, testing software means that you yourself had access to the computer code.
I guess we just have a difference of opinion there. To me losing 75% of pass line bets in a sample of 328 bets is all the evidence I need. Meanwhile, I invite you to play there with your own money, since I don't have access to the source code, and am so out of line in my warning.
It sure is!
>and am so out of line in my warning.
I said nothing about it being "so out of line". You asked for comments. I suggested a more restrained statement that does not suggest you had access to the underlying computer instructions which is what, to me at least, is meant by testing the software.
Its clear that the site is giving invalid results that cheat the player.
Its not quite so clear to me from the original poster's initial data file but it is clear from your table setting forth the results of your test.
I don't think anyone who reads the review is ever going to play there with his own money.
"tested the software", according to Wizard: "I ran it to see how well it approximated real dice."
"tested the software" according to Fleastiff: "I looked at the code to see what was there."
To a casual gambler looking for advice, I think Wiz's is fine. I don't think anyone would assume he had access to the code.
Quote: MoscaTo a casual gambler looking for advice, I think Wiz's is fine. I don't think anyone would assume he had access to the code.
Thank you.
Quote: Jufo81Perhaps: "Tested the software" -> "Tested the game". I think that would be right.
Point taken. I prefer the word "software" because the same company supplies software to five different casinos. The word "game" is too vague. I'm trying to say that the game was not programmed correctly, and the results do not conform to statistical norms. The problem goes to the root source of the code and I think the word "software" expresses that better.
While I thank all for their opinions, I have to disagree with FleaStiff that I must examine the source code to say I tested the software. As if they are going to show it to me.
I hardly ever look at the source code.
If that helps :)
I hardly ever show the testers the source code (not that they would want or need to see it).
Quote: kpI write software for a living.
I hardly ever show the testers the source code (not that they would want or need to see it).
Very occasionally it's useful for hard core white box testing... I'll go and get it on some high risk security or key algorithm based code, but most of the time the developers do code reviews anyways, so that type of thing is already covered. I think I've gone to the code level about 4 times in this role in the last 18 months, and twice in all my last 8 years before that. And in this role it's only cos the code is easier and the dev trust me as I write automated test cases, so they know I have half an idea of what code -does- :)
Quote: FleaStiffNo, I'm not referring to the name of the entities involved.
I'm referring to the actual software... which to me implies access to the actual computer code.
I would simply avoid saying that you tested software when you yourself did not have access to the underlying code. Perhaps "tested the operation of the site". To me, testing software means that you yourself had access to the computer code.
There are two kinds of software testing. White-box is when you look at the code and analyze and profile it, subject it to unit tests, etc. Black-box is when you run it and evaluate its behavior according to external inputs and expected success and failure outcomes. What Wiz did was black-box testing.
Quote: WizardHere is a preview of my BLR Technologies warning on my Odds site. I have not announced it yet, but plan to later today. First, I would like to give the community here a chance to peer review the math and let me know if you find any typos.
This looks great, Mike. Here's a few thoughts.
(1) The software is still definitely in use at World Wide Wagering and Legends. I tested it myself to see that the screen information is the same. At Heritage, I could not get in to view as I don't have an account there. At 5 Dimes and Loose Lines, it appears that they have moved to different software but I can't verify. I am not sure whether the Casino City is up to date, but you've covered yourself well by stating that another site is claiming that these sites run the software.
(2) I also test software for a living, and when it is buggy, I look at the source code (as I have access to it), as it my job to uncover bugs and fix them. That said, you are indeed TESTING the software as you quote.
(3) You might want to put up TELIOT's striking test where he 7ened out immediately after placing all the numbers after a pass bet (on 41 of 74 occasions).
---
For my part, I hope that Clempops, Teliot, and you are making efforts to get your money back. For my part, I'm posting each day on SBR.com (the leading sportsbook forum site) and this is getting notice on CasinoCity. Why I'm doing this I have no idea. I'd love to keep abreast of this.
For me, it's a sad day for online betting. I realize that there was an era where online casinos were cheating people, but I had thought that a casino was simply making enough money by existing and the house advantage. This clearly is not true, and it puts all online casinos and their claims into question. This software's cheatware is so obvious however.
Wizard and TELIOT, you both have interests in a couple of these casinos.
Teliot, you did a statement of fairness for 5 dimes, link here. However, this applies only to the Bonus Casino and not the Cash Back casino. Wizard, I think you did one as well dates back in 2003 but I could not find it.
Quote: 4ofaKind... presently being used at New York Rancinos and other Indian Casinos; that provide outcomes (including video poker) based on pre-set house hold settings? We know Vegas and Atlantic City use Class 111 video poker machines based on 100% random draw.
What about bodog's more-basic attempted circumventions of the Canadian criminal code by "placing" servers there but on Indian Reserves? Story at http://www.osga.com/artman/publish/printer_6376.shtml .
I was never asked to perform, and never conducted an audit of the Cash Back Casino. My certification is for the Bonus Casino software, which is the exclusive platform I was asked to audit. I believe this fact is clear from the 5Dimes website.Quote: boymimboTeliot, you did a statement of fairness for 5 dimes, link here. However, this applies only to the Bonus Casino and not the Cash Back casino.
I would not like my results posted. I did not record them on video so they are not defensible. Also, I cannot refer back to a video to double check my work.Quote: boymimbo(3) You might want to put up TELIOT's striking test where he 7ened out immediately after placing all the numbers after a pass bet (on 41 of 74 occasions).
For the record, here are my results:
| Roll | Come out | Roll #1 | Roll #2 | Roll #3 | Roll #4 | Roll #5 | Roll #6 | Roll #7 |
|---|---|---|---|---|---|---|---|---|
| 1 | 9 | 8 | 7 | |||||
| 2 | 4 | 4 | ||||||
| 3 | 7 | |||||||
| 4 | 9 | 7 | ||||||
| 5 | 7 | |||||||
| 6 | 9 | 7 | ||||||
| 7 | 8 | 11 | 9 | 7 | ||||
| 8 | 11 | |||||||
| 9 | 4 | 7 | ||||||
| 10 | 6 | 3 | 6 | |||||
| 11 | 7 | |||||||
| 12 | 4 | 7 | ||||||
| 13 | 11 | |||||||
| 14 | 8 | 7 | ||||||
| 15 | 6 | 7 | ||||||
| 16 | 8 | 6 | 7 | |||||
| 17 | 11 | |||||||
| 18 | 6 | 7 | ||||||
| 19 | 9 | 8 | 3 | 5 | 5 | 6 | 10 | 9 |
| 20 | 6 | 7 | ||||||
| 21 | 11 | |||||||
| 22 | 8 | 4 | 7 | |||||
| 23 | 3 | |||||||
| 24 | 5 | 7 | ||||||
| 25 | 9 | 7 | ||||||
| 26 | 6 | 7 | ||||||
| 27 | 8 | 7 | ||||||
| 28 | 3 | |||||||
| 29 | 2 | |||||||
| 30 | 4 | 4 | ||||||
| 31 | 7 | |||||||
| 32 | 5 | 7 | ||||||
| 33 | 5 | 10 | 7 | |||||
| 34 | 4 | 7 | ||||||
| 35 | 5 | 8 | 9 | 2 | 9 | 4 | 9 | 7 |
| 36 | 8 | 7 | ||||||
| 37 | 4 | 7 | ||||||
| 38 | 2 | |||||||
| 39 | 5 | 11 | 7 | |||||
| 40 | 11 | |||||||
| 41 | 5 | 7 | ||||||
| 42 | 2 | |||||||
| 43 | 10 | 7 | ||||||
| 44 | 9 | 7 | ||||||
| 45 | 5 | 11 | 3 | 5 | ||||
| 46 | 8 | 7 | ||||||
| 47 | 9 | 6 | 7 | |||||
| 48 | 8 | 6 | 2 | 7 | ||||
| 49 | 4 | 5 | 9 | 7 | ||||
| 50 | 3 | |||||||
| 51 | 9 | 7 | ||||||
| 52 | 5 | 7 | ||||||
| 53 | 12 | |||||||
| 54 | 4 | 7 | ||||||
| 55 | 6 | 4 | 7 | |||||
| 56 | 12 | |||||||
| 57 | 9 | 8 | 8 | 7 | ||||
| 58 | 9 | 7 | ||||||
| 59 | 7 | |||||||
| 60 | 8 | 7 | ||||||
| 61 | 10 | 6 | 12 | 8 | 11 | 6 | 12 | 7 |
| 62 | 6 | 9 | 7 | |||||
| 63 | 11 | |||||||
| 64 | 8 | 6 | 7 | |||||
| 65 | 9 | 9 | ||||||
| 66 | 7 | |||||||
| 67 | 9 | 7 | ||||||
| 68 | 6 | 7 | ||||||
| 69 | 6 | 5 | 9 | 6 | ||||
| 70 | 6 | 7 | ||||||
| 71 | 8 | 2 | 8 | |||||
| 72 | 6 | 8 | 7 | |||||
| 73 | 4 | 10 | 2 | 3 | 7 | |||
| 74 | 10 | 11 | 4 | 7 | ||||
| 75 | 7 | |||||||
| 76 | 8 | 7 | ||||||
| 77 | 9 | 7 | ||||||
| 78 | 9 | 6 | 4 | 7 | ||||
| 79 | 7 | |||||||
| 80 | 5 | 7 | ||||||
| 81 | 12 | |||||||
| 82 | 9 | 8 | 10 | 7 | ||||
| 83 | 10 | 7 | ||||||
| 84 | 11 | |||||||
| 85 | 4 | 9 | 5 | 8 | 4 | |||
| 86 | 12 | |||||||
| 87 | 6 | 8 | 6 | |||||
| 88 | 6 | 7 | ||||||
| 89 | 3 | |||||||
| 90 | 8 | 7 | ||||||
| 91 | 8 | 8 | ||||||
| 92 | 9 | 7 | ||||||
| 93 | 10 | 12 | 7 | |||||
| 94 | 4 | 7 | ||||||
| 95 | 5 | 7 | ||||||
| 96 | 9 | 5 | 5 | 5 | 6 | 6 | 7 | |
| 97 | 4 | 7 | ||||||
| 98 | 8 | 7 | ||||||
| 99 | 8 | 7 | ||||||
| 100 | 9 | 7 | ||||||
| Total 7s: | 8 | 41 | 11 | 7 | 1 | 0 | 1 | 2 |
| Total rolls: | 100 | 74 | 29 | 15 | 6 | 4 | 4 | 3 |
1. 5Dimes just shut it down, and issued this statement:
Quote: 5DimesWith the present accusations and facts against the BLR casino platform, the 5Dimes group has chosen to remove this casino platform from our multiple casino lineup.
BLR was given ample time to address the concerns of forum members as well as our direct questions. No acceptable answers were given, so the casino platform was removed.
A management decision was made to eliminate that casino and focus on expanded offerings with the other casino platforms.
To that, I say "bravo," well done 5Dimes.
I would be interested to know if any other casinos remove their BLR-driven casino.
2. If it was my job to fix the problem, indeed, I would ask to see the source code. However, I feel my job as a player advocate is completed by just demonstrating the game is not playing fairly, I don't need to know exactly how or why.
3. I thought about it. I didn't want to overwhelm the reader with a lot of math, and I do indicate the link.
4. 5Dimes is where I played and they did refund the $166 I lost.
5. Yes, I did an audit of 5Dimes back in 2003, of another brand of software, which I deemed to be fine. I'm happy to see 5Dimes did the right thing and closed their BLR casino. They are a reputable organization and I'm sure had no knowledge of what was going on until the story broke here.
From the response from 5 Dimes, it appears that the blame is on the software provider and not the sportsbook itself. I suspect that they farm out the casino to BLR and they have a revenue sharing arrangement.
Quote: boymimboThe cash back casino at 5 Dimes appears to still be up. Anyone want to confirm?
When I tried to log in I got a message that my IP was restricted. It is possible they can easily just swap in another platform. Can you comment on whether the craps game looks like the image in my BLR warning?
I guess I should have gone to WikiPedia or something like that before suggesting that the use of the term "software testing" requires access to the actual computer code. Testing expected outputs against actual outputs without ever seeing a line of code does appear to be known as a particular type of software testing known as Black Box Testing.Quote: MathExtremistThere are two kinds of software testing. White-box is when you look at the code and analyze and profile it, subject it to unit tests, etc. Black-box is when you run it and evaluate its behavior according to external inputs and expected success and failure outcomes. What Wiz did was black-box testing.
Quote: FleaStiffI guess I should have gone to WikiPedia or something like that before suggesting that the use of the term "software testing" requires access to the actual computer code. Testing expected outputs against actual outputs without ever seeing a line of code does appear to be known as a particular type of software testing known as Black Box Testing.
Correct, in my world, we have unit testing, where the developers test the code; QA Testing (testing for quality assurance), or system testing which tests the direct functionality, integration testing, which tests the software with the applicable parts of the rest of the suite, and user acceptance testing, where the end users test the software.
The code is really only revealed during the user tests. The code is usually found in the technical specifications but is only reviewed by developers and QA. I usually take a good look at the coding as well even though I'm further down the test chain as I tend not to trust my developers as they are usually offshore and don't understand the functional requirements.
Take a bow, man!!!
-Ms. D.
Will ClemPops get his money back?
Will I return to my day job?
Will the other four casinos turn off their software?
Will BLR Tech go out of business?
And what about Naomi?Quote: boymimboThe question is, now that the proof is out there and the Wizard has confirmed,
Will ClemPops get his money back?
Will I return to my day job?
Will the other four casinos turn off their software?
Will BLR Tech go out of business?
Quote: Wizard
4. 5Dimes is where I played and they did refund the $166 I lost.
.
I had not known about this issue til someone posted on SBR about it. I encountered this back in August and posted about it on SBR.
Under my handle mighty maron I encountered the craps game at the rebate/cash back casino. I got killed fast for at least 100.
To be fair, the bonus casino craps I have made more than I have lost. The rebate casino was instant death for craps for me
TY for finding this out, I emailed about losses
Quote: boymimboAnd who is Naomi?
Some say she is the mother of Jake and Maggie Gyllenhaal...
On the original Electric Company, there was a segment called Love of Chair which ended with a series of questions, the last of which was always "And what about Naomi?" There was a Naomi working on the show who went on to give birth to the actors.
