BLR Tech casino software appears to be rogue ...
Poll
 | 26 votes (92.85%) | ||
| No votes (0%) | |||
| No votes (0%) | |||
| | 1 vote (3.57%) | ||
| | 1 vote (3.57%) | ||
| No votes (0%) |
28 members have voted
Split from previous thread "Impossible 843 of 3200 wins"
My opinion is that the software created by BLR Tech is rogue ... I've never seen anything like it ... everywhere you turn there is some little thing (or big thing) that works against the bettor, each individually is highly unlikely, but in combination the evidence is overwhelming ...
The original poster collected data over a long period of time, but unfortunately all of that data is "hearsay" and is not usable without something to back it up, but the wonder of it all is that he videoed himself playing hundreds of hands and posted those to you-tube, with results that are just as unbelievable as all the unusable data he has...
Just amazing ...
--Ms. D.
I ain't no programmer. I'll see what the comments are. I recall when I first read that thread it seemed unsettled as to suspicious results or reasonable results and I don't see why there should be doubt.
If there is some little thing or some big thing that works against the bettor but never any little thing or big thing that works against the casino, then the question is answered right then and there. You can't have simply a bad programmer if all the mistakes are one sided.
Just as I thought, the thread that started out complaining of a lack of raw data and featured an original poster who seemed to not want to respond directly to comments about his data file format eventually grew into a thread wherein the math types did seem to work with the processed file.
I don't know if this "casino" is letting the bettor play on one wheel or what. At one point the Wizard suggested we not inquire as to HOW the player might be cheated but just see if a particular type of bet has a normal response.
I've not followed the math but it seems that this is a question which should be answered definitively even if the data is presented as bet on X but then it later turned out to be Seven Out.
5Dimes Casino and Sportsbook, Heritage Sports, Legend Sports, WorldWide Wagering, and Loose Lines Casino and Sportsbook.
So, before we stop using these five sites, let's confirm that the are indeed still using BLR software before incriminating them.
Quote: IbeatyouracesThis is why I will never wager online no matter what site or who endorses it, sorry wiz. I will stick to bricks and morter casinos and live equipment.
I agree, unless you are a shut in why not save yourself for the real thing? Logistically it seems like a pain too, they have to skirt a lot of grey areas just to to get your money in and a bunch to get your money back to you.
With proper oversight it could be regulated in the US (pray its not the SEC regulating) and done correctly. It might force bookies to get a real job but they aren't paying taxes anyway so f'em!
I used to be a programmer and once had to write a random number generator - for those who are used to writing normal applications that have expected results from given inputs (cf black box testing) you can understand how difficult it is to prove it works. For instance if you order a book online, and get the right book delivered, simple test. If a customer uses an ATM to withdraw cash, they are various scenarios, but essentially you charge if money is given out and don't otherwise.
However if you have to pick a random card, it's difficult to prove. The same applies to the roulette machines that [UK highstreet] bookmakers have.
The way I tested the generator was using a variety of statistical tests.
(1) The first was the results were reasonably distributed, that is the percentage of results were close to expected values. In the craps games I would hope if I ran 100,000 games about half win half lose (more accurately about 49% 51%). Technically I would be worried if in a large number of tests the results were not quite close to the expected mean. If I ran 20 runs of 100,000 games, 19 of them would have to be within acceptable margins (someone probably knows the "best" method here, but 19 out of 20 worked for me).
(2) Next is that unexpected results do occur. It's no good if exactly 49 games in EVERY 100 win - one does expect good runs and bad runs. Without going into too much detail - these can be quantified and the range of results (e.g. 30 wins in 100 happened x times, 31 wins in 100 y times, etc.) tested to ensure most are close but there are a few unexpected ones as well.
(3) Use the output results and massage them, then test those. For instance in craps how many rolls between 7's (even though for the punter some of these cause a win and others a loss). This can be calculated to an expected distribution and compared as above. There are other interesting tests such as "parking taxis".
I should suggest stay away unless the organisation is very well known (in UK there are several well known high-street bookmakers or casino chains) as they have too much to lose. I use the same logic when considering the chances of a land-based casino cheating punters - they're making enough money already, why would they risk losing it all?
Personally I never gamble online (except freebies).
Quote: charliepatrickIt is surely in the interest of casinos to be able to show their software has been independently verified
I don't think it is really in their interest. If they wanted to prove they are not cheating, there is a really simple, fool-proof way to do that. Let the customer download an encrypted file with, say 500 outcomes. After the user makes 500 bets (or less, in which case the rest of the file is discarded), you let him download the key to decrypt the file, and verify that it matches the outcomes in the game he experienced.
There used to be some online casinos decades ago that did this ... and then the whole thing disappeared as if it never existed. I wonder why ;)
Heritage is off the black list. LooseLines also seems to not be using the software, leaving Legends and WorldWide Wagering. 5 Dimes discontinued their relationship with BLR today. Legends is denying any wrongdoing:
Quote: SBR Gaming forum
Greetings from Legends!
Thank you for contacting us, the Wizard of Odds is not a company that regulates casino softwares, Our casino is certified by a random number generator and is responsability of each customers to make bets, unfortunately we will not be able to conpensate your bets since they were place and played by you, we do offer a 5% casino rebate based on net losses per week.
If you have any other question or doubt please do not hesitate and contact us.
Best Regards,
Jason Devore
Customer Service Department
I've noticed that there is no notice on those two sites regarding their randomness or any testing. BLR Tech also has no randomness on their site either. I've heard boo from clempops for a couple of days and wonder if he's getting anywhere. I doubt any of these players will get their money back.
This is priceless!
Quote: SBR Gaming forum
Greetings from Legends!
Thank you for contacting us, the Wizard of Odds is not a company that regulates casino softwares, Our casino is certified by a random number generator and is responsability of each customers to make bets, unfortunately we will not be able to conpensate your bets since they were place and played by you, we do offer a 5% casino rebate based on net losses per week.
If you have any other question or doubt please do not hesitate and contact us.
Best Regards,
Jason Devore
Customer Service Department
any company that can send out something like this should be avoided.
Quote: boymimboHeritage is off the black list. LooseLines also seems to not be using the software, leaving Legends and WorldWide Wagering.
Can you elaborate on how you know that Loose Lines and Heritage are not using this software any longer? I certainly want to remove them from the BLR list if that is true.
Actually, a major part of my business used to be auditing online casino software. About the RNG -- that just speaks for itself.
Of course all of this is heresay, Mike, until you check it for yourself.
Quote: IbeatyouracesThis is why I will never wager online no matter what site or who endorses it, sorry wiz. I will stick to bricks and morter casinos and live equipment.
Unless you have played extensively online and 'tested' their software, you don't know that, in fact, there are many reputable and fair establishments (and software/s) to play at. I have played online for many years. Undoubtedly, I have probably played some crooked software, but the huge majority of my play has been with reputable and fair establishments. My results are testament to that.
Several years ago I collated about 1.5 million hands of online blackjack and my result was well within the expectation. From memory, my actual result in that sample was a loss rate of about 0.4%. Now let's say the online casinos were cheating and the game was 'rigged' to have a house edge of 1%, instead of 0.45% as per the rules (and assume playing perfect basic strategy). My result of -0.4% would be about 7 SD's from expectations. I reckon it's safe to say that I received a fair game.
Not all establishments and software companies are crooked.
An excellent response though one that is thought too restrictive by many. The wisdom of the approach is illustrated by the recent online poker sites set up by the Brick Casinos in Las Vegas all set to take advantage of online poker once it becomes legal. The brick casinos have licenses and strong regulatory agencies and thus the associated online sites are thought to have a giant head start in customer recognition and trust.Quote: IbeatyouracesThis is why I will never wager online no matter what site or who endorses it, sorry wiz. I will stick to bricks and morter casinos and live equipment.
I don't think I've voted in the poll yet, but its probably proper that I abstain due to lack of math skills but such abstention is not in any way due to a lack of being utterly convinced that the software is crooked.
Does anyone know who is behind that company and where they are located?
Quote: weaselmanI don't think it is really in their interest. If they wanted to prove they are not cheating, there is a really simple, fool-proof way to do that. Let the customer download an encrypted file with, say 500 outcomes. After the user makes 500 bets (or less, in which case the rest of the file is discarded), you let him download the key to decrypt the file, and verify that it matches the outcomes in the game he experienced.
There used to be some online casinos decades ago that did this ... and then the whole thing disappeared as if it never existed. I wonder why ;)
Wow, online casino decades ago... Interesting! I'd like to see what an online casino in 1980's looked like. Slots with ASCII graphics ;)
Anyways, the encryption method you mentioned does exist at one online casino - Betvoyager - which is the only online casino I know to certify it's fairness this way. See: http://www.betvoyager.com/games/randomness/
Quote: Jufo81Wow, online casino decades ago... Interesting! I'd like to see what an online casino in 1980's looked like. Slots with ASCII graphics ;)
Not, in the 80's. More like in 96 ...
Quote:Anyways, the encryption method you mentioned does exist at one online casino - Betvoyager - which is the only online casino I know to certify it's fairness this way. See: http://www.betvoyager.com/games/randomness/
Ah, interesting. Gotta check it out.
Quote: betvoyager.com
By clicking on the Calculate checksum button, the player will see the checksum for the text information mentioned above. The player can then compare this checksum to the one received at the beginning of the hand. The fact that these two sums correspond proves that the the game used the same deck that was created at the start.
Well ... this is better than nothing of course ... Except, how do I know that their program will calculate the checksum of the actual shoe that was just played when I push the button, and not simply output the same string that was displayed in the beginning?
I guess, I could just record all cards, and compute the checksum on my own, but ... (1) how many people know how to do that, and (2) one would have to know the exact algorithm they use to represent the deck, because for the checksum to match, the to representations must be absolutely identical, character to character.
Well ... like I said, better than nothing, but by far not as assuring as that thing I described earlier, which is the *only* way I know to reasonably assure the player that the online game is fare. With all due respect to Wizard, I just don't see how his audit can be worth much if anything at all. If they can turn the "rogue mode" on and off at will, no audit is ever going to uncover anything. The only way to catch them would be, like in this case, for someone to invest their own money, and meticulously document everything. And even then, the casino would have to get really greedy and stupid to get caught.
I finally checked out this thread, and looked at your results.Quote: WizardPlease read my post own results before you vote.
Um, WOW!
But, unless I missed something, it looks like those kinds of results would be a dark sider's dream.
Do the results change, to be similarly skewed towards a loss, when betting the Don't Pass?
I would assume so.
And that begs the question: What does the software do if you're betting doey-dont?
Quote: weaselman
Well ... this is better than nothing of course ... Except, how do I know that their program will calculate the checksum of the actual shoe that was just played when I push the button, and not simply output the same string that was displayed in the beginning?
I guess, I could just record all cards, and compute the checksum on my own, but ... (1) how many people know how to do that, and (2) one would have to know the exact algorithm they use to represent the deck, because for the checksum to match, the to representations must be absolutely identical, character to character.
I am no expert but they (Betvoyager) advise to calculate their checksum with any external independent website:
"The checksum is being computed with the help of the program used in our casino. Players can use other programs to compute the checksum for text information according to the SHA-256 algorithm. Players can find such programs on the following websites:
http://jssha.sourceforge.net
http://www.farfarfar.com/scripts/encrypt
http://www.fileformat.info/tool/hash.htm"
Also I believe that the key (code sequence) you get after the hand to unlock the checksum is unique, so there are no two ways to decode the checksum, making it impossible for the casino to choose a key that suits them the best.
If you are still unconvinced, why don't you test their Zero-house edge Craps (no registration or anything needed to play in demo mode):
http://www.betvoyager.com/games/demo/craps-equal/
In the top tab menu bar you can shift the values of the dice by any amount, so if the software initially chose 1-2, then user's shift of 3-3 would deliver outcome 4-5 = Nine. The third value is the number of outcomes generated for one checksum (default = 10). After you have done the specified amount of rolls, you can press the V button to view the results and verify checksum.
Quote: Jufo81
Also I believe that the key (code sequence) you get after the hand to unlock the checksum is unique, so there are no two ways to decode the checksum, making it impossible for the casino to choose a key that suits them the best.
Yes, that is true.
Quote:If you are still unconvinced, why don't you test their Zero-house edge Craps (no registration or anything needed to play in demo mode):
http://www.betvoyager.com/games/demo/craps-equal/
Heh. What does demo mode prove? It's kinda as convincing as the "third party audit" online casinos like bragging about so much. Surely, they would never cheat in demo mode or when they know they are being audited.
Anyhow, I checked out the betvoyager a little more, and tend to agree with you, it looks pretty convincing. To the extent of sounding "too good to be true" :) They are offering many 0 house edge games, including blackjack, which is also countable ... If they are also not cheating, how in the world do they expect to make any money? Just player mistakes?
Quote: weaselman
Heh. What does demo mode prove? It's kinda as convincing as the "third party audit" online casinos like bragging about so much. Surely, they would never cheat in demo mode or when they know they are being audited.
I wanted you to try the demo mode to get first-hand experience in how the "randomness control" is implemented, not to prove anything.
Quote: weaselman
Anyhow, I checked out the betvoyager a little more, and tend to agree with you, it looks pretty convincing. To the extent of sounding "too good to be true" :) They are offering many 0 house edge games, including blackjack, which is also countable ... If they are also not cheating, how in the world do they expect to make any money? Just player mistakes?
The shoes are always shuffled after every hand (new random ordering of cards is generated between hands) so there is no counting involved. They probably make money from players playing worse than optimal strategy and they also charge a 10% fee of net winnings made on 0% house edge games. So if you were to turn $100 into $1000 playing zero house edge games, you could only withdraw $910 (10% of $900 profit is deducted from your next withdrawal). Even though at first glance this sounds terrible, it's actually a cheaper gamble than to play against constant house edge over long-term.
Quote: Jufo81
The shoes are always shuffled after every hand (new random ordering of cards is generated between hands) so there is no counting involved.
It is? I did not realize that, because they kept talking about verifying the signature after the deck is completed.
But it looks like you are right, it (the signature) changes after every hand.
Quote:They probably make money from players playing worse than optimal strategy and they also charge a 10% fee of net winnings made on 0% house edge games. So if you were to turn $100 into $1000 playing zero house edge games, you could only withdraw $910 (10% of $900 profit is deducted from your next withdrawal). Even though at first glance this sounds terrible, it's actually a cheaper gamble than to play against constant house edge over long-term.
Hm. I did not realise that either, did not see them mention that anywhere at all.
I don't see how it is a cheaper option - with zero house edge, you are expected to lose 1/2 and win 1/2, but if they take 10% of winnings back, then your true expectation is -0.5 + 0.45 = -.05, for a horrific 5% house edge. :-/
I also don't quite see why they claim their BJ is 0 house edge to begin with. It has all the standard rules except AK spades pays 4 to 1. I get the probability of that happening as
8/312*8/311*(1 - 256/310*32/309) = 0.00060317. The difference in payout with the regular game is 4 - 3/2 = 5/2.
So, the decrease in house edge is 0.00060317 * 5/2 = 0.15%
Now, the regular game's house edge is about 0.37%, and decreasing it by 0.15% does not bring it to zero ...
Unless I screwed up somewhere in this math, it looks like my initial hunch about it being too good to be true was right on :)
Quote: weaselmanIt is? I did not realize that, because they kept talking about verifying the signature after the deck is completed.
But it looks like you are right, it (the signature) changes after every hand.
It is safe to say that at every online casino the shoe/deck is shuffled after every hand, apart from some rare exceptions. This shouldn't be news to anyone who has ever gambled online.
Quote: weaselman
Hm. I did not realise that either, did not see them mention that anywhere at all.
I don't see how it is a cheaper option - with zero house edge, you are expected to lose 1/2 and win 1/2, but if they take 10% of winnings back, then your true expectation is -0.5 + 0.45 = -.05, for a horrific 5% house edge. :-/
More details: http://www.betcruise.com/en/payrules/
You missed that the 10% commission is only deducted upon after cashing out once you have finished your play, not between every hand/outcome. Suppose that you played one thousand $10 hands in zero house edge blackjack and ended up ahead $100. As a result a $10 commission would be deducted from your winnings, but given that you wagered 1000*$10 = $10000 in total, it would translate only to 0.1% house edge ($10/$10000 = 0.1%), which is vastly lower house edge than a typical BJ game.
If you compare this commission scheme to regular house edge, you will see that the house edge punishes the player the longer he plays, because it will always catch him eventually and lead to a quaranteed net loss. But with commission system you can play as long as you want, even for years, with zero expected loss, and only pay 10% commission from net profit when you eventually cash out from the portal (it should be noted that they also pay 7.3% yearly interest for the total account balance you hold there: http://www.betcruise.com/en/daily-interest/).
Quote: weaselman
I also don't quite see why they claim their BJ is 0 house edge to begin with. It has all the standard rules except AK spades pays 4 to 1. I get the probability of that happening as
8/312*8/311*(1 - 256/310*32/309) = 0.00060317. The difference in payout with the regular game is 4 - 3/2 = 5/2.
So, the decrease in house edge is 0.00060317 * 5/2 = 0.15%
Now, the regular game's house edge is about 0.37%, and decreasing it by 0.15% does not bring it to zero ...
Assuming you refer to American BJ (AK spades pays 4:1 in the zero HE version), it has a house edge of 0.23% and the AK Spades payout brings it down by 0.15% to 0.08%. The definiton of zero house edge games (see: http://www.betvoyager.com/games/equal-odds/) includes a +/- 0.1% error margin so it's correctly classified as zero house edge with this definition.
Quote: weaselman
Unless I screwed up somewhere in this math, it looks like my initial hunch about it being too good to be true was right on :)
Well, it's just an online casino that does things a bit differently than the rest (randomness quarantee, commission scheme). This discussion is getting off-topic as the topic was BLR software. Basically I just wanted to let people know that it is possible to have online implementations out there that quarantee fairness for the player, which is what you sought after.
The sad thing is: There are estimated 4,000 online casinos in the business at the moment and if only ONE of them has an implementation that quarantees fairness for it's players, the online industry is going to have a lot to improve on.
Quote: Jufo81It is safe to say that at every online casino the shoe/deck is shuffled after every hand, apart from some rare exceptions.
Well, this one is an exception with its "randomness control" thingy, so could as well be an exception in the other thing too ...
Quote:This shouldn't be news to anyone who has ever gambled online.
I haven't. :)
Quote:You missed that the 10% commission is only deducted upon after cashing out once you have finished your play, not between every hand/outcome. Suppose that you played one thousand $10 hands in zero house edge blackjack and ended up ahead $100. As a result a $10 commission would be deducted from your winnings, but given that you wagered 1000*$10 = $10000 in total, it would translate only to 0.1% house edge ($10/$10000 = 0.1%), which is vastly lower house edge than a typical BJ game.
I stand corrected.
Quote:Assuming you refer to American BJ (AK spades pays 4:1 in the zero HE version), it has a house edge of 0.23% and the AK Spades payout brings it down by 0.15% to 0.08%.
Wizard's calculator says 0.2847. Reducing that by cut card effect (0.014) brings it down to 0.2707. Further decreasing by 0.15% leaves 0.12%. Still greater than 0.1.
Did I forget some rule?
Quote:Basically I just wanted to let people know that it is possible to have online implementations out there that quarantee fairness for the player, which is what you sought after.
This is noted. And kudos to them for doing this.
Quote: weaselmanQuote: Jufo81It is safe to say that at every online casino the shoe/deck is shuffled after every hand, apart from some rare exceptions.
Well, this one is an exception with its "randomness control" thingy, so could as well be an exception in the other thing too ...
Yes, fair enough. I have seen that card counting opportunities are very rarely offered online probably due to the instant access to computing power which would enable player to optimize his play perfectly by using some calculator, so he wouldn't even need to master any counting system. It's a kind of a different ballgame at B&M casinos where you have to all the work inside your head. So it's not a big surprise that if there is a Live Dealer Blackjack shoe online, the game has very stingy rules such as 8 decks, 50% penetration and burning several cards after every deal.
As for comparing the "house edge scheme" and "commission scheme with 0% house edge" I made a brief example using Roulette as an example. I wanted to make the example simple enough so that it doesn't need a computer simulation to get the results and is workable with pen and paper. That's why I chose roulette and 1:1 payouts, which enables using analytical risk of ruin formula.
I set the following parameters:
Player deposits $100 and wagers $25 per spin on Red/Black until he a) busts or b) reaches $300 and cashes out $200 profit.
Game #1
Betvoyager No Zero Roulette (0% house edge), and 10% commission on winnings.
Chance to reach target: P = 1/3 (independent of bet pattern used). Net gain if successful: $200 - 10% * $200 = $180
Expected value of play = 2/3*(-$100) + 1/3*($180) = -$6.667
So, the player pays a $6.7 cost for the betting action to triple his bankroll.
Game #2
Single Zero roulette (2.7% house edge), no commission on winnings.
Parameters for risk of ruin formula:
p = 18/37
q = 19/37
K = 4 (initial bankroll is 4 units at $25 per spin)
T = 12 (target bankroll is 12 units at $25 per spin)
Chance to reach target: P = [1-(p/q)^K]/[1-(p/q)^T] = 0.264.
Bust probability = 1 - 0.264 = 0.736.
Expected value of play = 0.7356*(-$100) + 0.264*(+$200) = -$20.69
So, we see that playing regular roulette with house edge is considerably worse for the player for this betting pattern and win goal.
Of course, the situation for game #2 changes if, instead of a sequence of smaller wagers on red/black, the player puts the whole $100 to one dozen for one round.
Now the EV is: 25/37*(-$100) + 12/37*(+$200) = -$2.7
So in the latter case, it would be more favourable to play the house edge version than 0% house edge version with commission. This is because in the latter case, the player plays only one round to reach the target with minimum number of bets.
So as a general guideline, I would say that the "0% house edge plus 10% commission" becomes cheaper for the player if the habit is to play long playing sessions with smaller bets per round (ie. typical recreational gamblers), whereas for a person who tries to win target goal with minimum number of bets (a gambler who is not looking for longetivity of play), the house edge version would be better.
Quote: weaselman
Wizard's calculator says 0.2847. Reducing that by cut card effect (0.014) brings it down to 0.2707. Further decreasing by 0.15% leaves 0.12%. Still greater than 0.1.
Did I forget some rule?
I went through some of my old posts and I found that I had already calculated numbers for this game in a thread about Betvoyager. See:
https://wizardofvegas.com/forum/gambling/blackjack/1183-betvoyager-com-zero-house-edge-blackjack-is-this-right/3/#post51055
I got a 0.263% house edge by using a different house edge calculator (very close to your value) and the AK spades payout increases the return by 0.1855%, thus pushing it slightly over the 99.9% threshold.
Of course, this game sucks, because the return is indeed less than 100% and you would be subject to commission in addition to being at disadvantage. For this reason I used to play Doublet Blackjack variation (european BJ with double down on any number of cards and double down rescue) that has ~99.85% return for the standard version and is not subject to any commission. I think that a 99.85% game without commission is far better than a 99.92% game with 10% commission on all winnings.
So to summarize: sometimes it is better to choose the standard house edge version if the house edge is low enough and also depending on how long you plan to play the game. When I used to play at Betvoyager, I used to mix between the 0% house edge and non-zero house edge games depending on which I thought would give me a cheaper gamble. Overall while playing there I felt I got a decent bang for my buck.
Quote: Jufo81AK rule increases return by 0.1855% pushing it slightly over the 99.9% threshold.
How does it increase it by 0.1855? What part of my math do you disagree with?
(4-3/2)*8/312*8/311*(1 - 256/310*32/309) = 0.00060317*5/2=0.001508 = 0.1508%
Quote: weaselmanHow does it increase it by 0.1855? What part of my math do you disagree with?
(4-3/2)*8/312*8/311*(1 - 256/310*32/309)= 0.00060317*5/2=0.001508 = 0.1508%
Could you explain the values in your formula? My formula was:
2.5 * 12/312 * 6 / 311
(2.5 = extra payout, 12 out of 312 cards for player's first card (A or K spades), 6 out of 311 for player's second card).
I also assumed that player gets paid independent on whether dealer has blackjack or not (the rules don't specify this), but even if player doesn't get paid with dealer blackjack, it only reduces the 0.186% gain to around 0.177%.
I'll go now, wipe the egg off my face ...
Quote: weaselmanNever mind ... I mixed up 6 and 8 decks (thus the 8's in the nominator, but 312 in the denominator), and also forgot to double the player's probability to account for the order).
I'll go now, wipe the egg off my face ...
Lol don't worry ;) These should be discussions, not competitions in who is right.
