RNG's - This needs to be addressed

Mugwump

Dormant account
Joined
Jul 20, 2004
Location
Oho
I'm sorry for the insane length of this post, but this behavior by online casino's is a pet peeve of mine, and the only way its going to end is for us, the patrons, to stop seeing it as acceptable.

------------------------------------------------------------------------
From a post in the "is Crypto rigged?" thread

kmartinusa said:
The best way that I can describe it is that the RNG's of machines vs. RNG's online are each random, but each in a little different way....The RNG's of online casinos seem to be set up to generate the appropriate number of winning hands that a mathmatical analysis would say that should be generated over a long period of time. The machines in land based casinos are random in a different way in that the individual cards (rather than winning hands) are dealt randomly and therefore the pay schedule probabilities match what happens. In both cases the game is fair, but is just done a little differently.
After a hiatus, I've started playing again at a few casinos. And am now ready to quit again because I've noticed behavior similar to what kmartin describes.

kmartin also believes that a results-based RNG is as fair as an individual-event based RNG.

Repeat after me...
THIS IS NOT TRUE AND IT IS UNACCEPTABLE!!!

Most of us on these forums who gamble away our dollars online choose to do so in games where we feel we have some control over the ultimate result. Games such as blackjack, video poker, etc, where we make decisions. Part of the reason for this is that the game is more interesting this way.

When we first hear it, kmartin's claim that the result-based RNG is fair will probably make some degree of sense. This is not true...and as casino patrons, we have a right to expect better out of the casinos.

There are several problems with this form of RNG

1.) In choosing a game such as video poker or blackjack, a patron has elected to play a game where skill influences the final outcome. They still do not have an advantage over the casino (usually), but they can (in theory) lose less over time with proper play. This is no longer the case when this type of RNG is used. They are now no longer playing the game they thought they were, and are now just playing an elaborately disguised slot machine. This is deception by the casino, and the player is not recieving the product he ordered. An analogy would be buying scallops and recieving shark-meat cutouts instead. I may not be able to tell the difference, but I'm still being cheated.

2.) There are times in which a results-based RNG 'win' is far different from what it ought to be.

For instance, lets talk craps...

I place a $10 pass bet down and the dice result in a point of 6. I then place the double odds bet on the 6. I now have $30 on the table, and will lose it if a 7 hits, and win $34 if a 6 hits.

Before the next throw I place a $10 come bet. Dice roll a result of eight. I place the odds down, and will now win $34 if a 6 or 8 hits on the next roll, and lose $60 if a 7 hits.

Now lets say that the results based RNG already determined that both the pass and the come bet would win by making their points....this is great, because I just won $68 right?

wrong.

Next roll is a 6...I win $34, YAY!!!! The point is now off, and on most online casinos my come odds bet is no longer in effect until a new point is set (the come bet still is). I place $10 on pass...

Next roll is an 8...and since my come odds are not active, I only win my initial come bet ($10). Instead of winning $68, I have now won $44 saving the casino $24.

Now I suppose that it is theoretically possible to code an algorhythm that would account for all of the possible permutations and generate 'fair' game results, but this will fail over time, because in order to achieve this the game will end up skewing the dice to where they need to be in order to accomplish the result, and the skew will show that the game isn't random....so they will have to adjust for the skew. Which will result in an event like I had today after a relatively long series where no 7 appeared. To adjust its history to correspond to its baseline, and appear 'normal', the program gave me something like the following 10 consecutive rolls:
7-7-7-12-7-11-7-12-12-7-3-2-7
(sorry bodog, that was my last deposit)

I don't know what the odds are of 10 rolls in a row without hitting a 'point' number are....but they've got to be pretty astronomical.

added later:
I also played vp there this week, and in a few 1000 hands recieved an above average number of full houses. Most of these were kings over queens, and most the k over q full houses all occurred either pat or after 3 card draws after the software had drilled my 100 unit stack to 45 or 50 units in an incredibly short period, pushing the stack buck up to the 90 or 95. Again, a win of sorts, but it sure didn't feel legit​

3.) Lastly, this form of RNG invites abuse by the casinos. If I've decided that you are going to lose your next hand of blackjack....which is the better hand for me to deal you? 6-5 vs dealer 6? or 10-9 vs dealer 10? Obviously the 6-5 is, as the player following standard BJ guidelines is going to lose double his bet there. All I have to do to increase my overall odds as a casino then is to toss a couple more of these than usual into a player's results during a session. The player will curse his bad luck....but will be hard-pressed to prove anything odd is going on. He did, after all win the % of hands that he was supposed to, and he was only dealt 3 11's, so any results there are statistically irrelevant.

If casino's kept logs of all of their game event results (many, if not all, do) and allowed auditors to examine these logs, this kind of behavior by casino software would be readily detectable. Unfortunately, I have never seen evidence of such an audit at an online casino. Auditors check the overall result (game returned x% to players) and compare it to the cash-in, cash-out totals, but they do not do the statistical analysis of individual events necessary to detect this form of manipulation. I say this because I am absolutely positive that if a casino did perform such an audit, they would certainly post evidence of the results for the public to view, assuming they passed, and if reasonably skilled auditors were working, the only way to pass such an audit would be to use an event-based rng. To date, I have seen no evidence of such an audit, except on www.pokerroom.com's website. www.pokerroom.com's audit however, only refers to poker hands though, and not its casino games.

Sorry for taking so much time to rant here, but I feel that its time we started expecting more out of online gaming, and the blind acceptance of this algorhythm is starting to really piss me off.
 
Last edited:
Mugwump said:
The RNG's of online casinos seem to be set up to generate the appropriate number of winning hands that a mathmatical analysis would say that should be generated over a long period of time.

Which casinos are using these false RNGs and what proof do you have?
 
A totally valid question, and I do not have an acceptable response to it.

From what I have seen over the past year in blackjack, craps, and video poker. In my experience (and the second hand reports of others) Cryptologic and RTG both regularly give results that don't feel 'right' (even on wins) and I've seen this as a common explanation as to why happens.

Like everyone else on these forums, I do not have an endless supply of cash to make bets with. If I did, I would happily hire a bunch of people to generate enough data, analyze the results, confront the companies with their fraud, and demand they conform to our expectations. Since this is so, we can only hope that the companies themselves submit to a regulatory body at some point in the future. It will doubtless happen someday, but I suspect its going to be a long time coming.
 
Last edited:
What casinos are using fair RNGs, and what proof do you have of this?

Wait...wait...PriceWaterhouseCooper and eCogra. Bingo - I knew I'd find it if I thought long enough.

LOL.

It's true that you hear a lot about "non random cards but fair results" and it's also true that that's a total crock of bollocks. If it's non-random it's rigged. Whether or not the claim actually has any weight is another matter, but it remains rubbish either way. "Non random but fair" - gimme a break. I love that one.
 
this rather cute, i would like to add an analogy to the topic of the thread regading a situation i just encountered today

i play semi-often at a reputable casino on a reputable software platform

however, in the pass i've noticed that blackjack at this casino is played literally in streaks

i mean when dealer is on streak, you know even if you get dealt 20 its no good, at most its a tie

however, this casino also go on insane streaks for the player, i've literally won 15 hands in a row before. so i counteract this strategy by betting small when dealer is on fire and making bigger bets when i'm on fire.

so in this instance, i was in the middle of my hot streak where i was betting 50$ a hand and think i won the last 6-7 hands, i get dealt 10, 6 dealer has 3 up. the funniest thing was that i was distracted by something else and pressed hit by mistake. as i watch in astonishment as a 5 floats to me and dealer busts on a 3, 6

..makes you wonder sometimes
 
Random RNG confirmation is the key to the entire industry going mainstream.

Until then, we are playing a minority sport.
 
ezc3m said:
i play semi-often at a reputable casino on a reputable software platform

however, in the pass i've noticed that blackjack at this casino is played literally in streaks
This is exactly what I'm talking about, and its because I'm sure others have reported the same type of 'curious' results at the same casino. And we've all decided to accept it.

I've been guilty of this too, and because of this, my resolution this year has been a simple one. If I suspect that a casino is not playing a truly random game, I can make one more deposit. If I still don't feel its random after that deposit, even if I've won. I remove the software and send their cs an email explaining why they will not recieve any more of my money. So far the only responses I've recieved have been indignant denials, which is to be expected, but I feel its my responsibility to let them know why I won't be giving them any more deposits. I also suggest that they conduct a real 3rd party audit of game event results and publish the audit results if they want to regain my trust.

So far, I've uninstalled bodog and global player casino, I suspect intercasino is going to be next, but I haven't played there this year yet. Uninstalling GPC really hurt...I like the variety of their vp, and their paytables, but after winning my first (and only) vp royal flush there, I immediately sufferred some of the worst losses I've ever seen over several months in their happy hour deuces wild game, 19 or 20 one hour sessions, 1 winning session, down almost 4000 coins in approx 17000 hands and no quad deuces (@ 1/5000 hand frequency), they were first to go after Jan 1.

If nothing else, at least I'm getting rid of all this crap on my hard drive. But the email does give me a little bit of self-rightious ego buffing, and worse come to worst, I should be saving money by the end of the year because I won't be able to find an online casino to play at.
 
caruso said:
It's true that you hear a lot about "non random cards but fair results" and it's also true that that's a total crock of bollocks. . "Non random but fair" - gimme a break. I love that one.

Hi Caruso,

I've rarely posted but lurked for years. I appreciate the experience and insight that you have brought to this board. This message board really only has value when it has posters who are knowledgeable and who are willing to take time out of their day to post their thoughts for "the masses" (me included) to learn from.

I have never heard the theory of non random cards but fair results forwarded before. I've played literally tens and probably hundreds of thousands of hands of VP both online and offline. My observations are in no way scientific, so please know that I don't pretend to be the definitive authority. Please don't discount my observations, either although they are strictly anecdotal, they are based on lots of experience.

I have noticed that MG VP doesn't play anything like what I've experienced from a machine in a casino. My experience is exactly as I described, almost as if the outcome is predetermined rather than each card being dealt randomly. I disagree with the other guy in that Crypto seems more random to me than MG, and I haven't played enough playtech lately to really have a strong opinion either way.

I guess that my question to you is why do you feel so strongly that the non random but fair argument is completely without merit? Why do you feel that the non random card but fair results is also a crock of "bullocks"? If my memory serves me (without spending the time to research it), you seem to be someone is guided by facts, so I am interested in your thoughts.

I'm not questioning your integrity or anything else. I have no contact with other online gamblers other than this forum and a couple of internet acquaintances with whom I have infrequent contact. I am genuinely interested in this topic because I've never seen this particular discussion put to the test and this is a subject that I've always kind of been curious about.

Thanks for your time in responding.

Sincrely,

Kevin
 
kmartin,

I hope you don't feel that I'm calling you to task for your comments. That was not my intent at all. I'm a lurker too, and I've seen the 'non-random but fair' theory floated out there several times on these boards. My grief is not with you, or others who feel that 'non-random but fair' is possible. It is with the casinos that appear to engage in it. For you and others who feel as you do, its merely an issue of education. For the casino operators who appear to use this form of RNG, its an issue of deception and fraud.
 
Mugwump,

No problem at all from me...In fact I found your thoughts enlightening and interesting.

I really don't know why I am so interested in this subject or the details of this discussion. In the scheme of things it really doesn't matter in my world.

My only suggestion to you is that you don't quit altogether...I'm not sure that the number of hands that you or I could afford to play would constitute definitive proof of cheating by anyone. Besides, I don't believe that the casinos, online or otherwise have any reason to cheat with the odds already stacked in their favor. They'll win in any case and the risk of getting caught cheating far outweighs the small rewards gained by doing so. Our sport tends to be streaky.

I do understand your point relative to the difference of how a RNG operates. This is especially interesting although it hasn't victimized me personally as I'm ahead of the casinos and have been for several years. I'm really not convinced that it matters either way in the long run.

Unfortunately, I'm not smart enough to figure out what a fair game is mathmatically. I rely on others who appear to be legitimate to make that call for me...Bryan, Gonegambling, posters here who have done the appropriate analysis, etc. So far I have been very satisfied with my personal results relative to what I would have expected, so I guess that those who I rely on for advice have done OK by me so far.

Thanks for taking the time to head things off, but you didn't offend me in any way. I appreciate your concern.

Kevin
 
Last edited:
kmartin also believes that a results-based RNG is as fair as an individual-event based RNG.

Here we go again. I have addressed this in a number of threads in the past.

An RNG is an RNG. If it is NOT an RNG it is a NRNG.

In other words:

A Random Number Generator is a Random Number Generator. If it is NOT a Random Number Generator it is a Non Random Number Generator.

There is no such thing as a results-based RNG. There is no such thing as an individual-event based RNG.

The result of the bet can be non-random - but the RNG *cannot* be non-random.
 
QUOTE: What casinos are using fair RNGs, and what proof do you have of this?

Wait...wait...PriceWaterhouseCooper and eCogra. Bingo - I knew I'd find it if I thought long enough.UNQUOTE

Good heavens, Caruso has experienced a Road to Damascus conversion to the truth at last LOL!
 
spearmaster said:
There is no such thing as a results-based RNG. There is no such thing as an individual-event based RNG.

The result of the bet can be non-random - but the RNG *cannot* be non-random.

Spearmaster,

Thank you for your ever so patient lesson in semantics. Although I thought I had explained the premise well enough, I apparently need to be a little more precise.

I agree with you absolutely that an RNG works or doesn't work. (i.e. it is random or not random). All an RNG does is generate results from 0 to 1 that is manipulated to create the game.

That is why I made the distinction between event-based and result-based. And there is indeed a difference, and that difference in the algorhythms used to build the game.

In a slot machine, every pull is both an event and a result. There is no distinction. This is partially the reason why (most of us on these forums) find slot machines boring as hell.

Lets imagine a more complex game - but one where it is easier to explain the distinction that I'm trying to make than in a more complex casino game

Rules of the game
1.) I place a bet
2.) We each roll a die
3.) I compare my number to yours
4.) I can now optionally double my bet
5.) We now roll one more die.
6.) High total wins, dealer wins ties

I know I probably have an advantage in this game...but thats to show the distinction.

The most natural use of the RNG in this case would be to use the RNG to develop the 4 die rolls with results of 1-6. 4 dice = 4 events = 1 game. This is the model I refer to when I mean event based.

Now the "non-random" or result-driven use of the RNG would be to use it in a few different methods, both of which I've seen proposed in the past.

1.) Take the random number generated, 0-.49999 = you win .50-.9999 = I win.
2.) Generate (using the RNG again) a couple of dice, show them
3.) Let me make my side bet
4.) Generate a couple more dice that show the final results, reroll or adjust the dice until they match the result determined in step 1.

This abuse of the RNG would still generate a game that appeared random to an individual over a typical gaming session. However, the effect and purpose of my side bet is now neutralized, giving you back the advantage. I would play, and most likely sadly curse my luck at the end of the session.

Or in a more sophisticated fashion....
1.) Over time, generate using either method 1 or 2.
2.) Keep track of the results
3.) If the results exceed some given deviation, use the RNG to determine whether its time for a 'correction'. Say if I'm 3 rolls ahead of you in a 10 roll series, assign an 80% chance start skewing my results towards losses. If you're ahead, assign a 60% chance to start skewing me towards wins. Skew using method 2, with changed win %

This second use is actually what I think would be a more likely explanation for many of the bizzarre behaviors we've seen at some casinos. The reason a casino would want to use an algorhythm such as this would be to limit their risk exposure. Given enough individual events, the casino is statistically guaranteed to show a profit. As you are no doubt aware, there are many online casinos out there that are struggling to generate enough users to generate enough events to negate the effects of game volatility. This second algorhythm would reduce the exposure of the casino to large wins. They would still happen, but not nearly as often.

Note that in all of these cases we are using a random-number generator. Note also that in all of these cases, the results of an individual gaming session would fall within the parameters of expected results. Note also that in the last two cases, the number generated by the RNG is massaged to fulfill the casino's needs, hence the belief in the "non-random" RNG. In actuality, what we are talking about here are the algorhthms used to generate the games, but for interpretation by an average non-technical person, I figured that using the term RNG would suffice, and it was easier to type than "result-based algorhythm used to transform the number generated by the RNG into a game result" or ""event-based algorhythm used to transform the number generated by the RNG into a game result" It was a tremendous error on my part, and I apologize for failing to make such an important distinction. I hope that what I intended to communicate is clearer now and perhaps maybe acceptable to you.

Oh by the way, I studied English in college. You should correct your post.

"An RNG is an RNG. If it is NOT an RNG it is a NRNG." is wrong. It should be worded as:

"An RNG is an RNG. If it is NOT an RNG it is an NRNG."
or
"An RNG is an RNG. If it is NOT an RNG it is a Non-Random Number Generator."

And wouldn't a non-random number generator, just be a number generator? So maybe it should just be:
"An RNG is an RNG. If it is NOT an RNG it is an NG."
or
"An RNG is an RNG. If it is NOT an RNG it is a number generator."

But what is a number generator anyways? Well, obviously, it would be a numerical sequence in mathematical terms. So perhaps
"An RNG is an RNG. If it is NOT an RNG it is an NS."
or
"An RNG is an RNG. If it is NOT an RNG it is a numerical sequence."

Perhaps if you stated it this way, you would not have to address this issue again in the future.

Alternatively, you could also just pull the stick out of your ass.
 
I apologize to any that think that my response to Spearmaster was out of line, and to Spearmaster if he truly believes that what he was saying was important. As we're all aware, sometimes we type things that come across a bit more harshly than we meant to. Unfortunately, I'm in a bad mood today, and in no mood for pointless flames. Perhaps I interpreted his remarks to be a little more arrogant and hostile than he intended, and if so, I apologize in advance.
 
Mugwump said:
I apologize to any that think that my response to Spearmaster was out of line, and to Spearmaster if he truly believes that what he was saying was important. As we're all aware, sometimes we type things that come across a bit more harshly than we meant to. Unfortunately, I'm in a bad mood today, and in no mood for pointless flames. Perhaps I interpreted his remarks to be a little more arrogant and hostile than he intended, and if so, I apologize in advance.
At least you caught it before I did. Please don't force me back into the roll of "Forum-cop". Let's either take a few bong hits, or chill pills, or listen to Enya with the lights turned off before we post, okay?

Or simply ten deep breaths...that will usually do the trick.
 
Last edited:
Excellent posts Mugwump.

It all depends on whether an individual believes that documents such as the attached actually merit any credence, I for one do not.

My experience/records of online Blackjack hands/trends throughout a session lead me to believe in an RNG determined game outcome rather than that produced by RNG generated individual cards. I may well be proved wrong but the behaviour of the game software results in sufficient trends for me to take advantage and a regular profit.
 
vgyhnji said:
My experience/records of online Blackjack hands/trends throughout a session lead me to believe in an RNG determined game outcome rather than that produced by RNG generated individual cards.

I wonder if the actual mechanism by which it's rigged isn't just increasing and decreasing the likelihood of being dealt face/10 cards - to create the sort of trends card counters wait for but which shouldn't happen on-line if random cards were dealt each hand.

I'd also agree it's so blatant as to be exploitable - before anyone suggests the casinos would have gone bust because of it you have to bear in mind that extreme 'streakiness' will make most players lose money quicker. The extra variance means players go bust more easily, and the common negative progressions (or the even commoner 'chasing'!) will be more likely to fail. Casinos know that after long winning streaks they can also usually rely on the player to give them most of the profit back (especially if there's a wagering requirement involved).
 
Vesuvio said:
I wonder if the actual mechanism by which it's rigged isn't just increasing and decreasing the likelihood of being dealt face/10 cards - to create the sort of trends card counters wait for but which shouldn't happen on-line if random cards were dealt each hand.

if you decrease the likelihood of face cards and aces, you don't even have to wait for the player to bet big to gain more of a house edge

the whole basis for card-counting in blackjack is based on the theory that face cards are good for the player and smaller cards are good for the house.

if small cards are more likely to come up, then house naturally gained an extra edge with or without the player making progressively larger bets
 
more thoughts on event-dependent vs. result-dependent outcomes

wizard denotes the following on pull tabs, a result-driven mechanism that looks like a regular video poker machine:

"So a pull tab is like a gum ball machine. As a visual aid the game may show how much the player won in the form of a slot machine of video poker game. However don't be fooled, there is never any skill in a pull tab game. Even if the game looks like a five card draw video poker game your outcome is predestined. For example if you get a royal flush on the deal and throw all of it away you would get another royal flush on the draw. If you kept only one card to a royal you would get four wild cards on the draw."

now who amongst us is brave enough to test this hypothesis on the major software providers by throwing away a pat royal? :D :D
 
ezc3m said:
if you decrease the likelihood of face cards and aces, you don't even have to wait for the player to bet big to gain more of a house edge

I should have made it clear that I think Microgaming has streaks in the player's favour as well as streaks in the casino's favour - so overall you get the expected result, but due to the way people usually play the casino ends up making more money.
 
vgyhnji said:
Excellent posts Mugwump.
It all depends on whether an individual believes that documents such as the attached actually merit any credence, I for one do not.

<attached document of pwc audit results summary>
Thanks.

I've seen a few sites post audits from pwc, usually for 1 or 2 games, and usually out of date. Perhaps the audit was conducted with reasonable dilligence, but we will never know.

In the justice system, we go with the premise of innocent until proven guilty. This does not work if you are an auditor. As an auditor, you need to assume that there is something fishy going on, and then request evidence from the customer to prove otherwise until you are satisfied. I do not believe that this is the case with most audits, and not just in the gambling world. Enron and the collapse of Andersen are sad examples of what happens when audits are too lax.

There needs to be a certified process by which a company can prove that a game works. This should involve both data analysis, as well as an analysis of the code used to generate the games. The process used, data used, and code analyzed, should all be publicly available.

Game companies who cry foul when asked to open up specific portions of thier software for examination do not have a legitimate complaint. In any land-based casino, there is a specific procedure to run each game. These procedures are all public (except for slots), otherwise we would not have the world of gambling literature out there providing advice. The code used to generate a game is really not that much different. I may use different techniques to code it, but if you compare working programs written by two authors to generate a blackjack game, the code will look remarkably similar. Software differences between gaming platforms should involve the graphics used in the display, design of the user interface, and in the security algorhythms used for client/server communication. It is reasonable that a company would want to keep these private, the game logic does not need to be considered a company secret, and should not be.

Certifying these companies' games (except for slots, which I believe are uncertifiable, although the players are :D) would NOT be that difficult for a neutral third party. The only reason I can imagine that a company would be reluctant to undergo such an audit, is that they believe they would not pass such scrutiny. After all, wouldn't the end result of such an audit be a safer gaming environment, and thus more people willing to risk their money on the turns of the virtual cards?

Again...sorry for the length of the post. I'm bipolar and currently in a mild manic phase. In addition to making me irritable, it also makes me wordy. If someone has a guide on short and sweet message board posting for dummies, send me the link, I'd like to learn.
 
Last edited:
kmartinusa said:
I guess that my question to you is why do you feel so strongly that the non random but fair argument is completely without merit?

Blackjack and video poker games ostensibly run on free running RNGs (or whatever the correct term is) with no weighting. If the RNG is NOT free running as claimed (or as one is led to believe), then its unfair. Casino Bar is an example. Although using an RNG, other factors are involved to produce the seconds-dealing result. Thats unfair. If the Casino Bar software were somehow programmed to deal extra player blackjacks to compensate for the fix, that would still be unfair because the ostensibly free results are tampered with though it remains a mystery to me why a casino would claim to deal a random game via their RNG, then in fact deal a non-random game, but produce results commensurate with a random game! Neither do I have the foggiest idea how they'd achieve that, LOL. To rig a game to still return the correct theoretical amount is a bit contradictory. I figure if they're cheating it's probably going to show up in the game payout. If it doesnt, its probably fair. If it does, its probably rigged.

If the casino claims a random game which is actually functioning other than claimed, via weighting or whatever other principle, it's unfair. If it's a blackjack "slot" masquerading as genuine game, it's unfair.

jetset said:
Good heavens, Caruso has experienced a Road to Damascus conversion to the truth at last LOL!

Yup, even the great and the good occasionally fall from their pedestal and end up rooting around in the muck and the slime. Sucks.
 
caruso said:
though it remains a mystery to me why a casino would claim to deal a random game via their RNG, then in fact deal a non-random game, but produce results commensurate with a random game! Neither do I have the foggiest idea how they'd achieve that, LOL.
My theory is that they do it to assure themselves the house take. Most online casinos do not get nearly the traffic that we tend to give them credit for, which means that volatility is a significant issue for them.

And you are right, I do not think it would be possible to create a non-random game that appeared random in a way that would be undetectible if we were given enough data to analyze.

Short and sweet? Did I do it?

YAY!!!!
 

Users who are viewing this thread

Meister Ratings

Back
Top