Are RNG's Truly Random? Not On your Life

[Captain Chaos]

IIRC, MG use a centralized RNG? So these results should be easily replicated at any other MG casino? At one casino only it could be classed as a bad run. If you really wanted to prove that the RNG at MG in this case was not as they say then doing the same at other MG casinos would add significant weight to your results if they achieved broadly similar outcomes.

Do you still have the original data or spreadsheets by any chance?

 

[Harry_BKK]

IIRC, MG use a centralized RNG? So these results should be easily replicated at any other MG casino? At one casino only it could be classed as a bad run. If you really wanted to prove that the RNG at MG in this case was not as they say then doing the same at other MG casinos would add significant weight to your results if they achieved broadly similar outcomes.

Do you still have the original data or spreadsheets by any chance?

here from my previous post:

To make it broader I did them across a few casinos

i did a second one over a much longer period and a number of casinos

Trust me i did my scientific due diligence with regards to sample sizes, different times of the day, different casinos etc.

However, i have thought very long about posting this as i knew i would probably open a can of worms. Yes, i do still have the data recorded on a spreadsheet but sorry, i will not share nor publish it as i don't want to get involved in any kind of dispute. The results are clear to me.

We all have to remember that this is a piece of software written by people (MG, Netent etc) to make profit for an operator. Anything involving making a profit is rarely fair and square or completely random in the case of betting on a slot or roulette.

 

[Harry_BKK]

I have one more example which for me as a scientist starts the alarm bells ringing:

Mega Moolah - Progressive Jackpot slots.

If you click "View Payout" it says on the top of the first page: "The higher you bet the greater your chances of winning the progressive jackpot". Yet when I see the winners of the Mega Jackpot (which starts at 1Mio) it was won by an unusually high number of people betting 0.25 or 0.50 cents. Very rarely it was won by somebody playing max bets for these slots.

To me this is as well intended by the operators to encourage players to play high bets, hence getting the jackpot higher to attract more players and ultimately higher profits for the operators.

Yet higher bets do not seem to get the big jackpot pays more often then small bets as the above quoted sentence implies.

 

[dunover]

I have one more example which for me as a scientist starts the alarm bells ringing:

Mega Moolah - Progressive Jackpot slots.

If you click "View Payout" it says on the top of the first page: "The higher you bet the greater your chances of winning the progressive jackpot". Yet when I see the winners of the Mega Jackpot (which starts at 1Mio) it was won by an unusually high number of people betting 0.25 or 0.50 cents. Very rarely it was won by somebody playing max bets for these slots.

To me this is as well intended by the operators to encourage players to play high bets, hence getting the jackpot higher to attract more players and ultimately higher profits for the operators.

Yet higher bets do not seem to get the big jackpot pays more often then small bets as the above quoted sentence implies.

Because the majority of stakes are placed by small bettors.
These pooled jackpots work on a 'lottery' basis. One minimum 1c unit CAN win a prize, so betting 25c gives you 25 chances and so-on. Therefore as they say, betting bigger DOES improve your chance on any particular spin.
Given that at each moment the majority of play is on low stakes, overall the jackpot will target smaller bettors.

P.S. Your evidence suggests that the pool for the RNG must change in order to reduce the casino's liability as stakes as raised. This is similar to what the OP is suggesting on the roulette game.

 

[mathsboy1975]

here from my previous post:





Trust me i did my scientific due diligence with regards to sample sizes, different times of the day, different casinos etc.

However, i have thought very long about posting this as i knew i would probably open a can of worms. Yes, i do still have the data recorded on a spreadsheet but sorry, i will not share nor publish it as i don't want to get involved in any kind of dispute. The results are clear to me.

We all have to remember that this is a piece of software written by people (MG, Netent etc) to make profit for an operator. Anything involving making a profit is rarely fair and square or completely random in the case of betting on a slot or roulette.

What you are effectively claiming here is that for slot machines, the TRTP decreases as the stake increases but this should be fairly easy to test. I am not questioning your figures or analysis but I disagree with your conclusion.

If we adopt the conspiracy theory stance, then surely it makes more sense that the casinos encourage higher stake players, not penalize them with lower RTP. I am sorry but I just cannot see this happening. WHY would operators would do this when the system is in their favour already.

Having said all that it would be simple enough for anyone to access their playcheck logs, stratify their gaming history by stake value and analyse the RTP on a per stake per slot basis.

 

[Harry_BKK]

Because the majority of stakes are placed by small bettors.
These pooled jackpots work on a 'lottery' basis. One minimum 1c unit CAN win a prize, so betting 25c gives you 25 chances and so-on. Therefore as they say, betting bigger DOES improve your chance on any particular spin.
Given that at each moment the majority of play is on low stakes, overall the jackpot will target smaller bettors.

I understand that. So let's do some maths on the basis that every unit you bet higher will give you a better chance at winning the jackpot, e.g. 25 cents = 25 chances, 50 cents = 50 chances etc. and we have this case at one point across the network:

200 players betting @ 25 cents = 5,000 chances
10 players betting @ 625 cents = 6,250 chances

Statistically, one of the 10 players should win more often the mega jackpot than one of the 200 players, and this is by far not the case. You can follow the whole history of the MG Mega Moolah Jackpot winners ever since it was released more than a decade ago and will find that most were won with small bets, although statistically it should be different.

 

[mathsboy1975]

Mathsboy - I think the OP is suggesting he's pretty sure that the results can be 'reactive' at any given time. In other words there is a dynamic function which ensures the house edge doesn't deviate over a certain point in favour of the bettors, ascertained by the stakes placed before each spin and circumventing the fact the house will have the edge in the long term anyway. How he thinks this is applied I don't know. The easiest way to do this would be to have a fluid pool of RNG outcomes which meant on a spin-by-spin basis numbers with minimal or zero stakes were weighted in the pool.
This though would mean the software is illegal as I read it.
So, his question could be reworded to:
"Is there a lawful way the game could have built-in protection from excessive house-exposure and still be described as random?"

If we stick with roulette for now, then surely the only feasible way to implement the game is with a random number generator that produces an integer from 0 to 36 inclusive. It simply cannot operate like a slot machine (i.e. return 1.5 x stake profit for the next spin for example) as the win outcomes have fixed odds. Therefore the fairness can be determined only by the distribution of the spin outcomes. If you throw into the mix a weighting such that a bet placed that puts the house exposure over a threshold results in the probability of that outcome occurring being reduced then this surely HAS to be illegal? In fact surely any RNG that takes a players stake (or some function of the stake) as an INPUT should be illegal??? Furthermore, such manipulations would ultimately affect the TRTP in a measurable way such that higher stake players would see evidence of a lower TRTP??? This could be statistically measured.

It is very different in the situation in a real casino where the croupier can attempt (with some degree of success I am led to believe) to target a section of the wheel that is favourable to the house when the spread of bets across ALL players can result in a big house loss should the ball lands in a certain area.

When it comes to roulette, I simply do not understand why a casino operator would risk its reputation by trying to increase its edge when the game in its natural state is mathematically proven to make them money regardless!

With all of these conspiracy theories floating about, it makes me think it would be a good idea to start a company that produces open source slot machines so that everything is above board and players, operators and gambling regulators know exactly what is going on.

 

[Harry_BKK]

What you are effectively claiming here is that for slot machines, the TRTP decreases as the stake increases but this should be fairly easy to test. I am not questioning your figures or analysis but I disagree with your conclusion.

If we adopt the conspiracy theory stance, then surely it makes more sense that the casinos encourage higher stake players, not penalize them with lower RTP. I am sorry but I just cannot see this happening. WHY would operators would do this when the system is in their favour already.

Having said all that it would be simple enough for anyone to access their playcheck logs, stratify their gaming history by stake value and analyse the RTP on a per stake per slot basis.

I did not want to believe it either as i was a medium-to-high roller for years and was hardly ever able to make a withdrawal and always thought that i am just having bad luck.

Hence the reason why after the first test i did a second one with a much larger sample size at more MG casinos from different groups over a longer period of time. Although the results are not exactly the same, the trend shows clearly that the RTP decreases with higher bets.

For me as a scientist a truly random result from a limited number of available outcomes (millions on the newer slots) should average out over time with a large enough sample size, no matter what you bet. And this was never the case as my results are showing.

I might be completely wrong at this, just wanted to share my experience.

 

[mathsboy1975]

I understand that. So let's do some maths on the basis that every unit you bet higher will give you a better chance at winning the jackpot, e.g. 25 cents = 25 chances, 50 cents = 50 chances etc. and we have this case at one point across the network:

200 players betting @ 25 cents = 5,000 chances
10 players betting @ 625 cents = 6,250 chances

Statistically, one of the 10 players should win more often the mega jackpot than one of the 200 players, and this is by far not the case. You can follow the whole history of the MG Mega Moolah Jackpot winners ever since it was released more than a decade ago and will find that most were won with small bets, although statistically it should be different.

But the number of jackpot wins itself is actually a very small number compared to the amount of jackpot entries so it is hard to read too much into that. Compare the situation with playtech marvel jackpots. I have played playtech games for a long time as they are my favourite. In terms of turnover I have staked literally hundreds of thousands of pounds on these machines. Last week I hit the lowest marvel jackpot possible (£63) after a prolonged period of betting over £15 per spin (I had a good run and was up over £4k at one point so upped my stake considerably). So in terms of the marvel jackpot setup, my personal experience DOES match what you are proposing in your post.

Edit: Should mention that this was the ONLY time ever that I had a shot at the marvel jackpot

 

[dwa36]

All online RNG gaming isn't what it seems. I agree with the slot games biased towards lower stakes. I will say this if you play online roulette your numbers seem to go missing. But if you play £1 red/black just to see the results the numbers you would choose come in more regularly than when the chips on. The gaming companies online and in bookmakers are fully aware the games aren't honest same as the government but as long as the money comes in is all that matters to them. If you want to see how easy for the companies to cheat players go to a site virg....gami and watch online bingo with 75% of the players bots winning the prices. It is legalised robbery.

 

[Captain Chaos]

Seeing as from what I can remember, bingo sites automatically or at least have the option of automatically marking the numbers etc, I rather fail to see what a bingo bot could add? I thought most people went on their, paid, let the software do the work while they all chatted away in the chat box? I'm no online bingo expert admittedly. But that's how I thought it worked anyway. Why you would need a bot I've no idea!

 

[cncas123]

I am not a mathmetician, I am not an overly intelligent person, but what I do know is that in my experience of online roulette, at some stage of playing you will always hit a freezing cold run where no matter what numbers you are on they will not come in. And worse than that is that when you change your bet/stake, the numbers that previously carried the larger stakes but have had them removed, invariably fall in. Maybe its just that we notice this more on a losing streak, and we dont appreciate the randomness of the whole process, or maybe we just need the tinfoil hats, but it makes a myth of the randomness of a RNG on roulette, either online or in the bookies.

 

[Captain Chaos]

I can perhaps concur with the suspicion about bookie machines. I see the FOBTs as simply glorified fruit machines. I refuse to play them.

 

[RichieRespin]

Please remember that the source code for the game servers of the suppliers (MG, NetEnt, etc) are heavily checked and verified by several independent compliance agencies. The test houses do their own runs on the software, to check the correctness of the software, and the server software itself is heavily scrutinized.

As a software engineer, I can tell you how easy it would be to detect any type of foul play in these types of applications. And if a company like MG were to lose their license in one jurisdiction due to a systematic and conscious effort to cheat not just the players but the regulatory agencies, they would get kicked out of other jurisdictions very swiftly.

Why on earth would they risk getting booted out of this industry? To make a few extra bucks? They are already raking in cash and this kind of complete cheating makes no sense at all.

 

[spoton]

So 23 came up once in 426 spins, and never in 276 when carrying £1.55. Over this same period, all the other numbers on the wheel came up at least once.
There are numerous examples of this kind of gross bias to be seen within the data submitted.

The chance of not hitting one number over 276 rolls aint that small (37/38)^276=0.00064, vs for example the probability of hitting the same number twice in a row that would be (1/38)^2=0.00069 if I'm not thinking wrong. These things do happend, and if you are to proove something you need more than 5-15 such occurences.

 

[dunover]

Please remember that the source code for the game servers of the suppliers (MG, NetEnt, etc) are heavily checked and verified by several independent compliance agencies. The test houses do their own runs on the software, to check the correctness of the software, and the server software itself is heavily scrutinized.

As a software engineer, I can tell you how easy it would be to detect any type of foul play in these types of applications. And if a company like MG were to lose their license in one jurisdiction due to a systematic and conscious effort to cheat not just the players but the regulatory agencies, they would get kicked out of other jurisdictions very swiftly.

Why on earth would they risk getting booted out of this industry? To make a few extra bucks? They are already raking in cash and this kind of complete cheating makes no sense at all.

Are you aware by any chance of how these checks are conducted? Are the checks made of real-play or just a fun-mode basis? This could be important as the allegations are that firstly (and this HAS been proven in both the Spielo affair and on some iffy software mentioned in the rogue section) the game reacts differently to real-money play by being 'stake-sensitive' and secondly the real-play game differs from fun-play. Both are forbidden.

So, it's all very well testing a game's programming by pumping it for millions of results to establish there is a fair, reasonable and expected spread of outcomes, i.e. on BJ and roulette - that just means the basic RNG and overall TRTP function is being confirmed. But do these tests actually use random bet amounts/stakes and analyse if the RNG then produces a different result? This is where some people allege the disparity is - winning is being 'managed' somehow.

 

[Harry_BKK]

Please remember that the source code for the game servers of the suppliers (MG, NetEnt, etc) are heavily checked and verified by several independent compliance agencies. The test houses do their own runs on the software, to check the correctness of the software, and the server software itself is heavily scrutinized.

As a software engineer, I can tell you how easy it would be to detect any type of foul play in these types of applications. And if a company like MG were to lose their license in one jurisdiction due to a systematic and conscious effort to cheat not just the players but the regulatory agencies, they would get kicked out of other jurisdictions very swiftly.

Why on earth would they risk getting booted out of this industry? To make a few extra bucks? They are already raking in cash and this kind of complete cheating makes no sense at all.

That reads itself like something from dreamland Richie. Authorities, agencies that care about players, that would be something new to me.

Do you really think that these agencies & authorities, which are very often funded by casinos, are employing software geniuses who are able to detect complex software tweaks?

Did you know that for years RTG software operators could set an RTP themselves for every player!!! And the authorities, agencies etc were all OK with it because they were told that the RTP values could only be changed in a certain bandwidth.

"To make a few extra bucks?" - we are talking here possible 10's or even 100's of millions in extra profit every year, not just a few bucks. Always room for greed when these kind of money is involved.

 

[Westland Bowl]

Are you aware by any chance of how these checks are conducted? Are the checks made of real-play or just a fun-mode basis? This could be important as the allegations are that firstly (and this HAS been proven in both the Spielo affair and on some iffy software mentioned in the rogue section) the game reacts differently to real-money play by being 'stake-sensitive' and secondly the real-play game differs from fun-play. Both are forbidden.

So, it's all very well testing a game's programming by pumping it for millions of results to establish there is a fair, reasonable and expected spread of outcomes, i.e. on BJ and roulette - that just means the basic RNG and overall TRTP function is being confirmed. But do these tests actually use random bet amounts/stakes and analyse if the RNG then produces a different result? This is where some people allege the disparity is - winning is being 'managed' somehow.

If casino software is 'reactive' to the size of bets and/or betting strategy, then audits or 'checks' using flat-betting isn't going to reveal a rigged game.

 

[dunover]

If casino software is 'reactive' to the size of bets and/or betting strategy, then audits or 'checks' using flat-betting isn't going to reveal a rigged game.

Yes, that's what I was getting at - it's all HOW the testing of the software is done. Would it be possible to get this sort of reactive mechanism past the checking system?

 

[nisosbar]

....

 

[RichieRespin]

Are you aware by any chance of how these checks are conducted? Are the checks made of real-play or just a fun-mode basis? This could be important as the allegations are that firstly (and this HAS been proven in both the Spielo affair and on some iffy software mentioned in the rogue section) the game reacts differently to real-money play by being 'stake-sensitive' and secondly the real-play game differs from fun-play. Both are forbidden.

So, it's all very well testing a game's programming by pumping it for millions of results to establish there is a fair, reasonable and expected spread of outcomes, i.e. on BJ and roulette - that just means the basic RNG and overall TRTP function is being confirmed. But do these tests actually use random bet amounts/stakes and analyse if the RNG then produces a different result? This is where some people allege the disparity is - winning is being 'managed' somehow.

I don't know exactly how these tests are conducted, but I do know that the testing houses aren't fools, and that they examine the entire chain of the game round.

The most likely scenario is that the game result server, the server that spits out the gaming result, is not aware of the difference between a "for fun"-mode and a "real money" mode. It makes very little sense to have the same piece of software handle both the monetary transaction AND the RNG/reel-set win management code.

The way I think it is being set up is that one server handles the money (checking weather the the cost of the bet can be covered, stc), and then a request is sent over to the RNG-server that handles the actual game result.

The reason I believe this is the case is, not only from a code design point of view, but also from a "real world data" point of view: If you remember some issues that we have seen reported on here, players say "It runs fine in Free to Play mode, but it takes a looong time to get the spin results when playing for real". I've had that happen quite frequently, and I am pretty sure that there's been a number of threads talking about this delay.

In the free to play-scenario, the money-server doesn't have to do anything, so there is no delay, but in the "money"-scenario, the money server has to store play sessions into a database, keep track of all the play rounds, etc. (In "For Fun"-mode, the game history isn't available for the games, that only exists in "money"-mode, which means that a server does a lot more in the "Money"-mode. )

I may be wrong, naturally, but this is what makes the most sense to me, from observed delays during game play, and as a software engineer in general.

 

[RichieRespin]

If casino software is 'reactive' to the size of bets and/or betting strategy, then audits or 'checks' using flat-betting isn't going to reveal a rigged game.

A thorough check of the software will reveal this kind of cheating within minutes.

 

[RichieRespin]

That reads itself like something from dreamland Richie. Authorities, agencies that care about players, that would be something new to me.

Do you really think that these agencies & authorities, which are very often funded by casinos, are employing software geniuses who are able to detect complex software tweaks?

Did you know that for years RTG software operators could set an RTP themselves for every player!!! And the authorities, agencies etc were all OK with it because they were told that the RTP values could only be changed in a certain bandwidth.

"To make a few extra bucks?" - we are talking here possible 10's or even 100's of millions in extra profit every year, not just a few bucks. Always room for greed when these kind of money is involved.

I'm not going to stoop to "dreamland" insults. Have you _ever_ had your own software checked by any kind of authoritative testing house? I have. I know the rigor they use. And, no, it doesn't require a "software genius" to detect foul play in a slot game server, AND I think that if you have software auditors who check server implementations from a large number of software vendors, they *ARE* experts. No one else on the planet will have spent that much time looking at different types of server solutions as those guys.

I have had code vetted by testing houses, and I *know* that they are good.

Of course, rogue providers will have rogue solutions, but it is a vast step from a few historical anecdotes, to say that vendors like MG, IGT, WMS or NetEnt, who have been vetted by several test houses, by several authorities, in pretty harsh countries like Denmark and Italy, are cheating both the players, the test houses AND keeping all casinos in the dark about this.

 

[RichieRespin]

Yes, that's what I was getting at - it's all HOW the testing of the software is done. Would it be possible to get this sort of reactive mechanism past the checking system?

I sincerely doubt that. The source code of the software is being read. It is not just tested "black box"-style, the server code, most likely Java code, is being read by software engineers in testing houses. They would find these kind of tricks in minutes of reading the code.

EDIT: Sorry for spamming the thread with replies, maybe I should have put them all in one big post. My apologies.

 

[mathsboy1975]

I sincerely doubt that. The source code of the software is being read. It is not just tested "black box"-style, the server code, most likely Java code, is being read by software engineers in testing houses. They would find these kind of tricks in minutes of reading the code.

EDIT: Sorry for spamming the thread with replies, maybe I should have put them all in one big post. My apologies.

If every game provider has to submit their source code to an external agency for fairness monitoring then that is great - but is this actually a requirement before a gaming license can be granted??