Regulatory Issues

[Ron (r273)]

Excuse me while I open the bottom right-hand drawer of my desk, retreive my (not supposed to there) bottle of sour mash and pour myself a couple of fingers - Jeez-Louise...

Finally something I can relate to in this disscussion 

Very interesting subject you have started DealingwithNIGC542, keep up the good work guys.

Ron

[Op-Bell]

All of the training I've taken up to this point says there ARE (theoretical) cycles in Class III games. The first (elementary) training used a very familiar, very popular 3 reel, 64 stop game - cycle 262,144. NIGC classifies this game as Class III and the evaluation process I was taught translated very nicely to other games of this type (we're still running many of these games). What you're saying flies in the face of that. Are you saying that the time-tested, many times reviewed process I was taught is nothing more than smoke and mirrors?

No, it's not smoke and mirrors, but people misunderstand it, including people who ought to know better. The "cycle" isn't a guide to actual game outcomes, but to the probability of game outcomes. In a large number of game trials, the number of times you hit an actual outcome will trend towards the probability of hitting that outcome. In an infinite number of games they will exactly match it, but you can never achieve this, because after you've tried it an infinite number of times you can still try it again, and then again after that, for ever.

Take something simpler as an example, throwing a pair of 6 sided dice. The dice have a "cycle" of 36. In any group of 36 throws, snake eyes should occur exactly once. However, it's quite likely that you'll throw a pair of ones twice, or not at all, in 36 throws. It's possible to throw snake eyes 36 times in a row - highly unlikely, but possible. If you throw the dice 36 million times, the probability is that you'll throw a pair of ones a million times, but the chances of it being exactly a million is quite small. In the short run, though, probability is full of surprises - this is why betting systems never work and casinos manage to stay in business. Take the Martingale system for Roulette. You bet red or black, and every time you lose, you double your bet. When you hit your color, you get back everything you lost, plus your original bet as winnings. Sounds foolproof! The trouble is, you can get a string of enough losing bets to break your bankroll, or take you over the house limit. Even casinos don't always win. The people who built the old Thunderbird lost the casino on opening night to a lucky craps player.

In the process of designing a game, the designers simply add together all the possible winning outcomes over the whole game cycle, divide by the number of games in the cycle, and declare that to be the percentage. In the case of our pair of dice, say we pay 10 for a pair of sixes and 4 for a total of 7. In the "cycle" of 36, we get one payout of 10, and six payouts of 4, for a total of 34. This game then pays 34/36, or 94.4%. Will you get a payout of 34 every 36 games? Try it and see. But in 36 million games, you'll get quite close to 34 million payout. No game designer, or regulatory test, can go any further than this! It's a game of chance, goddammit. Sometimes chance will favor the player, and sometimes the house. All you can say is that in an infinite series of tests, the results will trend towards the theoretical.

[brichter]

P.S. brichter - "pencil-whipped" is now part of my regular vocabulary - give folks arouind here something to ponder...

That's a term I learned in the military, it applies equally well to civilian life. 

[stayouttadabunker]

Hi DealingwithNIGC542,
I'm following your posts with interest and I have this to offer.
I ran a S+ through 100,000 pulls about two weeks ago.
The reel prom had a 92.85% chip installed and I cleared the ram before running the test.
After the machine completed the pulls ( I ran the machine with a "Auto-play" switch),
I calculated the % and came up with 87.34% payback.
Now, this was done with EVERY pull playing the "Max" bet of 3 coins.
And took nearly four days to complete.
This machine would definitely be out of the "standard deviation" range of +/- 3% your casino is trying stay within.
I believe that 100,000 pulls is way too short of a complete "cycle" of the newer games you're talking about.
Because slots work with randomness properties,
I cannot see how any machine could fall into the +/-3% range...that's just too tight!
Esp. when this measurement is supposed to be taken with as few as 100,000 pulls...
gee, that's almost like taking a ball and throwing it as hard you can and
see if you can hit a running deer on the other side of a farmer's cornfield.
Just my 2 cents...

[jay]

Taxonomy is a wonderful thing as both of these definitions differ greatly from how I interpreted the term cycle.
My definition is closer to how SOB was explaining the sampling rate.

I was defining the cycle as the number of spins or plays needed to reach statistical norms.
In the S+ this is based on 10 million spins. While it would be a simple testing program to "play through" the RNG 10mm times the actual casino play to reach this is very high.
Over a year this would mean the machine would need  to be played 27398 times per day, 1142 times per hour, 20 times per min. Given the time it takes to complete a spin this is impossible.
To get this amount of play you are probably looking between 5 and 7 years at a busy casino.

With 128virtual stops on a 3 reel unit there are just over 2mm combos and as pointed out by opbell any particular combination may never actually hit.  

To DealingwithNIGC542 withs point it is possible that some themes (in a competitive market) have come and gone in 3 years and will never reach their statistical norm. If infact the norms are being based on larger than 10mm Ie 100mm then that time frame would be 10 fold ... 50 years.

[Op-Bell]

Taxonomy is a wonderful thing as both of these definitions differ greatly from how I interpreted the term cycle.
My definition is closer to how SOB was explaining the sampling rate.

I have always understood the term "cycle" to be the total of all possible outcomes, or the product of the number of reel stops in slot machine terms. This is what's always used to work out the percentage, and the resulting probabilities are always exact. The number of times a particular outcome appears in practice, however, is not exact - it follows a Gaussian or "normal" distribution (warning - head-exploding maths  ), which is governed by a different set of probability rules. Consequently you can't test a game for an exact percentage. All you can do is run the game to accumulate a total, then see where that total falls on the distribution curve and make an judgment as to whether it's reasonable. In a Gaussian distribution, 99.7% of results lie within 3 standard deviations of the mean, so any result outside that merits further study. The related Chi-squared distribution is the standard test for random number generators, but note that in this case, a perfect result is a FAIL (not random enough). It's strangely easy to achieve a fail - since mathematical RNGs have a fixed and known cycle length, you just run it over some multiple of the cycle length, and the longer the run, the more perfect the result. This has resulted in some manufacturers adopting a 64-bit or longer algorithm, so that no practical test ever gets near the full cycle.

The gaming control people in Britain used to carry out random inspections on AWP arcades. An inspector would show up anonymous and unannounced with a large bag of coins, pick a machine, play the coins through once, and count what came out. If this was too far below the legally mandated 80% he'd produce his ID and have the machine hauled away for further testing.

[jay]

Of course if he hit the jackpot no one would ever hear from him again ..... 

[stayouttadabunker]

  haha...what a dumb way to inspect a machines' payout... ....
I wonder how many coins he put in the bag

[rickhunter]

In all these machines, the less volatile the game, the more likely you are going to be close to the theoretical holdback percentage.  Take for example Double Diamond 90.95% chip.  The volatility predicts:

                 90% CONFIDENCE VALUES     VOLATILITY INDEX =   10.524
   HANDLE PULLS     LOWER         UPPER
                  PERCENTAGE   PERCENTAGE
        1000.      57.67         124.23
       10000.      80.43         101.47
      100000.      87.62          94.28
     1000000.      89.90          92.00
    10000000.      90.62          91.28

So theoretically within 100,000 pulls you are withing your 3.x% variance.  The key if you want to stay around this number is to get games that don't have ridiculously high top awards.  Something like 800 to 1000 coins max per coin bet would get your there.  The higher the volatility, the less likely you are to get within the holdback with a smaller sample size.

[stayouttadabunker]

I agree with you Rick,
however, DealingwithNIGC542's problem lies with the companies
giving him PAR sheets that don't even have the volatility index shown for the games he has?
Also, he doesn't really have a decent way to even check if his games are in line with what the
casino's are supposed to be making with these machines he needs to stay competitive with neighboring casinos.
Just about everyone he talks to passes the buck to somebody else.
The problem lies mainly with slow    regulatory bodies who are usually a step or two behind super fast advancing slot technology.
 

[DealingwithNIGC542]

To all those who have contributed - I can't thank you enough for what you've done up to this point. I have (obviously) a lot more research to do and you've provided me with some direction. I have something of a different question to pose.

I have taken pains (as much as I could) to not "point fingers" or to mention specific manufacturers or properties (including my own); there is a reason for that. Up to now, the dialogue has been focused on the technical aspects of this issue; I'd like to ask a question of a more ah, “philosophical nature�. Bear with me, I need to phrase this carefully:

Go back a little and you'll find comments that I made about the intent behind regulation(s) - protect the players, protect the casinos, protect the integrity of the industry. I’ve also alluded to what I perceive to be the “connection� between perceived effective regulatory oversight and the affect that has on patron confidence. That said:

Based on what you've all learned (and I have) up to this point, has your opinion of casino regulation changed at all? If you visit a casino, do you think you would view the operation and/or the games on that floor in a different light?

The reason I'm asking these questions: Last night as I was going home, I was pondering what real change would come from the dialogue we've been having (other than an exponential expansion of my technical background). In reality, I'm a nobody - I work in a little bitty casino on the fringe of no where (around here, they say that the only thing that separates this place from the Arctic Circle is a barbed wire fence). I don’t have the “voice� to effect any change.

Then I considered the “court of public opinion�. There are obviously a number of highly skilled, very experienced and opinionated people regularly visiting this site (last time I looked, the topic had been viewed over 1200 times). And, there was a point in time when Randy Fromm (publisher, Slot Tech magazine) indicated that he thought the regulatory issue was worth an article. I don’t hold a regular subscription to Slot Tech but the casino does; I’ve checked periodically and I haven’t seen anything. I wonder what Mr. From would think of the opinions that players (potential or otherwise) have about the issue; if that would throw enough fuel on the fire to get “folks� to pay attention. I don’t think we’ve violated any nondisclosure agreements (but I’m no attorney); no mention was made of any specific manufacturer nor of any specific game(s). GLI was mentioned, but anyone who deals with gaming knows that GLI is a key testing and approval lab. NIGC regs are a matter of public record; no harm there. Maybe the thing to do – stop by your local tribally-owned casino and inquire:

“Pardon me, but is your property in full compliance with 25 CFR 542.13 (h)(18)?�

All kidding aside (and please don’t get me wrong – I’m not trying to grab the tiger by the tail) – I’m just curious what you-all think of the idea. Opinions?? Options??

[StatFreak]

...  I’m not trying to grab the tiger by the tail ...

Murphy's Law dictated that you would finally post your conundrum just as I took a three-day hiatus from the forum, and between Jay, Brichter, Op-Bell and others, most of what I had to say has been said. I can't help chuckling, however, at your last comment, which appears on the second page and which, I swear, was exactly  the phrase that I uttered after reading your post. Okay, the exact phrase that I uttered was "Wow, has he got a tiger by the tail."

Since Op-Bell has covered the mathematics thoroughly, I'll table that end of it, except to say that I agree with him regarding the definition of a "cycle" of all permutations in probability mechanics, and in his distinction between the probability of outcomes vs. observed outcomes in random models.

I also happen to be a software developer (I deliberately refrain from using the term "engineer") and can only nod in agreement when reading the PowerPoint presentation that you posted regarding the catastrophic failures that can occur from the smallest programming errors and their causes. I have personally seen failures due to lack of communication between management, developers and clients, poor management, feature creep, and yes, software developers who aren't up to the task.

However, this seems to be more of an economic issue than a technical one. Computers do only what they are told to do, and people make mistakes. Theoretically, a company should tailor its level of QA against the consequence of error or failure, both for the company and for the end user. Obviously, QA in the release of a new version of a PC video game is not as critical as it is when developing software that manipulates people's money, which is not as critical as developing software that could kill thousands or millions of people should it malfunction, although one would hope that the game developers would care enough about their customers and be fearful enough of the potential loss to their bottom line to put a good QA program in place. Fear and greed are the basic drives here.

Gaming software isn't going to kill anyone, and while it affects the financial bottom line of the casinos and their patrons, it doesn't directly affect the finances of those designing it (the gaming company) unless they are participating directly in the game (WAP). It is, therefore, understandable that gaming software QA and documentation is going to be governed by the fear of loss of sales revenue due to poor product reception, and the greed and profit margins of the gaming companies wanting to produce new product at the lowest cost. They will never be as concerned about these data as you because it isn't their money that's at stake.  Their desire to minimize the costs of QA and documentation will be mitigated by the tolerance (or lack thereof) of their customers (you, and others in the industry) for accepting poorly or incompletely tested and documented software, the rigorousness with which the regulatory bodies enforce compliance, and the speed at which the laws are kept up to date. You can count on the laws always being a decade behind and the regulators being underpaid, understaffed, overworked, and slow to respond.

What that leaves are people in the industry, like yourself, doing just what you are doing. The only way that you are going to get the industry to change is to continue to be a squeaky wheel and to gather momentum and support from other casino operators, and I agree with you that the continued slacking-off of slot manufacturers producing proper testing and documentation could be a serious problem, particularly if server-based gaming takes off.

Having said that, I have to wonder if the reason that you are being stonewalled by the GLI and the manufacturers when asking for detailed volatility indexes and 90% (or 95% ?) confidence value tables is that they are simply unable to produce them and are unwilling to admit it. Last year I was briefly in touch with an Australian gentleman who had designed software to test newly developed Australian "Reel Power" slot machine games. The software was quite thorough, but one statement that he made that I took note of was that in some cases, the play and bonuses became so complex that in was not feasible to run through every permutation to arrive at a definitive conclusion as to the payback of one complete cycle. Instead, his program essentially simulated play by running hundreds of millions of plays (or more) to test the game and generate the SD and volatility data. He had also mentioned that some spreadsheets used by game designers could take over a month to setup correctly -- and that was just for an analysis spreadsheet -- which only emphasizes how complex these games have become.

To get back to the math side of it, the question might be: is it acceptable to operate class III games that are too complex to make a full analysis feasible, but that can be demonstrated to payback an expected percentage with an estimated volatility based on simulations instead of a full-cycle analysis, and if so, are we condemning ourselves an eternity to low-volatility/high hit-frequency games to keep the machines paying within a narrow range of payback percentage given that a machine may never see more than 0.1% to 1% of a complete cycle in its tenure on the floor of a casino?

DealingwithNIGC542, regarding your question, I can't say that my opinion of gaming regulation has changed -- yet, but I am glad to know that there are people like you who are conscientious enough to bring issues like this to light and to pursue them with vigor. I can definitely say that I am not pleased with the direction that slots have taken in recent years.

They've taken the psychology of gaming to its extreme, and even beyond (studies show that intermittent, random reward is much more addicting than no reward or guaranteed reward)  in that they have designed slots that pay out on almost every spin, only just not quite as much as was bet, and with numbers so large that people are duped into thinking that they've won something wonderful when, in fact, they haven't. What does a 15,000-coin win mean when it took 450 coin bets to win it? That's a win of only 33.33 to 1. I'll take my slots with top-prize odds that are slightly less than the number of stars in the universe, thank you. Give me a three-reeler with 262,144 combinations and prizes that pay over 500 to 1 that I can actually WIN. It's as Op-Bell pointed out with regards to 100-play poker. It's no fun if there is no volatility.

[DealingwithNIGC542]

StatFreak - Thank you for the input (and again to everyone else). Much appreciated. To be honest, I thought I might have gone a little too far in soliciting for input on how to proceed from here. I was inspired by the responses I had received up to that point and, while I do admit I felt more than a little trepidation about taking that step, I thought I could go a little farther out on the limb. I scared myself when responses dryed up. You made my day.

I'm going to spend some time thinking over your take on this issue; you've introduced a VERY interesting perspective that I hadn't thought possible - we aren't getting the tools we need because they are too difficult to produce or too expensive to produce and (probably) a combination of the two. If there is no squeaky wheel, I'm betting you're right - these damn things will saturate the market.

The studies you referred to - are any of them available on the web? I'd like to dig through them.

Thanks again, StatFreak. I'll be back...

P.S. - After we get the problems with 25 CFR 542.13(h)(18) all cleaned up, we can chat about 25 CFR 542.13(h)(2) - evaluation of multi-game/multi-denomination cabinets - a whole 'nother can of worms.

[StatFreak]

Okay, this is actually the last article that I found, but it's the most tongue-in-cheek, the easiest read, and the least ostentatious, so I'll lead off with it! 
I make no representation as to the anything else that you find on this site, and the article contains some offensive language (but none that Joey hasn't used here ) http://therawness.com/the-compliance-recipe-part-3-intermittent-rewards/



One study with rats and methanol: http://alcalc.oxfordjournals.org/cgi/content/full/41/3/225  (you will need to create an account to read their articles)



This is a patent application for a method of using non-random inverse adjustment of reward to behavior to produce a desired moderate level of the behavior in the subject. Although this is not the addictive response we're discussing, it is interesting nonetheless. http://www.freepatentsonline.com/WO2008089084.html
Here is a quote from that patent application:

Behavior reinforcement is the formal term for a process that uses reward or punishment to increase the frequency of a desired behavior. For practical reasons, the most suitable reinforcement in commerce is usually a reward. Important parameters in behavior therapy are the schedule and immediacy of reinforcement. It is important to draw the distinction between a continuous schedule, whereby every instance of desired behavior is rewarded, and an intermittent reward schedule, whereby rewards occur only with some instances of the desired behavior. A specific type of intermittent schedule is the variable-ratio schedule, in which reward frequency fluctuates. Likewise, the value of rewards can vary, even if reward frequency is continuous or fixed. Variable -ratio schedules tend to motivate people more than if the same amount of reward were distributed on a continuous schedule. This is partly because variable-ratio intermittent reward leads to emotions of anticipation, suspense, and pleasure. It appears to involve key reward centers in the brain. The immediacy of the reward, i.e., the delay between the behavior and the reinforcement, is ideal if kept to a minimum. The variable-ratio schedule is to be contrasted with a fixed-ratio intermittent schedule, for example, a predictable reward every third time a desired behavior is performed. Familiar examples of continuous rewards include: "20 percent off all athletic shoes;" "buy a pint of ice cream and receive a free bottle of chocolate syrup;"

I can find more if you wish.

If you were also referring to the Australian information, I'm afraid that the person in question wanted to remain anonymous, so I cannot go into any more specifics.

[DealingwithNIGC542]

StatFreak - some things to pass along:

Our slot operations guy just got back from a show in Arizona and brought back some interesting news. You know that company that produced that very common, very popular 3 reel, 64 stop game? The one that has been investing in server-based games for a while now? Turns out that they have a kind of beefed up "conversion kit" hitting the market. I haven't dug for details (all I got from the slot ops guy was a thumbnail), but it appears that these folks have found a way to (my term) "retrofit" some of their cabinets to enhance performance. Any guess what kind of games they are offering with these retrofits?

The links were great - I've passed them along to some other folks; really got their attention. Some of them weren't too keen on the concept of being manipulated in the way that they are; can't say I blame them. Most of them LOVE the way that those games "play with them" - their words...

I also excerpted some of the really relevant posts and forwarded them to a member of our Gaming Commission, the one that attended the GLI forum back in March. I'm hoping to get some time with her to talk this over; try to gain a concensus about what to do next. I forwarded the same set of excerpts to Randy Fromm; he got right back to me, saying he will be in contact - maybe some renewed interest there.

StatFreak - thanks again for the wealth of information. Just out of curiosity - if you were me, what would you do next? I've looked around (not real hard) trying to find other folks that have the same problem(s), but it isn't easy. I have it in my mind that there is a lot more "let the tiger sleep" attitude out there than there is the alternative.

Thanks again StatFreak and everybody else. Talk to you soon...

P.S. - I've had the pleasure of meeting and working with a number of Aussies when I was much younger. A more quiet, reserved group of citizens you couldn't find anywhere. I'm glad you had the opportunity to chat with him and he had the almost clairvoyant sense to talk to you. Ta....

[jay]

I didn't want to be the first to answer....

Pardon me, but is your property in full compliance with 25 CFR 542.13 (h)(18)?�

As a casino patron walking in I think asking this question would probably get me a blank stare from 90% of the employees. I might be told to wait while they got a floor manager who then might ask me for my credentials, for which I have none.
     (Save my Elvis Aron Presley drivers license and Hall of Justice honary membership card for sending in 15 Marvel Comic book proof of purchases... kidding).

They might give me the run around that they don't know what I am talking about, or they might tell me that is restricted information. I would equate this to walking up to the slot tech and asking him what the percetage payback was on a particular machine. Or asking him to go into statistcs mode to tell me what the last big jackpot was. The latter having been the cause of me being walked out for trying to obtain "advantage" information. Heck he was working on the machine anyways.
Its not like I used my own Key.....

Having read through this I am in some doubt that even if I got a polite "Yes" that I would NOW have some disbelief that anyone is capable of actually complying with the reg. In anycase I suspect I would be standing around for 20-30 min all the while getting the evil look from the better half with the occasional comment - "is this really necessary". 

Having participated in this post does this make me look at gaming any different ??
What I believe is that the probability of hitting the big one is low but not impossible. It has been done.
The brand name games (IGT, Bally, Knomi, Aristrocat, sigma, Universal ..,. etc) bring with them a certain amount of credibility since they are used enmass.
Publically traded casinos etc have a certain amount of profressional standards to uphold and just the relative sofistication that would be needed to "RIG" the machine so it would not pay off would make this a non-starter.

OpBell pointed out that the Casino already has the license to fleece so why steal..... excellent point.

People often say the "odds" are or it hasn't hit for a while so its "due", I usually don't spend the time to correct them that the odds are the same each and every spin
Given that I do not know, and no one will tell me expected payback, I still play the game, and this has not changed. I play for entertainment, not because I consider this gambling.
To me this is more like a lottery. Low cost to get in, and it gives me some ability to dream what IF ..... it only lasts 30 seconds.

For the most part I like to play BlackJack. The rules, odds, volitility indexs are well understood. I play the house and make conversation with the dealer.

I know this is your job (and no offense is in anyway intended), but as a technologist I see your role as a duplicated (wasteful) cost that is sholdered by the gaming public. There is no reason why all of these machines could not be "online" and report their real-time statistics directly back to the  gaming commission. This would allow them to monitor variences, compliance with % minimums, errors etc. After all it is they (he commission) who granted the manufacturer the go-ahead to sell a particular theme. There would also be no conflict of interest. IE you work for the casino, so would you turn in your employer for perhaps putting too-low a %% chip in. Whisle Blowers (think Enron) while appreciated by the public are generally not often able to find work in any industry. The manufacuter is driven by sales so if they are not permitted to sell the machines by the gaming comission, then they have to address this (the industry change you are looking for). Put the monitoring where the enforcement will be effective. At present your just a thorn in their side and the last thing they want to do is provide you the information that would perhaps interfere with a sale.

[DealingwithNIGC542]

Jay - When I made that coment, I was trying to be wry - you know, facetious. A typical floor person in a tribal casino would have no more idea about what compliance to 25 CFR 542.13 (h)(18) means than they would about any other subsection of the reg. More than likely, a floor manager or exec wouldn't have a clue either. It's not something they're expected to know. What I was trying to do was solicit ideas and/or opinions about how to proceed.

Playing blackjack IS a smart thing to do if you're going to gamble in a casino. It's a game of skill - you are directly involved in the outcome of each hand.

No offense taken insofar as my "slot" (pun intended) in the regulatory scheme, Jay. To clarify, in tribal gaming, there isn't a Nevada-style Gaming Commission overseeing the entire jurisdiction (state). Each casino has it's own Gaming Commission (as provided by Federal law); those individual bodies oversee their own properties. Games approved for play in my jurisdiction must be approved by GLI, then the state clears the game for play. The state only has regulatory oversight over those regulations connected to the Tribal-State compact; these compacts are negoitiated tribe by tribe. ALL other regulatory requirements are dictated by Federal law and represent the BULK of the reg we must follow.

I'm not trying to play the role of a whistleblower, Jay. I'm trying to build a mechanism to come into compliance with a particular reg and I was (at first) looking for other people with the same problem; I know the're out there. As it turns out, the research and development of the mechanism has taken me down a path I did not expect - one that I couldn't see it. The gracious folks here helped me better understand the real nature of the problem; I thought if they could help me with that, they might be able to see a way to proceed - you know - "I know this guy who knows a guy - let me get you in touch" -  that kind of thing.

Anyway, thanks for the input all the same - I appreciate it.

[jay]

You might want to take in the annual G2E tradeshow. This is the industry gaming expo. You can usually get a free ticket from Randy Fromm's slot tech mag. I was there about 3 years ago. They had several sessions that were specifically aimed at Tribal gaming, and other sections on regulation. I was too late registering but I also understand the property tours are worth the expense. Here is the link. Its Nov 17-19.
http://www.globalgamingexpo.com/images/100490/2007_graphix/index.html
Global Gaming Expo (G2E) Three-Day Conference Program
November 17-19

Highlights include:
  Casino Design 
  Corporate Social Responsibility (CSR)
  C ompliance, Law & Regulation 
  Finance
  Gaming Technology 
  Global Gaming & Development
  Human Resources 
  Indian Gaming
  Marketing  Players Clubs & Incentives
  Racino—NEW! 
  Retail, Dining & Entertainment (R, D & E)—NEW!
  Security & Surveillance 
  Table Games
  Wild Card 

[StatFreak]

...
I know this is your job (and no offense is in anyway intended), but as a technologist I see your role as a duplicated (wasteful) cost that is sholdered by the gaming public. There is no reason why all of these machines could not be "online" and report their real-time statistics directly back to the  gaming commission. This would allow them to monitor variences, compliance with % minimums, errors etc. After all it is they (he commission) who granted the manufacturer the go-ahead to sell a particular theme. There would also be no conflict of interest.

Jay, I'm not sure that I agree with you completely. In theory what you say makes sense. In reality, we know that NGC, and I'm guessing GLI, are underfunded and could not possibly police casinos at this level, even with automated technology spotlighting offending machines.

DealingwithNIGC542 has already defended his position as not being that of a whistle blower. I don't see his efforts in that way, either. It is often the people "in the trenches" who are the first to recognize potential or real problems within their areas of expertise. Regulatory agencies, like most law makers, are slow to react to changes, which is, perhaps a good thing overall, but which has proven to be a serious problem when confronted with rapid advances in technology, as we have seen more than once in the last 15 years since the internet moved from geek street to main street.

If people like DealingwithNIGC542 don't bring these issues to the attention of regulators, who will? If we take his word that the PAR sheets for the newer, complex games don't have complete data, I would consider this a problem. Once the gaming companies know that they can get away with providing substandard specifications of their games to their customers, how long before they start slacking off in their due diligence when designing the games? It's a slippery slope. Consider how long it's probably going to take for the GLI (and perhaps the NGC) to address an issue like this, even with people like DealingwithNIGC542 making noise. Imagine how long it would take them to get to it if no one said anything?

As a customer, I'm already concerned about the integrity of server based gaming, and it isn't even here yet! For 20 years people like us have had to dispel the myth that the casinos could change payout percentages on a whim, and that they did this on weekends and holidays, etc. If server based gaming is introduced as it is intended to be introduced, that myth becomes reality. Do I want to be playing next to someone who gets a better payback than I do because he plays more at that casino? The simple answer is no. I won't play slots, or any other electronic game, if I can't be confident that I'm getting the same shake as every other patron who plays the same game at the same denomination as I am playing.

I realize that my rant in the previous paragraph doesn't really have anything to do with the issue at bar, except that it is yet one more way that the advances in technology that we're seeing could potentially damage the gaming industry's reputation if not implemented carefully. Most players don't know what a PAR sheet is or care, for that matter. Most of them wouldn't understand how to interpret the math even if they saw one. Even members here have that problem, particularly with regards to the confidence and volatility indexes. But what they WILL notice after a while is what happens to their bankrolls when playing.

I'd site the 6:5 blackjack phenomenon as an example. Most blackjack idiots players accepted the rule because they didn't understand how much of the overall payback in blackjack was tied to that 3:2 payout. At first, the difference in the payouts didn't seem to be much and the casinos sugar-coated the rule change by adding some other favorable rules that didn't come close to compensating for the loss in BJ payouts, but which placated the players. When the casinos discovered that the players would swallow this guff accept the rule change, it swept through Vegas like Bubonic Plague. They kept it at the low-limit tables because they are the most frequented by tourists and new players who don't know better, and because their profit margins at the higher limit tables was probably sufficient and they knew that the more seasoned players wouldn't accept the new rule. But have you noticed the trend is starting to reverse? I'm seeing more 3:2 low-limit games now. It took a couple of years, but players finally started realizing that they were losing much more than they used to when playing their old favorite. I could foresee the same rebellion happening with slot players after they realize that they never seem to be able to get ahead playing these low-volatility games.

On the flip side, as a casino operator, I would be pissed off if the gaming manufacturers could not or would not provide me with detailed analyses of the games they were selling/leasing to me. I'm also not sure that I would want a game with a cycle so large that it would see less than 1% of a complete cycle in its life on my casino floor unless the game's volatility was so low that I could be comfortable with the expected deviations from the norm given the amount of play I would expect on my floor. In other words, I would only want high hit frequency, low volatility games... and that's precisely what these multi-lined, multi-bonused monsters are.

So what happens in a few years when the casino has filled itself with these extra-low volatility games to protect itself from the extreme variations that cycles in the billions can produce and customers begin demanding the return of higher volatility games where they actually have a chance of winning something substantial in the short-term?

[jay]

Jay, I'm not sure that I agree with you completely. In theory what you say makes sense. In reality, we know that NGC, and I'm guessing GLI, are underfunded and could not possibly police casinos at this level, even with automated technology spotlighting offending machines.
Stat - this is exactly my point, it should be done via vi automation - big brother stuff -  that could poll stats from all over and recognise variences between "like" machines. I know this is not done, the programs and systems don't exist but it is the way that it should be done. While I am not a proponent of server based (terminal) gaming the central collection, monitoring and analysis would be both feasable and effective not to mention cool

DealingwithNIGC542 has already defended his position as not being that of a whistle blower. I don't see his efforts in that way, either. It is often the people "in the trenches" who are the first to recognize potential or real problems within their areas of expertise. Regulatory agencies, like most law makers, are slow to react to changes, which is, perhaps a good thing overall, but which has proven to be a serious problem when confronted with rapid advances in technology, as we have seen more than once in the last 15 years since the internet moved from geek street to main street.

All I am pointing out is that if the goal is to ensure gaming purity - as in - machines working within operating parameters, protecting the casinos from fraud, protecting the gaming patron - I think it is very ultraristic to expect self compliance. First off the analysis tools and cooperation from the manufacturers are not there - so how can you gain this cooporation, as an industry regulator you force it, at the casino level you grin and bear it. .As a single casino - even if you did find a game that was particulary bad you have no insight if this is a single anomilie, or a more wide spead problem and what can you really do. Shut if off ?? If its popular or earning the casino would want it back on and at the end of the day they sign your paycheck. The position would not even exist if it was not a requirement to hold the gaming license - I can't see the casino doing anything to fund this position unless it was towards technology that would be cheaper than having to hire a second person. Once again I mean NO disrepect to DealingwithNIGC542 as this is like being between a rock and a hard place.

Over all I think if the casinos were charged xx $$ and this money then went towards a independent technology oversite group that could develop analsys technology, tools etc they would then have the power to manage some of these issues as an industry and be more effective.


If people like DealingwithNIGC542 don't bring these issues to the attention of regulators, who will? If we take his word that the PAR sheets for the newer, complex games don't have complete data, I would consider this a problem. Once the gaming companies know that they can get away with providing substandard specifications of their games to their customers, how long before they start slacking off in their due diligence when designing the games? It's a slippery slope. Consider how long it's probably going to take for the GLI (and perhaps the NGC) to address an issue like this, even with people like DealingwithNIGC542 making noise. Imagine how long it would take them to get to it if no one said anything?

No argument from me, this is needed !! - but again it is perhaps a case of lips without teeth.  The manufaturer on the other hand could look a casino owner in the eye and say " your buying 20 new machiens here - I can discount these by $1500 each if you get rid of nosiy-bob over there. Ethical ? Moral ? practiced ? - who knows - but from experience I have come to understand what is known as the Golden Rule  --- those with the gold make the rules. 

As a customer, I'm already concerned about the integrity of server based gaming, and it isn't even here yet! For 20 years people like us have had to dispel the myth that the casinos could change payout percentages on a whim, and that they did this on weekends and holidays, etc. If server based gaming is introduced as it is intended to be introduced, that myth becomes reality. Do I want to be playing next to someone who gets a better payback than I do because he plays more at that casino? The simple answer is no. I won't play slots, or any other electronic game, if I can't be confident that I'm getting the same shake as every other patron who plays the same game at the same denomination as I am playing.

I realize that my rant in the previous paragraph doesn't really have anything to do with the issue at bar, except that it is yet one more way that the advances in technology that we're seeing could potentially damage the gaming industry's reputation if not implemented carefully. Most players don't know what a PAR sheet is or care, for that matter. Most of them wouldn't understand how to interpret the math even if they saw one. Even members here have that problem, particularly with regards to the confidence and volatility indexes. But what they WILL notice after a while is what happens to their bankrolls when playing.

I'd site the 6:5 blackjack phenomenon as an example. Most blackjack idiots players accepted the rule because they didn't understand how much of the overall payback in blackjack was tied to that 3:2 payout. At first, the difference in the payouts didn't seem to be much and the casinos sugar-coated the rule change by adding some other favorable rules that didn't come close to compensating for the loss in BJ payouts, but which placated the players. When the casinos discovered that the players would swallow this guff accept the rule change, it swept through Vegas like Bubonic Plague. They kept it at the low-limit tables because they are the most frequented by tourists and new players who don't know better, and because their profit margins at the higher limit tables was probably sufficient and they knew that the more seasoned players wouldn't accept the new rule. But have you noticed the trend is starting to reverse? I'm seeing more 3:2 low-limit games now. It took a couple of years, but players finally started realizing that they were losing much more than they used to when playing their old favorite. I could foresee the same rebellion happening with slot players after they realize that they never seem to be able to get ahead playing these low-volatility games.

On the flip side, as a casino operator, I would be pissed off if the gaming manufacturers could not or would not provide me with detailed analyses of the games they were selling/leasing to me. I'm also not sure that I would want a game with a cycle so large that it would see less than 1% of a complete cycle in its life on my casino floor unless the game's volatility was so low that I could be comfortable with the expected deviations from the norm given the amount of play I would expect on my floor. In other words, I would only want high hit frequency, low volatility games... and that's precisely what these multi-lined, multi-bonused monsters are.

So what happens in a few years when the casino has filled itself with these extra-low volatility games to protect itself from the extreme variations that cycles in the billions can produce and customers begin demanding the return of higher volatility games where they actually have a chance of winning something substantial in the short-term?

Well said - when I mess with my Williams Pinball machince the manual takes great pride in explaining how the various paramaeters can be tweeked to maximize earnings - and how to interpurt the stats that are played. It seems the slot manufacturer has gone in the opposite direction and are attempting to use those comforting words - Trust Us

[DealingwithNIGC542]

StatFreak, Jay - hello...

A quick post - StatFreak's vibe in regards to High-hit/Low volatility games sounds like it's coming true - I took a stroll around one of our competitors floors yesterday - they are running a bunch of the same themes - there were people (more or less) lining up to play them. Interesting note (something I'll check on when I have a chance) - some of those games were running as in-house progressives; a twist I hadn't seen. Do you think that that's how the manufacturers are "getting around" the low volatility angle - offering a shot at a progressive (the banks that I looked at offered either three or four tiers of payouts)? Sounds like something that might work (satisfy the "want" for the larger payout) ....

Jay - my casino is scheduled for a upgrade down to a new floor accounting platform in the next couple of months - been doing some reading on the enjanced functionality - It's not what you envisualized, BUT it's been beefed up quite a bit - interesting.. The G2E thing - I know we'll have people going - might have to see if I can tag along..

Randy Fromm has been in touch - he is anxious to publish something on the issue, but I'm not sure if it would be in an op-ed format or what.  I'm not sure what to think about that... I'd still like to find other folks that are having this problem if I can - see what they are doing - maybe get them to chime in...

More later - thanks, folks...

[TZtech]

Hi

Posted a few response when you originally raised this issue on the old site.

In the previous company i worked for we used a spreadsheet to calculate the standard error from the hits and rewards (The Sw you linked to also has this feature and most manufacturers still make this info available) - This figure was captured into the online system when a new machine was commissioned. The system used this together with the machines monthly figures to calculate what we called a score (Similar to standard deviation) Anything above or below 3 we investigated.

I dont see large cycles as being a problem - Do however agree that evaluating accuracy is difficult to do and that regulations are not realistic for these types of games and that we cannot trust the manufacturers or the gaming labs to test products 100%

There was an interesting article in November 2008 Slot tech magazines about random number generators and a device which is now available that uses quantam theory to generate truly random numbers. Manufacturer is here - http://www.idquantique.com/products/quantis.htm.
i see great application for this in the gaming industry - RNG will no longer be software dependand and therefore not prone to errors that go along with SW developement.

Regards
Ian

[DealingwithNIGC542]

Hey, Ian - thanks for the input - some questions:

In the previous company i worked for we used a spreadsheet to calculate the standard error from the hits and rewards (The Sw you linked to also has this feature and most manufacturers still make this info available) - This figure was captured into the online system when a new machine was commissioned. The system used this together with the machines monthly figures to calculate what we called a score (Similar to standard deviation) Anything above or below 3 we investigated.

Glad you looked at that link, Ian. I've actually used that calculator in a few instances, but the only way that calculator works efficiently is with the volatility/SD info from the par sheet - With games/par sheets with no volatility index/SD info you have to manually load all the paytables - some of these games, the tables are 20 and 30 pages long. The other thing with the calculator is that you can't save the loaded data after calculation has been completed.  It DOES work, but very, very labor intensive. That said, when you performed your investigations, how did you approach the task? Did you work with Month-to-Date data, Lifetime-to-Date? What did you consider to be plausible explanations? Say, if a game had a score of 5 and there were no major payouts out of  the game, what came next?

I dont see large cycles as being a problem - Do however agree that evaluating accuracy is difficult to do and that regulations are not realistic for these types of games and that we cannot trust the manufacturers or the gaming labs to test products 100%.

Knowing what you do (now), would you have any trepidation about running these games on your floor?

Thanks, Ian -

[/quote]

[DealingwithNIGC542]

Sorry folks - yellow wasn't the best color choice -

Hey, Ian - thanks for the input - some questions:

In the previous company i worked for we used a spreadsheet to calculate the standard error from the hits and rewards (The Sw you linked to also has this feature and most manufacturers still make this info available) - This figure was captured into the online system when a new machine was commissioned. The system used this together with the machines monthly figures to calculate what we called a score (Similar to standard deviation) Anything above or below 3 we investigated.

Glad you looked at that link, Ian. I've actually used that calculator in a few instances, but the only way that calculator works efficiently is with the volatility/SD info from the par sheet - With games/par sheets with no volatility index/SD info you have to manually load all the paytables - some of these games, the tables are 20 and 30 pages long. The other thing with the calculator is that you can't save the loaded data after calculation has been completed.  It DOES work, but very, very labor intensive. That said, when you performed your investigations, how did you approach the task? Did you work with Month-to-Date data, Lifetime-to-Date? What did you consider to be plausible explanations? Say, if a game had a score of 5 and there were no major payouts out of  the game, what came next?

I dont see large cycles as being a problem - Do however agree that evaluating accuracy is difficult to do and that regulations are not realistic for these types of games and that we cannot trust the manufacturers or the gaming labs to test products 100%.

Knowing what you do (now), would you have any trepidation about running these games on your floor?

Thanks, Ian -

[/quote]

[StatFreak]

Tip: You can click on modify instead of quote to change your own posts. There are no time limits on modifications.


TZTech, thanks for the link to the Quantum RNG device.   COOL!!