Saturday, December 22, 2012

A Wart's Opinion - Casino Police Still "On the Pad" - Part 1

Towards the end of an earlier post (A Wart's Opinion - Casino Police 'On the Pad'?) I made the following statement:  "Regarding locking threads - here is a link to another Casinomeister thread which will be the subject of an upcoming post."

This is that "upcoming post".  In writing it, I found that I needed to revisit that earlier CM thread (the Casino Club thread) in order to extract some workable data.  In gathering this data I discovered that everything that I wanted to say, everything that I felt needed to be said, about all of this was producing a VERY LONG post..

(The devil is in the details, and writing about the details usually takes a whole lot of words.)

I managed to move some content into my "'On the Pad' Addendum" post.  When I looked at what was left, however, I decided that I needed to break it into 2 parts.

Even then, this Part 1 post is truly formidable in its scope and magnitude.  Brace yourself for a bit of a mountaineering adventure - the climb may be a little long, but the view (in the totally biased opinion of the writer) is certainly panoramic and hopefully educational.

Onward to Casino Police Still "On the Pad" - Part 1.
______________________________________________________________________

My colleague made the post "Avoid Betfred (And Casinomeister) Like The Plague" on Sep 7 2012.  In this post he linked to (what he termed) a full-on shitstorm at Casinomeister (CM).

His post dropped quickly to the bottom line - CM is bought and paid for.  He then left it to the reader to be entertained by the linked CM shitstorm thread.

In this post I'm going to spend a little more time than my colleague on this CM/BetFred issue.  As indicated above I'm going to break this post into 2 parts.  In this first part I'm going to revisit the CM thread that was the subject of my previous post (Casino Club) and extract what I can from there (data, methods, conclusions).  In the second part I'll then apply that knowledge to this BetFred shitstorm.

In my previous post  (Casino Police "On the Pad"):

#1 - The issue occurred in Nov 2008.

#2 - It involved a very profitable CM affiliate (Casino Club).

#3 - It included the suspected use of a "robot" (aka "bot") by the Player.

#4 - It included the incompetent and subjective (some, like the writer, added lazy, useless and ultimately "cover your ass" corrupt) attempt at proving robot use.

In this new post  (Casino Police Still "On the Pad"):

#1 - The issue occurred in Aug 2012 (4 years later).

#2 - It involved a very profitable CM affiliate (BetFred).

#3 - It included the suspected use of a "robot" (aka "bot") by the Player.

#4 - It included the incompetent and subjective (some, like the writer, added lazy, useless and ultimately "cover your ass" corrupt) attempt at proving robot use.

In other words - same shit, different day.

Let's see if, after 4 years, anything has changed.

To do that I think we should have some way of measuring change.  I'll create a ruler, an Integri-MeterTM.  (I expect that it will be a useful tool for future posts as well.)  In several previous posts I've referred to the sale price of Grandma, so our Integri-MeterTM is going to measure Grandma's sale price.

One end is "pocket change", which I'll define as 1 dollar.  The other end is "priceless", which I'll define as 100 dollars.  (Yeah, I know - 100 bucks is priceless?  But a measuring stick with a top end of 900 trillion would be kind of tough to work with.)

We now have a measuring stick that goes from 1 to 100.  Let's apply it, step by step, to the previous post.  (Full disclosure - I'm not on any hallucinogens (at the moment).  This is a figurative measuring stick, and its application is about as subjective as you can get.)
______________________________________________________________________

#1.)  Player started thread on Nov 14, Player screwed by Bailey BlastTM on Nov 21.  Hidden courtroom - closed doors - judge, jury and executioner all rolled up into one - the gavel bangs - all done.

I think it is fair to assume that CM did actually contact the casino involved (Casino Club).  I also think it is fair to assume that CM took everything that Casino Club had to say at face value - the Player was presumed guilty from the start.  But I'll extend to CM the benefit of the doubt that some small effort was made to ensure that there was at least some proof of that guilt.

Integri-MeterTM rating = 10.


#2.)  Subsequent forum activity forces CM to have another look at (and perhaps for the first time actually analyze) the casino's data.  Continued forum activity exposes that the initial data "analysis", by both the casino and by CM, could most kindly be described as a complete fustercluck.  No one knew which end was up ... literally!

I suspect that the active and ongoing forum exposure forces CM to conclude that any further discussion would do nothing but cause (more) damage to both CM and Casino Club.  Casino Club agrees, pleads Nolo contendere and pays Player on Nov 28.

HOWEVER, because of Nolo contendere plea, no punitive actions were taken by CM.  That is, Casino Club remained on CM's Accredited List.

Integri-MeterTM rating = 15.  (After all, he did get the Player's money back.)


#3.)  Mr. Bailey and Mr. Drayman (Max Drayman, aka MaxD, the Manager of the CM PAB system) continue to post in an attempt to transform the now publicly exposed PAB System's odor of incompetence and corruption into something resembling clean laundry.  Sadly, these attempts simply increase the stench.  (I use the word sadly because what they were trying to do, and how they were trying to do it, was just ... sad.)

Integri-MeterTM rating = 10.


#4.)  Simmo!, one of the CM Forum Moderators, defends CM by describing the PAB system as a free service.  Mr. Bailey, in his Bailey BlastTM, threatened to shut down the PAB system for all new members.  Simmo!, Mr. Bailey, and Mr. Drayman finally add the icing to the cake - they bitch about people complaining when CM's "Secret PAB Process" has f**ked a Player out of a $10,000 credit.

Free Service?  Let's look at that.

CM's PAB process applies (almost) exclusively to the casinos on CM's "Accredited List".  To quote from CM:

"EVERY MEMBER MAY PAB ONE TIME ON ANY CASINO - ANY FUTURE PABs WILL ONLY BE ACCEPTED CONCERNING CASINOS LISTED AT CASINOMEISTER - WE RESERVE THE RIGHT TO TURN DOWN FURTHER COMPLAINTS."

In short, the PAB process is CM's "personal guarantee" of Player satisfaction for those casinos that he recommends.  How many sign ups do you think he gets (= how many affiliate dollars do you think he makes) as a result of this "PAB warrantee"?

Free service my ass.

Final Score - Integri-MeterTM rating = 5.

(Remember, this is a scale from 1 to 100, with 1 as "pocket change".  In this example, Mr. Bailey sold his Grandma for 5 bucks, which is still a bit higher than pocket change.)
______________________________________________________________________

Side Note **********

In a previous post "Casino Police 'On the Pad'? Addendum", I'd have to give CM an Integri-MeterTM rating of 1, which is the minimum value possible and equates to selling Grandma for pocket change.

It was obvious that "the Player was guilty until proven innocent", CM took everything that the casino (BetFair) had to say without question, there was NO effort expended to assist the Player in proving their innocence, there was NO thought applied by CM to the issue of "robot use", either in specific (this BetFair complaint) or in general (possible future complaints).

I mean, it just can't get any worse than this.  I think that an Integri-MeterTM rating of 1 for the issue discussed in the "On the Pad" Addendum is appropriate.  CM sold Grandma for pocket change.

End Side Note **********

In all of this Casino Club business, what was learned?  I mean, item #4 in the Casino Club list at the top of this post deals with proving the use of a robot.  There was a whole lot of discussion (182 posts in the thread) - were there any conclusions?

#1.)  Are robots intrinsically "bad"?

Mr. Bailey (Casinomeister), in the "Dealing with players" part of his Philosophy on Online Gaming and Mission Statement, had this to say about Bot use:  "... those operators that ban bots are either naive, unschooled in casino math, or are just inexperienced.  Bots will not change the game's math, nor will they foresee random numbers or predict the future.  Bots only speed game play, and if the odds are in the house's favor (which in most cases they are), bots will only allow the player to lose money at a faster rate."

However, within the somewhat heated context of the thread discussion, Mr. Bailey had this to say:  "... to use a bot is fraudulent play. I think anyone with more than three brain cells engaged will agree on this."  He also had this to say:  "The bottom line is do not use bots ..."

It's safe to say that Mr. Bailey doesn't know whether he is coming or going.


Mr. Drayman (MaxD) had this to say:  "I for one am now convinced that judging these case by what 'seems reasonable' isn't sufficient."  "... the scale of [ed. suspected] bot usage and the corresponding frequency of 'false positives' demands better of us."  "Either that, or as some have suggested, the issue of bot detection needs to be taken off the table entirely."

(NOTE:  I added suspected to the above quote, for what I consider to be obvious reasons.)

It's safe to say that a College graduate was required for this particular project, but Mr. Drayman is still in Grammar School.


Simmo!, one of the CM Moderators, had this to say:  "IMO the best that a Bot can achieve when a bonus isn't in play is optimum strategy. That still leaves a house edge, so a casino refusing to payout - even if Bot play is proved - is both wrong and unethical."  (At least he is consistent with the CM Philosophy statement.)


Mr. Bailey, in his Philosophy on Online Gaming and Mission Statement, also had a quote from Michael Shackleford (aka the Wizard of Odds):  "If I ran a casino I would fully allow bot use."


Mr. Bailey is wandering around in circles.  Mr Drayman is still at the lower end of the learning curve.  Simmo! and Mr. Shackleford are at least clear and definitive.

Conclusion - Are robots intrinsically "bad"?  NO.


#2.)  To what games does any concern for a Player's "robot use" apply?

On this one, and within the scope of this discussion, I think the answer is pretty clear.  A casino's concern for a Player's "robot use" only applies to those games that involve Player decisions during game play.  (A robot's advantage in "Speed Play" during time-limited Tournaments, though valid, is, in my mind anyway, not within the scope of this current discussion.)

A "robot" is programmed to apply "optimal strategy" to all of the games that it plays, where optimal strategy is the "statistically best" method of playing a particular hand.  So, if a game does not involve any Player decisions then the use of a "robot" is not an issue.

A short list of games for which Player "robot use" does apply - any/all forms of Blackjack, any/all forms of Table Poker, any/all forms of Video Poker.

(A short list of games for which Player "robot use" does not apply - any/all Slots, any/all forms of Roulette, Keno, Baccarat, Sic Bo, Craps.)


#3.)  If robots are not bad, then what's the big deal?

For those games which involve decisions during game play, human beings (that is, "non-robots") are not statistically perfect game playing machines.  It's been documented that "human error" adds about 1% to the house edge for these games.

Most casino software manufacturers actually provide robots for their slots and keno games (aka "Auto-Play Systems").  Surprisingly, even though Video Poker games are in the "does apply" list above, many casino software manufacturers provide robots for these games too (aka "Auto-Hold Systems").

I'd say that the biggest big deal with robots is their use as a "grind away tool" for meeting the wager requirement (WR) of a casino bonus.  The casino people who "design" these bonuses appear to have about half as much intelligence as is required to do a good design job.

That is, rather than including an estimate of how many Players are going to be using a robot to grind away the bonus WR, they assume that all Players are going to be seeing final results of (House Edge + 1%).  Robots may then shift the bonus program from a "money maker" for the casino through "break even" into a "money loser".

So, I'd say that this is the big deal with robots - their use removes human error, and the profit that goes along with it, from the casino's operation.


#4.)  How do you prove that a Player is using a robot?

A Side Note **********

Here is a link to what I consider to be a reasonable, well written discussion of the issue of proving "robot use" at a Casino "beyond a reasonable doubt".  I've put this Side Note link at the top of this discussion point because I plan to use the phrase "beyond a reasonable doubt" below.

(And, just for the sake of including it, here is a good follow up post to the link above.)

End Side Note **********

In the Casino Club "robot use" thread there were four "performance indicators" discussed throughout:

A - Rate of Play, in games per hour.

B - Total Duration of Play, in hours.

C - Ratio of "Break Time" to Total Duration of Play, as a percent.

D - All games demonstrate compliance with "optimal strategy".

Let's see if any consensus was reached (CM Moderators, knowledgeable and experienced Players, etc.) on specific numbers for each of these 4 criteria.  That is, if Rate of Play is being used by the Casino to prove that the Player used a robot, then what is the rate, in games per hour, above which means "robot" and below which means "human being"?
______________________________________________________________________

A - Rate of Play

An Important Note **********

The maximum (human being) Rate of Play is going to be different for different games.  Here are 3 examples:

1.  Caribbean Stud Poker.  After the Deal, your options are to click a "Raise" button or a "Fold" button.  Not a lot to do there.

2.  Video Poker.  After the Deal, you (usually) have to individually click from 1 to 4 cards before clicking Draw.  (Sometimes, you just click Draw.)  That may result in a little, or a lot, of mouse work.

3.  Blackjack.  After the Deal, you may click a "Split" button, a "Double" button, several clicks of a "Hit" button, etc.  Or you may just click "Stand".  Again, the result may be a little, or a lot, of mouse work.

For this Casino Club thread, the game involved was Blackjack.

End Important Note **********

I think that MaxD started the ball rolling when he stated in the following post:

"... a while back I handled a PAB where bot use was claimed and the proof given was a trace where the player was 'playing' at about 500 hands per hour (sustained). The casino claimed that was sufficient proof that a bot was in use and I agreed with them 100%. Would anybody here claim otherwise? I seriously doubt it. And so, voila!, we have a case where the play records and the conclusions drawn from them were pretty much beyond question."

Starting Point = 500 games per hour.

This was then followed by member after member (I won't burden you with the links) stating that 500 games per hour was not even cruising speed for them.  600, 900, and 1,000 games per hour were mentioned.

New Rate = 1000 games per hour.

Another Important Note **********

I think it's obvious that the maximum Rate of Play is tied into the Total Duration of Play.  That is, some people are obviously comfortable at a Rate of Play of 1000 games per hour, but for how long?  I mean - 1000 games per hour for 24 hours is obviously going to require some sort of pharmaceuticals.  (I wonder if that is spelled out in any Casino's T&Cs?)

End Important Note **********

I'm reminded of another Player problem in which their rate of play was used against them.  This one occurred in early 2011, and involved the Video Poker game played at EasyStreetsports (now named EZStreetSports).

Side Note **********

With this EZStreetSports complaint we have "Dueling Sportsbook Forums" - TheRXForum and SBRForum.  We also have a problem-specific post from the Wizard of Odds (WOO), as well as some level of discussion in the Wizard of Vegas (WOV) forum.

I'm not going to get into all of the gruesome details here.  The WOO article, linked below this side note, has a lot of links to follow if you've got the time.

But if you want to see some really weird shit concerning getting paid by a Casino, read the following post on TheRXForum.  In this one, the Player was required to take a trip to Costa Rica, submit to a polygraph test, and then demonstrate their game play in front of an audience.  Because they refused to do this they were labelled a "fraud".

End Side Note **********

The game involved at EZStreetSports was Video Poker.  The rate of play that was being defined by the EZStreetSports Casino as "statistically impossible" was 1,000 hands per hour, sustained over an 8-hour period, with each hand demonstrating "optimal strategy".

The WOO posted details, as well as his opinion, on this web page.

Finally, here's where we bring in that phrase "beyond a reasonable doubt" that I went to so much trouble to introduce above.  A large number of people are saying that 1000 games per hour (about 17 games per minute, or about 3.5 seconds per games) can be done by what - 1%, 3%, 5% - of the online gamblers that play the "robot use" games listed above.

So, is the possibility that 1, 3, or 5 Players out of 100 get screwed if they play 1000 games per hour proof "beyond a reasonable doubt" that a robot was used?  I, and MANY others, would say not even close.  1 Player out of a thousand?  Maybe - I could be persuaded.  1 Player out of 10,000?  OK, I'd agree with that.

Whatever the actual Rate of Play which only 1 Player out of 1,000, or 10,000, can achieve, it is certainly above 1000 games per hour.

Final Important Note **********

Let's say that a Player has a history of game play for pick-a-game at pick-a-casino (weeks, months, even years).  And then let's say that this history shows a game play rate per session (or even per day) of 300 to 500 games per hour.  And finally, let's say that this Player suddenly, on some Tuesday in June, starts playing at 1,000 games per hour sustained over many 6-hour sessions.

Well, I would conclude that this is a pretty big Red Flag that something new is now involved in the equation.  Is this absolute, "beyond a reasonable doubt", proof of a robot - well, no.  But I would consider it one hell of a Red Flag for a more detailed analysis, outlined below.

End Important Note **********
______________________________________________________________________

B - Total Duration of Play

I think that Mr. Bailey got this one going when he stated in the following post:

"I know a number of you feel that gambling for 14 hours straight ... is doable - especially in a land based environment.  Online?  I don't buy it.  Sorry, I just can't."

(His actual statement included the words "... 14 hours straight without a break ...".  I removed the words "without a break" because 1.)  later data analysis proved that this part of his statement was incorrect, and 2.)  the amount of break time is part of our next point of discussion - "Ratio of 'Break Time' to Total Duration of Play, as a percent".)

As with MaxD's Rate-of-Play starting point of 500 games per hour, this 14-hour duration statement was then followed by member after member (I won't burden you with the links) stating that 10, 15, 20, even 25 hours of play time was within their experience.  Several members even volunteered to contact Casinos at which they frequently played to get some actual data for their "marathon sessions".

Is 14 hours of play proof "beyond a reasonable doubt" that a robot was used?  Again, I, and MANY others, would say not.  As above, whatever the actual Total Duration of Play which only 1 Player out of 1,000, or 10,000, can achieve, it is certainly above 20 hours.
______________________________________________________________________

C - Ratio of "Break Time" to Total Duration of Play, as a percent
.
This particular performance indicator gets a little tricky, because it ties into the Total Duration of Play.  For example - 1 hour of play without any break is obviously not a problem, but 24 hours without any break is obviously a problem.

Further, this performance indicator as it applies to this specific Casino Club Player problem is really tough because the logs (the ones published in the CM thread anyway) only show the game's time value to the minute, NOT the second.

There are some posts which provide some "bathroom break data" and so forth.  But I think we're going to need to "wing it" a bit here.

Side Note **********

By the way, the Player that started this Casino Club thread, initially blocked from the forum by Mr. Bailey (as part of that by Bailey BlastTM), was then allowed to make further posts.  In the following post he stated:

"there i found enough breaks. in fourteen hours: eight times a 2 min; one time a 3 min; one time a 4 min; two times a 6 min; one time a 9 min; one time a 10 min; one time a 18 min; two times a 22 min. in all 110 min."

(My total in all of this is not 110 minutes, as the Player states, but rather 116 minutes.  14 hours is 840 minutes, giving this Player a "Break Time" to Total Duration of Play of about 14%.)

Thus, Mr. Bailey's statement, referenced above, in which the Player was supposed to have been "... gambling for 14 hours straight without a break ..." was made out of ignorance.  An amazing display of ignorance, actually, given that he was supposed to have already analyzed the Player's data before delivering his Bailey BlastTM.

End Side Note **********

And, just to complicate things even further, I'm going to try and bring in the maximum Rate of Play as well.  So, away we go.

In my "Max Rate of Play", I'm decreasing the rate value for longer durations for no other reason than the consideration of "mental fatigue".  However, I only go as low as 700 games per hour = a little over 5 seconds per game. 

In my "Min Total Break Time", I'm only thinking about bathroom breaks, snack breaks, "grab a beer" breaks, "FedEx Delivery at the door" breaks, and maybe even a couple of "make a cold cut sandwich" breaks.  I'm not thinking about "long phone call from Mom" breaks.

(Unfortunately, this blog's particular create/edit interface does not provide the ability to create tables.  Column alignment without tables basically sucks - things get all out of line.  So, I'll break it down based on Max Total Duration.  NOTE - This is contiguous play time, not play time spread across multiple sessions.)

Olympics Caliber Human Performance Values

Max Total Duration = 4 hours or less
 - Max Rate of Play = Average 1100 hands per hour (excluding break time)
 - Min Total Break Time (Percent) = 0 minutes (0%).

Max Total Duration = 4.01 to 6 hours
 - Max Rate of Play = Average 1100 hands per hour (excluding break time)
 - Min Total Break Time (Percent) = 5 minutes (1.4%).

Max Total Duration = 6.01 to 8 hours
 - Max Rate of Play = Average 1000 hands per hour (excluding break time)
 - Min Total Break Time (Percent) = 15 minutes (3.1%).

Max Total Duration = 8.01 to 10 hours
 - Max Rate of Play = Average 1000 hands per hour (excluding break time)
 - Min Total Break Time (Percent) = 25 minutes (4.2%).

Max Total Duration = 10.01 to 12 hours
 - Max Rate of Play = Average 900 hands per hour (excluding break time)
 - Min Total Break Time (Percent) = 30 minutes (4.2%).

Max Total Duration = 12.01 to 14 hours
 - Max Rate of Play = Average 900 hands per hour (excluding break time)
 - Min Total Break Time (Percent) = 35 minutes (4.2%).

Max Total Duration = 14.01 to 16 hours
 - Max Rate of Play = Average 800 hands per hour (excluding break time)
 - Min Total Break Time (Percent) = 50 minutes (5.2%).

Max Total Duration = 16.01 to 18 hours
 - Max Rate of Play = Average 800 hands per hour (excluding break time)
 - Min Total Break Time (Percent) = 60 minutes (5.6%).

Max Total Duration = 18.01 to 20 hours
 - Max Rate of Play = Average 700 hands per hour (excluding break time)
 - Min Total Break Time (Percent) = 80 minutes (6.7%).

Max Total Duration = 20.01 to 22 hours
 - Max Rate of Play = Average 700 hands per hour (excluding break time)
 - Min Total Break Time (Percent) = 90 minutes (6.8%).

Max Total Duration = 22.01 to 24 hours
 - Max Rate of Play = Average 700 hands per hour (excluding break time)
 - Min Total Break Time (Percent) = 100 minutes (6.9%).

So, this now becomes just a "look up list".  Given the Player's duration of play - if their average game count per hour is greater than the Max Rate of Play above and their total break time is less than the Min Total Break Time above then this Player is suspect.  (See Test D below for final confirmation requirements.)

Lookup #1 - The Player played pick-a-game for 13 hours, at an average rate of 800 games per hour, with a total "break time" of 50 minutes.  This Player's performance is not suspect.

Lookup #2 - The Player played pick-a-game for 13 hours, at an average rate of 1000 games per hour, with a total "break time" of 10 minutes.  This Player's performance is suspect.  (See Test D below for final confirmation requirements.)

(The only complication that I can see - let's say the Player has played 15 hours straight, then takes their allowed 50 minutes break in one shot.  They then re-enter game play mentally refreshed, allowing their max rate of play to jump from 800 games per hour back up to 900 or 1000 games per hour.)

If we look at the performance values for a 24-hour marathon session:

1.  I've got a human being playing at about 5 seconds per game across that entire 24 hours.  Given that they are spending the first 4 hours at 3 seconds per game, that represents a significant "mental fatigue" impact through the 24 hours to finally arrive at an average of 5 seconds per game.   (In other words, although their first 4 hours are at 3 seconds per game, their last 4 hours might be 8 seconds per game to get that 5-seconds-per-game 24-hour average.)

2.  I've allowed for 100 minutes (over an hour and a half) of break time.  That is a lot of bathroom breaks, snack breaks, and sandwich breaks.
______________________________________________________________________

D - All games demonstrate compliance with "optimal strategy"
.
This particular performance indicator got no discussion at all on the CM forum.  It was simply assumed, by everyone - CM Moderators, Players, Forum Members, Casino Reps - everyone!

I can tell you, as someone who has significant experience in both programming as well as statistics under their belt, that proving that all games were played using optimal strategy would be an amazingly huge bitch.  I'm talking Hoover Dam huge here. 

About two thirds of the way down the following post (which has also been linked above) is the only intelligent discussion (provided by the President of Galewind Software) that I have ever seen that addresses the specific problems with this performance indicator.  The critical bit is quoted below as "Item #2":

****************************************************
Item #2 - Statistically Perfect Game Play

The Casino has declared that the games were played using a robot, therefore those 10,000 games should demonstrate statistically perfect game play.

Who is going to spend the required time to look through 10,000 games, compare each of them to the required standard for 'statistically perfect', and determine 'yes' or 'no'? The Casino? Certainly not. Again, some independent source would need to be found.

I'm not aware of any software program for doing this. Someone would have to do this 10,000 game comparison manually. Wow! Brain freeze would set in by sample 1,000. What is the error rate for doing this kind of manual comparison? My guess - really large.

There is even the problem of the 'required standard' for comparison. I am aware of 3 different 'Optimal strategies' for Blackjack alone.
****************************************************

In our extreme example (24 hours, minus break time, at 800 games per hour) we're talking about either analyzing (with software) or comparing (with human eyes) almost 18,000 games!  That's a lot of games, a LOT of games.

(The President of Galewind Software states that he neither has this software, nor is he aware of where it may be purchased.  However, he also stated elsewhere in the thread that he didn't care about robot use in his Casino software, so he would not be inclined to build something like this, nor would he be inclined to do an exhaustive search for something like this.)

Perhaps one of the Certified Testing Labs (eCOGRA, TST, CFG, iTech Labs, et al.) has such a software program?  I checked - if they do, then none of them mention this anywhere on their web sites.

Even if they did, they would need a separate program for each game.  And each of these programs would need to be able to accommodate different definitions of "optimal strategy" for its specific game.  Otherwise, there would be a significant number of "false positives" and/or "missed positives" ("positive" meaning a deviation from "optimal strategy").

I also agree with Galewind's President concerning a manual comparison of game play versus "optimal strategy".  If someone were doing this comparison manually, "brain freeze" would set in after 1,000 games.  If the technician actually made it through all 18,000 games, the miscount problem, if any, would not be "false positives" so much as "missed positives".  That is, if the technician found a deviation then they would almost certainly double check to make sure.  It would be much more likely that the technician would miss a deviation entirely.

So, the availability of some tool which can prove that all of the Player's games demonstrate compliance with some "optimal strategy" is simply unknown.  No one is asking, no one is telling - the most challenging test of all, and absolutely everyone involved just assumes it is done and true.

In short - this writer, after a reasonably rigorous investigation, has concluded that there is no practical method currently available to prove "beyond a reasonable doubt" that a Player whose game count exceeds, say, 2,000 games played all 2,000 of those games with statistically perfect "optimal strategy".

A call should go out to all of the significant parties involved, or potentially involved, in this issue - Testing & Certification labs, casino software manufacturers, what have you - to either provide demonstrable proof of the existence of such a verification system, or agree that such a system is not available.
______________________________________________________________________

Well, if you have made it this far into this "Mega-Post", then either you are truly interested in this subject, or you're a speed reader of phenomenal abilities.

Either way, I think we've reached the point where we can start to sum up where we stand.  I'm going to label this:

The Bottom Line

1.  We've discussed that "Player Protection Watchdog" sites (In this case, CM) that used to apply at least marginally acceptable levels of integrity to Player advocacy conflicts "back in the day" have been transformed into "Gambling Protection Rackets".  (Player is "guilty until proven innocent", taking certain casinos simply at their word, Bailey BlastsTM of intimidation, ignorance and incompetence (and their subsequent cover-up) in addressing player conflicts, etc., etc.)

2.  We've defined a (totally subjective but better than nothing) Integri-MeterTM against which we can measure the actions of a "Player Protection Watchdog" in providing real Player protection against the claims of a casino that screws that Player out of their money.

3.  We've applied that Integri-MeterTM to one issue involving CM and BetFair (Casino Police "On the Pad"? Addendum), and gave CM an Integri-MeterTM rating of 1 in their handling of this problem.  (That's as low as our Integri-MeterTM goes.)

4.  We've applied, in excruciating detail, that Integri-MeterTM to another issue involving CM and Casino Club (Casino Police "On the Pad"?), and gave CM an Integri-MeterTM rating of 5 in their handling of this problem.

5.  Both of these problems (BetFair and Casino Club) involved claims by the Casino that the Player was using a "robot" to play their games.

5A - We've addressed the question:  "Are robots bad?" with the answer No.

5B - We've listed the games to which any casino's concern regarding the Player's use of a robot applies.

5C - We've addressed the question:  "If robots aren't bad, then what's the problem?" with the answer "Robots remove the 1% addition to a casino's bottom line that arises from human error."

5D - I think it is safe to say that we addressed the question:  "How do you prove that a Player is using a robot?" with an excruciatingly detailed answer.  We provided a simple "look it up list" with a matrix that included 1.)  Total Duration of Play, 2.)  Rate of Play (games per hour) and 3.)  Total Break Time (as a percentage of Total Duration).

And finally, we came to a wall through which no door is available - how do you prove "beyond a reasonable doubt" that all of the games that were supposedly played by this robot demonstrate compliance with "optimal strategy".  Our conclusion - you don't, no one has, you can't.  There is simply no door through this wall, and this fact is completely ignored by all parties.

Everyone assumes that it can be, and has been, done.  But this writer could find no shred of evidence proving this to be true.  This is truly the 800-pound gorilla in the living room.

I'm going to close this Part 1 (enjoy the Holidays, then move on to Part 2) by once again linking to a CM post to which I linked above.  Again, this is from the President of Galewind Software, and he had this to say:

"I have NEVER seen anything even closely resembling what I have outlined above provided as proof of a Player using a Robot.

IMO, I just don't think that it is possible. I mean, look at all of the variables, all of the complications, all of the unknowns, all of the guesswork that I've detailed above."

He then wrote an open invitation - "If anyone can find an instance where a Casino claimed "Robot Use" and the above level of proof was provided then I would be very interested in reading it."

There were no responses.


2 comments:

  1. "There were no responses."

    Of course there weren't.

    ReplyDelete
  2. The "Bailey Blast" was funny.but it was on the nose too, because I've seen some of these in other threads you haven't mentioned.

    It looks like you put a hell of a lot of effort into these posts because they are really long but they contain a lot of meaty content.Thanks you.

    ReplyDelete

All comments are moderated, so it may take up to 24 hours for your comment to appear.