Mathematical Proof that English Harbour is cheating

[soflat]

Oh, I thought they were randomly picking cards from a shuffled deck. I guess they can do whatever they want if it does not involve the impression they are dealing cards.

[spearmaster]

I was maybe tired and rushed. I did not even draw the distinction correctly - but just in case I did and it wasn't clear, the bonus game is NOT a doubling game - you pick a card and get a multiplier, that's it.

[HKGambler]

So now, according to Spearmaster, the buggy code was SOMEHOW uploaded by accident on May 13 to the live servers of various OddsOn-powered casinos: English Harbor, Hot Pepper, etc.?

So during the whole of those 19 critical days, not one, if not all, of the said joints had recognized or reported the discrepancies? And this need not be answered?

While Spearmaster has no idea why the upload had taken place, he is only pretty sure that the bug was never intended, how come nobody had requested for an explanation directly from OddsOn themselves? Why relying on Spearmaster finding the answers all along, as he alone is not able to answer some of the questions raised other than the codes at hand, as well as he sometimes has to double check for clarifications of some contradictions like the “Jan to May data period”? And by now he is seen to be very exhausted!

The source code is the property of OddsOn, who should be held totally responsible, who should have full knowledge as to what had been going on. Why had OddsOn been allowed to shut their mouth and be uninvolved? Has anyone who is helping pressed for one announcement from them?

I do not give credit to any of OddsOn or EH for most of what they did. The cooperation & good gesture they expressed to the third party for investigation, albeit prestigious, who either have past business dealings or current links with them; or the 120% restitution they had offered to the cheated, are only the most basic steps which they had no choice not to take, are far from enough to earn contentment of the players majority.

I’ll remain suspicious and criticising until I have most of my frustrations settled.

[Vesuvio]

JohnGalt - replace the word "cards" with something else and you should get the idea. We're not simulating a random card being drawn from a deck - or at least that's not the way I see it - we're drawing 4 elements from a selection of 52. These elements are weighted.

The patronising tone here and elsewhere doesn't do you any favours, spearmaster. If you read the posts you'd see we're as clear as we can be for now on how this phantom game would have worked - we just disagree with your overly generous interpretation of it.

You're asking us to believe that they would program a game to deal 4 cards (remarkably like the video poker doubling), but that these would be weighted. You haven't answered the logical objection that if explained to players this would destroy confidence in the cards dealt in VP, VP doubling and indeed all the other card games at the casino. How likely is it that they would ever really introduce such a game?

That's just one part of a weirdly unlikely chain of events:

1) This exciting new game code gets written in a game specifically targetted by bonus hunters (who would just ignore it, removing the most likely innocent explanation for making a change). It's unprecedented as a variation of video poker and unprecedented as a game which is going to openly deal weighted cards.

2) The code somehow gets uploaded to the server.

3) Remarkably rather than just doing nothing or producing a fatal error the new code integrates with the old and the game runs smoothly. The only outward difference is that the casino now has a large edge.

4) This flaw isn't spotted by any testing or audits for weeks (though you'd think if they were busy reprogramming VP there'd be some attention paid to it).

5) A player spots it and posts on Casinomeister and the software then automatically updates itself over the weekend to remove the flaw.

6) The casino, as casinos do, deny it, before realising that mathematically they're in a completely indefensible position. They then take the only option of trying to provide some sort of evidence of an innocent mistake, seeing as the alternative is to be 100% confirmed as cheats.

7) They refuse to provide logs going back over the last few years which might have the potential to blow their whole story out of the water.

Do you really think that anyone who finds this chain of events suspicious is biased and jumping to conclusions? You must realise the case you're stating isn't somehow self-evident. There's an outside chance it might be what actually happened, but for the moment (pending more details) no impartial jury would accept it.

[nafanny29]

As far as im concerned the jury is still out on this one!!

This is assuming for one minute that English Harbour were actually cheating.(im not saying they were but hypothetically speaking)

As they realised they were caught red handed they panicked and thought up a plausable explanation (ie the new feature excuse) They then got their programmers to write bogus code to back this up and pass it out for examination.

As I say this is all hypothetical but could this be possible?

[thelawnet]

3) Remarkably rather than just doing nothing or producing a fatal error the new code integrates with the old and the game runs smoothly. The only outward difference is that the casino now has a large edge.

4) This flaw isn't spotted by any testing or audits for weeks (though you'd think if they were busy reprogramming VP there'd be some attention paid to it).

5) A player spots it and posts on Casinomeister and the software then automatically updates itself over the weekend to remove the flaw.

This one is the unbelievable one isn't it.

They (claim that they) screw up their code. They notice it's screwed up and schedule it to be fixed.

Coincidentally that time is IMMEDIATELY after they are caught in public.

Just to get this straight:

They had the code in place for three weeks
They suddenly noticed that the code didn't work properly and was ripping off players to the tune of thousands of dollars per day
They realise they should fix it, but instead of immediately disabling the game they continue to leave the code they 'accidentally' released running, even though they know it's heavily biased against the player.
Instead of an emergency fix they leave the code running a bit longer, even though it appears from Spearmaster's analysis that it was shoddy and did not work properly, and was a serious liability - not an 'oops, we got the graphics a bit wrong error', but a whopping great 'we are cheating players out of thousands of dollars a day with broken code that isn't even finished'
No contact with the players. Nothing. What kind of person would see that they had this code in place and just do nothing about it? I would be seriously freaking out if I knew had released this kind of mess into a production system. I would be spending every second frantically working out what problems it had caused for the bank I work for and would not stop working on it until I had resolved the problem. But for English Harbour, it's no sweat, no investigation, just 'we don't give a shit', let's release an update to go out in the next release and hope that the mug punters don't notice.

And then magically the time that they scheduled this to be fixed is immediately after they are caught cheating on a public message board.

And then they send out this notice saying 'no, nothing was wrong', even though they knew perfectly well something was wrong as they had (they claim) scheduled the code to be fixed.

And then they subsequently say that their analysis showed nothing was wrong because they looked over too long a period, a claim which appears to be false - and despite the fact that they knew they had half-assed unfinished code with missing variables in place....

Whichever way you slice this there's no scenario that leaves them looking honest, decent or credible.

[ergopro]

Why had OddsOn been allowed to shut their mouth and be uninvolved? Has anyone who is helping pressed for one announcement from them?

Maybe because OddsOn is owned by EH New Ventures Gaming Inc. that operates all EH group casinos...

[Annorax]

If you don't mind, could you find an example of this standard for me?

In California, 5-card draw is legal. Poker variations with community cards are legal. But 7-card stud is illegal. What's the difference?

Under California law, draw poker and poker variants with community cards are considered games of skill, while stud poker is considered a game of chance, and thus is banned.

[spearmaster]

Originally Posted by spearmaster
JohnGalt - replace the word "cards" with something else and you should get the idea. We're not simulating a random card being drawn from a deck - or at least that's not the way I see it - we're drawing 4 elements from a selection of 52. These elements are weighted.


The patronising tone here and elsewhere doesn't do you any favours, spearmaster. If you read the posts you'd see we're as clear as we can be for now on how this phantom game would have worked - we just disagree with your overly generous interpretation of it.

There wasn't anything patronizing about that at all - certainly nothing intended, in any case. My whole point is that the nature of elements used in the bonus game were irrelevant - it was only after I wote that bit that I realized everyone was thinking that a dealer element was present - there is no dealer element in the bonus game. It is functionally equivalent to being asked to select one of four sticks, or colors, or whatever in the presentation. You do not have to beat the dealer to get your multiplier.

[quote]You're asking us to believe that they would program a game to deal 4 cards (remarkably like the video poker doubling), but that these would be weighted. You haven't answered the logical objection that if explained to players this would destroy confidence in the cards dealt in VP, VP doubling and indeed all the other card games at the casino. How likely is it that they would ever really introduce such a game?[/quote

See above. The doubling game must by nature include a dealer element - you must beat the dealer in order to win. The bonus game does not have a dealer element - it's more like picking a prize.

3) Remarkably rather than just doing nothing or producing a fatal error the new code integrates with the old and the game runs smoothly. The only outward difference is that the casino now has a large edge.

Smoothly is not the correct description. Just because a function throws an error doesn't always mean that execution stops. Functions usually return a null value - the problem is that the invalid return value triggered another comparison.

6) The casino, as casinos do, deny it, before realising that mathematically they're in a completely indefensible position. They then take the only option of trying to provide some sort of evidence of an innocent mistake, seeing as the alternative is to be 100% confirmed as cheats.

This is also invalid - because no casino has ever claimed anything but a bug in any software. EH simply allowed me to verify this fact in its code. Other operations might simply hope it all goes away and refuse anyone access to their code.

7) They refuse to provide logs going back over the last few years which might have the potential to blow their whole story out of the water.

Irrelevant. Considering the nature of the bug, it would not have been an issue in past few years unless somehow we allowed it to go undetected - and as the logs of April show, the game WAS normal before the bug was introduced.

Do you really think that anyone who finds this chain of events suspicious is biased and jumping to conclusions? You must realise the case you're stating isn't somehow self-evident. There's an outside chance it might be what actually happened, but for the moment (pending more details) no impartial jury would accept it.

It is not an outside chance at all. In studying the code it is plainly evident - but again this is not possible for me to explain clearly without a lot of skepticism, so as I said there is absolutely nothing more I can do than to post my interpretation of all the issues and let it go. I don't expect EH, or any other operator, will ever allow anyone to examine its code again because of the uproar it has caused. And I have already stated that I don't argue with your right to be skeptical - but then telling me that you don't believe what I am seeing, well, what do you expect from me?

[spearmaster]

Under California law, draw poker and poker variants with community cards are considered games of skill, while stud poker is considered a game of chance, and thus is banned.

Nice catch, Annorax But they are all still universally considered to be forms of poker...

Anyhow, this issue with the cards automatically constituting a card game is now irrelevant since I think we have established that there is no dealer element in the bonus game - by picking a card you either increase the size of your win, or decrease it, but you do not leave the bonus game with nothing since you don't have a dealer to beat.