Casino software caught cheating

[iGamingWriter]

Hey All,

I wanted to quickly alert everyone to a report on Amigotechs casino software I’ve just published - http://thepogg.com/skybook-cheating-...echs-software/

After we received a complaint from a player about odd results I subjected the player’s hand history to statistical analysis. The conclusion is that there’s simply no possibility that the results they received came from a fair game (in fact from 2 separate games).

I’d also be quick to point out that this is far from the first time that Amigotechs software has been found producing non-random results.

The operators that I'm aware of using Amigotechs software are;

AmericasBookies.com
BetMania.ag
BetPhoenix.com
BetRoyal.com
BetTrojan.com
BogartCasino.com
HorizonSports.es
PlayMe247.com
PokerHost.com
SBGGlobal.eu
Skybook.ag

I’d strongly encourage any affiliate working with any of the above properties to reconsider their relationship with these operators – cheating software isn’t just bad for players, it damages the reputation of the entire industry and is bad for all of us. Any casino that has to resort to cheating players likely wouldn’t think twice about shaving affiliates.

I’ll be directly contacting as many of the affiliate partners of the above sites as we could identify over the course of today.

ThePOGG

[ocreditor]

BetTrojan ?!... very inviting name...

[iGamingWriter]

BetTrojan ?!... very inviting name...

I'm going to guess they've been in business long enough that 'trojan' wasn't a word associated with computer viruses, but I've got to agree - that's where my mind went first.

[ocreditor]

I'm going to guess they've been in business long enough that 'trojan' wasn't a word associated with computer viruses, but I've got to agree - that's where my mind went first.

Well trojan is also refer to brave so should be a good name...seemingly

[-Shay-]

Thank you for all the research you put into that and for sharing.

[iGamingWriter]

It's just been pointed out to me that the main brand involved in this issue - Skybook (AffCorner) - are Silver sponsors here at GPWA. I'm going to flag this thread up for administrator attention.

[iGamingWriter]

I've received an email from one of the affiliates I've contacted confirming that they're going to pull their links to PokerHost. While I don't want to post up details of the affiliate in public, I do feel it's significant to say that this person indicated that they were also having trouble getting paid by this brand. Obviously I cannot confirm this and there could be more to the issue but given the situation I feel it's relevant information.

[ocreditor]

Do you know if any of these brands on your list associated with any of these aff program brands ?
http://www.commissionaccount.com/

[iGamingWriter]

Do you know if any of these brands on your list associated with any of these aff program brands ?
http://www.commissionaccount.com/

On gut instinct level I'd say it's certainly possible (i.e. I'm actively looking), but I have nothing tangible at the present time to suggest a connection. I'll certainly post if I find out anything.

These brands sprung to mind pretty quickly when considering the above operators and they're advertised by a lot of the same affiliates. That said they're all US facing (mainly) sportsbooks, so they're competing for the same market.

[Anthony]

It's just been pointed out to me that the main brand involved in this issue - Skybook (AffCorner) - are Silver sponsors here at GPWA. I'm going to flag this thread up for administrator attention.

Thanks for the notification. I will review your article and contact AffCorner.

[iGamingWriter]

To keep everything totally out in the open - the CEO of Amigotechs has got back in contact. They have stated that they tried to contact me after the last email I received (I've checked, we have no record of this email), but got no response.

They have requested a phone call to discuss this issue. While I do think discussion of the issue needs to occur, I have declined any phone call conversation. In this type of situation a written record of everything said is absolutely essential. I will be updating this thread as and when this develops, but in the name of fair representation of the situation I wanted to be open that Amigotechs are now aware and open to discussion of this issue.

[dfiocch]

Thanks, ThePogg. As usual, great work. Thanks for sharing. I will take a look. I have BetRoyal reviewed on site. Obviously... Blacklisted!

[-Shay-]

I really respect your approach in keeping things open and in writing. Far too many times, things are discussed and spun or deals are reached behind closed doors. You're indeed one of the good guys.

[TheGooner]

Just to understand the numbers - after reading the expose on your site - I'm still a little wooly on the stats.

How does the chance of "Three of a Kind or better" get to 10.9% in this game mate?
I don't know the specific game at all.

It's around 2.9% to get dealt 3oaK or better from any five cards ... (source : http://en.wikipedia.org/wiki/Poker_probability )
Plus a factor for receiving more cards but that's still a bit shy of nearly 11%.

Is the game in question a modified deck or extra cards or something ?

(I hate not understanding the full picture)

[iGamingWriter]

Hey TG,

The % is higher as you can discard and replace cards - this means more 3oaK or better hands will occur. The figures you're look at only consider being dealt 5 cards with no replacements. The figures I originally used were produced by http://www.beatingbonuses.com/vp_pay.php and are based on the assumption you play optimal strategy (you could easily inflate the 3oaK result if you played specifically to get that hand, discarding other winning hands to chase 3oaK). The default 2 Ways Royal paytable is the paytable that Amigotechs use. The figures that Eliot Jacobson provided I believe were produced with a similar calculator at the WizardofOdds, though I could be wrong about that.

[TheGooner]

Got it (after I worked out to hit the calculate button).


The maths that they use (optimising strategy according to paytable) is beyond me - I would not know where to start to work them out - but the percentages and house edge look to be robust and in the magnitude I'd have expected.

Thanks.

So after 560 games - the player should have seen 40 x 3oaK, 6 each of straight, flush and full house, and maybe a 4oaK.

Anything higher would have been a rarity, but that still offers an expectation of 59 hands with good payouts.
To see NONE - is extremely unlikely to naturally occur - as you've calculated.



----------------------------------------------------> 2 Ways Royal

Optimized House Edge:	0.20% (99.80%)
Default House Edge:	0.20% (99.80%)
Standard Deviation:	5.98

Final Hand	Pay	Percentage	1 in ...
Royal Flush	800	0.0025%	40216
Low Royal Flush (2-6)	800	0.0023%	44057
Straight Flush	50	0.0107%	9379
4 of a Kind	40	0.235%	426
Full House	6	1.15%	87
Flush	5	1.05%	95
Straight	4	1.13%	88
3 of a Kind	3	7.40%	14
2 Pair	2	12.9%	7.8
Jacks or Better	1	21.5%	4.7
Nothing	0	54.7%	1.8

[muffincrumbs]

My question is how many hands were analyzed? Must have been a huge number to prove it to be overwhelmingly unfair. My understanding is that it is hard to prove randomness except on very large scales, and I figured a couple of players would not be enough to prove anything.

[TheGooner]

ThePogg has correctly calculated that the chances of ZERO hands of 3oaK or higher is 1 in 18867924528301900000000000000

The reason the number blows out to such an unlikely probability after "just" 560 games is because the probability is nearly 11% each and every time. The player is expected to win a bigger hand roughly 1 in 9 or 1 in 10 ... so to be 0 from 560 is hugely unlikely.

Hugely unlikely ... it's such a big number that humans don't understand the number ...
My real world example it is like a coin 94 times and always coming up heads.

Try it sometime - I don't think that you will ever get to 10 straight heads.

-------------------------------

Taken from ThePoggs onsite analysis :
The odds a player buying a single ticket for three different National Lottery draws and winning the jackpot on each of the 3 draws is 1 in 2734494804479010000000.

-------------------------------

The problem is we get big numbers and our brain fades away.

One big number- then another big number - the human mind boggles. It's about the same isn't it?
NO!

Here is another way to look at it ...
- 1 in 2734494804479010000000 to win lottery 3 times.
- 1 in 18867924528301900000000000000 to not get at least 1 3oaK or better in 560 hands
- 1 in 19807040628566100000000000000 to coin toss 94 consecutive heads.

There are SEVEN more zeroes in the players losing streak than the lottery winner. That's approximately 10,000,000 or ten million!
This means that it is roughly 10 million times more likely to win the lottery 3 times than to have that online casino losing streak.

WTF?

Now - did we just see a very unlikely losing streak?
Or was the software malfunctioning / configured to not allow certain payouts?

You feel free to decide.

(The only thing I will say was WHY did the player keep playing ???)

[muffincrumbs]

Ok I see, and between you and I, we both see it is not fair, but just to argue which I love to do I doubt it would stand up in court or something like that. Yes there is no record of anything statistically that impossible to achieve I could find expect one.

The estimated probability of Life spontaneously forming is 1 in 10³⁹⁰, yet some people believe it happened while others do not. That is 390 zeros. I only see 29 zeros above.

[mojo]

Nice job ThePogg. I can appreciate the time and energy put in for the good of everyone.

I'm not so great at doing the math but it appears this isn't Amigotechs first rodeo.

Eliot Jacobson had a similar investigation in 2012 at WOV. Amigotechs Online Gaming Solutions Warning

I conclude that Amigotechs, the vendor who supplied the casino software to Youwager, is responsible for providing a product that was programmed to act in a malicious fashion to the detriment of the player and that this software did act in a malicious fashion on 19 occasions.