Playtech - Pick 'Em Poker Game Error

This article first appeared on the 10th of August, 2015

It was recently brought to our attention by the owner of BeatingBonuses.com due to posts on their message board (http://www.beatingbonuses.com/forums/showpost.php?p=92051&postcount=17) that one of the Video Poker games at Playtech casinos has been malfunctioning. The game in question is Pick ‘Em Poker.

The Basic Problem

All Video Poker games are dealt from a single deck of cards. After placing a wager the player receives their cards. They then select the cards they wish to hold and the cards they wish to discard. When the player discards cards, as would be expected from the name, the cards are placed in the discard pile and cannot be re-dealt that round. For example, if you received the ThJhQhKh4s and decided to discard the 4s, the re-deal could not produce the 4s.

This is the same across the board in all poker games.

The Pick ‘Em Poker games at some Playtech casinos are re-dealing discarded cards under certain circumstances. This has a significant impact on the House Edge and shifts it substantially away from the figure advertised in the game help file.

The Details

To demonstrate this error we tested at Paddy Power casino using the 100 hand version of the game (if an issue was present it would present far quicker playing the 100 hand game variant).

(Credits for video: ThePogg)

In 10 rounds of play, in every single round the discarded card was re-dealt at least once. The results of play are summarised below;

After establishing that the error did occur we looked at the single hand variant of the game;

(Credits for video: ThePogg)

After 300 rounds of play the discarded card had not been re-dealt once. While this isn’t mathematically certain, it seems fair to conclude that the error was not occurring in the single hand game.

Testing different numbers of hands, the error occurs on all multi-hand games, but seems not to occur on the first hand dealt in any multi-card game, which would tie in with the single hand game preforming correctly.

We then looked at a different casino to confirm whether this issue is across all Playtech casinos or only at specific venues. Testing at Boyle casino returned no re-dealt discards on the 100 hand variant after 10 rounds of play;

(Credits for video: ThePogg)

The fundamental difference appears to be that Boyle, at the time we tested, were running on version 14.10.18.2 while Paddy Power were running version 15.1.15.1 (the version number can be found at the end of the url on the game help file).

What impact does this have on the player

When Pick ‘Em Poker if functioning correctly, using the paytable offered by Playtech casinos and playing optimal strategy, the House Edge of the game is 0.05%.

Re-dealing discards on a single hand game would increase the House Edge of Pick ‘Em Poker to 0.98%*.

Working on the assumption that the first hand is never affected the House Edge for the different hand numbers are as follows;

*This figure was derived by the owner of BeatingBonuses.

As can be seen above, this error significantly increases the House Edge over the advertised 0.05% in the game help files provided by Playtech, multiplying the House Edge by almost 20 for the larger numbers of hands.

How did Playtech respond?

Playtech took several weeks but did respond to our communications and did issue an update to fix the problem. That’s a good start. We were told that from this point on it would take some time to realise updates as and when individual operators installed them.

At the time of publication of this article, Boyle casino was no longer showing this error but Paddy Power casino was.

We would advise players to avoid playing Pick 'Em Poker until such time as this issue is fully corrected.

What Playtech would not do is discuss in any way their testing procedures or how a game with this type of error came to be missed by their testing procedures, instead simply saying something to the effect of ‘accidents happen some times’. This is disappointing as from our perspective this is the crux of the issue as we’ll discuss in a moment.

Conclusion

It’s clear that this issue has mislead players and ultimately short changed them, however I also feel it’s fairly obvious that this hasn’t been an intentional attempt at deception. The inconsistency in the application of the re-dealt discards suggests an accident rather than intent especially as Pick 'Em Poker was the only game impacted.

Inadequate Game Testing

There’s a bigger problem here that this issue is just a symptom of and that is inadequate game testing at Playtech. Playtech are not only not unique in this, they are the norm. At the same time we’re publishing this article we’re also publishing an article detailing a game error encountered while playing a Baccarat game at both Dragonfish (888 group) and Rival software that again is the consequence of insufficient testing before the game was deployed. Michael Shackleford who is commonly known as the Wizard of Odds, published an article about Soft Magic Dice software that doesn't grasp that A2345 is considered a Straight when playing poker.

None of these groups are unusual in these failing and they shouldn’t be singled out for scorn, though that does not mean they shouldn’t be criticised for these failings. The entire industry needs to reflect on how and when games are tested and all software providers need to ensure that their procedures are robust. These sort of slip-ups hurt everyone involve and undermine trust in the industry.

Casino operators do not have the time or resources to test every game they offer to players. More than simply the time and resources, casino operators doing this job would be a phenomenal waste of resources, as hundreds of operators repeat the same testing of the same games over and over again.

Game testing has to be carried out by software providers.

Up till this point the industry has held to their ‘testing lab certificates’. On a common sense level, if these testing labs were providing the service that players are being told they are issues like this wouldn’t occur. They are occurring so clearly something isn’t working out like it should.

With regard to this particular issue, had this specific game been properly tested at the time it was updated (rather than a general test on the RNG or testing of previous versions) it would have very quickly become apparent that the game results were not correlating with the expected outcomes.

The Dragonfish and Rival issues are the same – had the Baccarat games in question been properly tested or had someone with a sound understanding of the fundamentals of gaming mathematics been on hand for the Game Development team, it would have been easy to establish that the advertised odds did not correlate with the actual return.

As stated earlier, these issues should not be used as clubs to beat Dragonfish, Rival or Playtech. The issues aren’t specific to these companies. This is an industry wide failing and an issue that the entire industry needs to address. What is more questionable is the lack of response and transparency that’s been apparent over the two issues.

There’s a very basic solution to these problems – software providers need to review their testing procedures and ensure that thorough and rigorous testing is conducted on each and every game both at the time of launch and before any update made to the game is rolled out to players.

As things currently are, these types of game issues often cause significant damage to the reputation of the casino provider where they’re detected and the software provider that developed the game. They also leave players out of pocket. The ‘win/win/win’ situation for software providers, operators and players is for software providers to place greater focus on their responsibility to ensure the products that they provide operate in a fair fashion that conforms to the standards that are advertised. These sorts of slip ups wouldn’t be accepted in the well regulated offline markets and shouldn’t be acceptable online.

Acknowledgements

We would like to thank the owner of BeatingBonuses and Michael Shackleford their assistance with reviewing the analysis and conclusions of this article.