Virgin Casino (Wagerworks software): forty years of non stop play needed to achieve these results in a fair game

Virgin Casino, franchise of the famous Virgin brand headed by entrepreneur Richard Branson, uses software provided by Wagerworks.

Yesterday, I wagered just over �6000 pounds at Virgin Casino, on a slot machine using initial starting bets of �1. In total, I made 1950 initial �1 bets.

After wagering those 1950 bets at �1 a pop, I had lost �1520.

This represents a loss rate of 78% on those initial wagers, and an overall loss rate of 23%.

The theoretical loss is �77, or 4%.

The probability of a losing �1520 in �1950 in initial wagering at �1 a time in this manner is one is a little less than 100,000.

This represents in the region of two royal flush "cycles" in video poker. And this is not just one hand, but 1950 hands.

As such, if you were to set out to attempt to replicate this monumental loss, you would need to play nearly 100,000 series of 1950 hands.

This would require a total of initial hands well in excess of 100,000,000.

Assuming it takes about four hours to wager in the region of 1500 hands in this manner, it would take in the region of 400,000 hours of play to achieve such a result.

This would take around 40 years to achieve. This is half the life span of the average human being.

There are some who will, quite rightly, say that anything can happen, and, as such, it is possible to see a result that would take half an average human lifetime to replicate in a fair game. That is the pragmatic stance.

My own stance?

I prefer to remain realistically and healthily cautious. In the face of such dismal evidence, it is not unreasonable to assume that there is a perfectly good liklihood that Virgin Casino, and WagerWorks software, is either 1) unfair or 2) can be adjusted to deliver an unfair game.

I will have solicited Virgin's feedback on this matter, and I will confront WagerWorks at their booth at the 2009 International Casino Exhibition. I will report back on their comments.


UPDATE January 20th

Virgin has got back to me. Since my results on the day were off the wall, they've looked at my overall play. Selected highlights:

On 14 January, your payout was - as you found - towards the lower end of what could be expected. If we take the whole game into account, the expected payout for the number of bets you made should be (using a 90% confidence interval) at least 76.7%. This means that your payout was within what could be expected, albeit unfortunate.

If we take the Gamble feature from the game specifically, this does not vary by game and so we can look at it across all games (you have also played Texas Tea and Wheel of Fortune Triple Action Frenzy). For this, your payout rate is 99.7% over 15,976 bets, slightly lower than the expected long-term payout of 100%.

Across all games that you have played, including the Gamble feature and the base game, your payout percentage is 99.83%. This is higher than the expected payout for any one of the games you have played, let alone the actual expected hold for your combination of games.

This doesn't tell me much I didn't already know, although the numbers are certainly interesting.

Does this prove that WagerWorks is rigged? No.

Does it prove that WagerWorks isn't rigged? No.

In fact, it doesn't say anything conclusive one way or the other. It's never possible to determine, taking results after the fact, whether software is rigged to deliver an unfair game, whether the odds are 100,000 to one or 10,000,000 to one.

The fact remains that in order to replicate an event as extreme as the event I experienced, it would be necessary to undertake my 1950 wagers close to 100,000 times.

The fact remains that this would take one person, playing non-stop, half an average human lifetime to achieve.

And in fairness, the fact also remains that my overall payout is 99.83%, which is better than expected.

Although I cannot prove that Virgin, or WagerWorks, is rigged, I remain sceptical. The software may be configfurable, when desired, to deliver an unfair game. Until definitive proof can be offered that proves Wagerworks fair, I will continue to treat it with extreme caution.


In 2004, a software project called "Fairdice" began development; its purpose was to establish whether online casino software was, in fact, fair, by combining numbers from random number generators both casino side and player side - there was some discussion in the Winneronline Fairdice thread.

Sadly, the project never got off the ground and appeared to close up. This is a pity, as it would have been an opportunity for online casinos to prove, once and for all, that their software was honest.

UPDATE February 23rd

Another weekend of heavy wagering on Wagerworks slots yielded less bad, but still disatrous, results: over �1000 lost on approximately �9000 initial wagering, excluding the gamble feature.

On the gamble feature, on the last day's play after logging in:

A few wins and losses.

Then eight straight losses.

One win.

Another eight straight losses.

A couple of wins.

Another eight straight losses.

That's twenty four losses with three wins.

When I did manage to string together a handful of wins, these were at the smallest and second-smallest, win denomination.

Each and every such bet after a bonus round (representing a largish win) lost.

I've given up on any remaining hopes I may have held out that the Wagerworks "gamble" feature, a theoretical 50/50 bet, is not either selectively riggable on an account basis or generally rigged as standard across all accounts. If it is "selectively rigged", then there is almost certainly a precedent for this in the Casino Bar test.

Hopefully, a full examination of all my gamble feature logs across all Wagerworks casinos will bear this out, although nothing will ever be proven. I will report on this as soon as such an analysis has been carried out.

UPDATE September 8th

Although I no longer play at Virgin, I've played the same slot and gamble feature elsewhere. Over just under �13,000 wagering I'm currently ahead �99. In addition to this, just prior to leaving Virgin for good I won �100 or so on a relatively small amount of wagering.

I'm hoping that the monumental loss that prompted this article was an aberration, particularly in consideration of the fact that my overall return from Virgin, at 99.83%, was very good.

Post a Comment

May 2005 | June 2005 | July 2005 | September 2005 | October 2005 | November 2005 | December 2005 | January 2006 | February 2006 | March 2006 | April 2006 | May 2006 | August 2006 | October 2006 | January 2007 | February 2007 | March 2007 | May 2007 | June 2007 | July 2007 | January 2008 | February 2008 | March 2008 | April 2008 | June 2008 | July 2008 | September 2008 | October 2008 | December 2008 | January 2009 | February 2009 | March 2009 | May 2009 | June 2009 | July 2009 | August 2009 | September 2009 | October 2009 | November 2009 | December 2009 | January 2010 | February 2010 | March 2010 |