The Illusion of Randomness in Video Games

Today I will write to those who are challenged with the thought process of regulating pay to win games.  This can be a difficult process when looking through the lens of research, reporting and public opinion.  Its similar to seeing the forest through the trees.  For this piece I will try and highlight things from a developers perspective.  I will be shining a light on what has been done by game developers, and what can be done in a manor that is considered unfair by most people with respects to randomness.  This does not mean that every developer employs these tactics, but that some do and will, because there is a vacuum of rules and oversight.

When regulating pure gambling games, there is a pure sense of odds, or randomness that is looked at.  It is usually very simple and defined.  A deck of cards is 1:52.  There are specific accepted and enforced game rules, that use simple odds calculations that can be enforced.  You can audit and review that the electronic gambling game does not pervert the odds illegally.

Video games however follow no rules, except to use randomness as a tool to balance a game, and in games with in game transactions, to monetize it.  There is no rule of law that states that every action have the same randomness from instance to instance, or player to player.  The video games are much more complex in where the randomness happens.  Each game can be completely new, the same with different formulas and skins, or a progression of another.  The only limit is the developers imagination.

For an example and demonstration of odds manipulation in a game.  I took a look at a simple game play design.  I ran a test on King of Bubble.  This is a simple type of  micro-transaction game that involves aiming a colored bubble at a group of colored bubbled to match and pop them.  The aim is to clear bubbles and move to the next level.  This is a simple explanation but to the point.  The game monetizes by selling lives that can be bought rather than waiting for them to refresh over time, watching ads or paying to remove them, and buying power up items that aid in game play.  This game was chosen at random and for its simple nature.  Other games exist similar, using variations of the theme and methods.


For the test I picked level 6.  I figured a few levels in, to test past the intro level mechanics.  I played this level 38 times taking starting screen shots of each play through.  The idea was originally to see if my starting ball color was randomized sufficiently to show the odds where not tampered with.  I discovered another controlled element that was easier to identify and demonstrate.  In the game you have two balls lined up to use.  You may select the alternate, or shoot the one that is next.  The game at level 6 uses 5 different color balls.  This means the odds of getting any particular color should be 1:5.  This also means the odds of getting two balls of the same color in the shooter are 1:5, assuming you do not care which color is the first being matched.

The test showed that 16 out of the 38 times played, the first two balls where a match.  This means odds where presented at 16:38 or a little less than 1:2.  This is far outside the natural odds of 1:5, and proved that the developers tweaked the odds.  This is not a statement of proof that this was done to increase monetization or improve game play balance.  This does show that randomness is completely controlled by the developer, and is not exclusive to loot boxes.

We have no way to know if the aim and deflection of balls off the walls had a prescribed “random” accuracy formula attached.  Random variation of aim is often used by developers in shooters to warrant a better gun.  A lower level gun will randomly be off in its aim by a greater degree than a better weapon achieved later in the game.

Famous game designer Danielle Bunten Berry had the following to say in an August 1992 interview1.

Asked how [she] balances [her] games, [she] opted for a one word answer, “Cheat.” Asked how [she] would defend that practice if a gamer caught [her], [she] said, “Lie!” in a folksy rhythm of misdirection. Then, [she] went on to reveal some of [her] actual methods. For example, in M.U.L.E. and Global Conquest, there are random events which can affect the flow of play. The programs are constructed, however, so that players in the lead can never have good luck and the trailing players can never have bad luck. Only for the players in the middle are the events truly random.

Gamers who play Mario Kart have dug into the code and published2 very detailed charts that show odds of getting specific power ups based on the distance a player is from the lead.


Another trick used by mobile developer MZ is concealing the amount of items needed to reach certain levels.  They advertise purchase of packs that guarantee a certain level, and promote getting to the highest.  The amount of special items needed is not only obscured, but increases multiplicity each level3.  This is not randomness but has the same effect of randomness.  It leaves the player hoping the next purchase will be enough, and purchasing based on a guess of weather its probable.

You can not regulate randomness in games, they are core in many designs, and near impossible to identify completely and understand.

It is still part of the addiction and monetization of pay to win games.  This is where randomness and its illusion in many instances, becomes a public concern. Games can use randomness to their monetary benefit.  A well known Activision patent4 describes perverting randomness in player match-ups.

A system and method is provided that drives microtransactions in multiplayer video games. The system may include a microtransaction arrange matches to influence game-related purchases. For instance, the system may match a more expert/marquee player with a junior player to encourage the junior player to make game-related purchases of items possessed/used by the marquee player. A junior player may wish to emulate the marquee player by obtaining weapons or other items used by the marquee player.

So in conclusion.  When regulating pay to win games, you need to look at what happens before, after and around the loot boxes.  They are an end to a means.

  1. Computer Gaming World Issue 97
  2. Mario Kart 8 item probability distributions
  4. System and method for driving microtransactions in multiplayer video games

5 thoughts on “The Illusion of Randomness in Video Games

  1. Doug,
    An excellent post, thank you.

    To establish true odds and randomness levels in loot boxes and gameplay would take an individual thousands of hours of data recording and even then, the results would be greeted by the gaming industry as incorrect and nobody be could be proven wrong.

    I play a fair bit of WSOP poker. I also play poker in real life with real people. The amount of full houses, flushed and straights being dealt in the WSOP game hugely outweighs the chances in real life. Like an action movie; everything is created to be more dramatic, to make it more exciting and to encourage players ultimately to spend real cash on game chips.

    When all said and done, the willingness to continue playing is the choice of the gamer…but they can be heavily persuaded and influenced by the greater power of the gaming industry. Many become addicted to such games and the unfortunate ones pay out money they cannot really afford in an attempt to achieve, impress and enjoy the game and the online friends they have made. In both cases they continue to make these relationships deeper as they progress further into the game.

    Of course this is not a trap that’s been baited by the gaming industry…but it sure can look and feel like it at times!

    David (Monty’s Pythons)

    1. Interestingly I checked out WSOP to see a little more about it. It is what is called and rated as simulated gambling. As you described, they are not bound to uphold any natural odds or randomness. They are also not bound by gambling law.
      The little disclaimer at the bottom of the site read as follows, “This product is intended for use by those 21 or older for amusement purposes only. The games do no offer “real money gambling” or an opportunity to win real money or prizes. Practices or success at social casino gaming does not imply future success at “real money gambling.”

      That makes me wonder how many people believe that because they are successful at the game, they can win at a real casino. The game perverts odds to keep players, playing, and spending. Real casinos will not let you win to keep you hooked. Does the game explain this clearly? Or is the disclaimer, that doesn’t explain why, the only warning?

  2. Doug, the game explains very little; at least not with clear transparency.

    I enquired with WSOP customer support on the subject; asking whether or not the cards were tampered with in any way and for any reason. Their answer:

    ‘Kindly note that we are not allowed to tamper with the cards, since they are dealt randomly, via a certified random shuffling algorithm.

    You can read more about it by following this link:’

    The main revenue is via the selling of in-game casino chips for real money. So, there is no gambling for real money as such… but as your game account grows bigger with experience points and in-depth performance stats, a player becomes engross and treats the in-game chips as if they were real money chips. And when they run out, there would be a strong desire to spend money to buy more.

    There are also plenty of other in-game marketing strategies; spinning a bonus wheel, slot machines of varying values and so on.

    So, if there were an incentive for the company to tamper with the cards, it could be to make the games more exciting or to identify likely spenders, build them up then drop them down just when they’d be most likely to buy more chips…like when they’re approaching a big in game progress milestone.

    One WSOP hand I remember well…
    I was holding AA. By the turn there was 3,7,K,K. Betting was high and when the river card was dealt and I saw another Ace I felt I was pretty much unbeatable. A full house AAAKK is a pretty rare hand! In fact the ONLY thing that could beat me was if my opponent was holding KK.
    I went all in and my opponent called – and guess what, I was beaten by four of a kind!

    I didn’t play the game for about 3 months after that as I lost faith in it. I’ve never spent any real money on the game. I just know now to expect the unexpected !

  3. Quoting a developer from 1992? Really stretching and scraping to make an argument if you need to take quotes from almost 30 years ago

    1. Actually it shows that faking randomness as a tool by developers to effect things in a game has been around and used for a long time.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.