Randomness and Game Design

For thousands of years, we’ve relied on randomness of various kinds to help our interactive systems work. While there will always be a place for randomness of all sorts in some kinds of interactive systems, I believe the current assumptions with regard to randomness in strategy games are largely wrong.

The major point I’d like to make is that noise injected between a player’s choice and the result (here referred to as output randomness) does not belong in a strategy game.

dice

What is “randomness”?

For the purposes of this article, randomness refers to “information that enters the game state which is not supposed to ever be predictable.” The process by which random information is generated is designed to be something that humans can never figure out. Classic examples of random systems are rolling dice, shuffling cards, or random number generators.

Technically speaking, a die’s rolling pattern is not actually “random”. It’s simply responding to physics, and a computer could take information about how a die was thrown and predict the number that would come up. We use dice precisely because a human being can’t do that. In fact, when we incorporate dice into our game designs, we do it under the assumption that no human will ever be able – nor likely even try – to predict the outcome.

In fact, trying to actually predict how the die will roll, by perhaps carefully tossing it with a specific, intended trajectory, so that it rolls to a side you intend, would likely be called out as “cheating” by any observers. The whole idea with a die is that you’re not supposed to know. It is noise that must remain noise, forever.

Part of the reason for this is the fact that we’re actually dealing with two separate, closed systems in a game that contains randomness. A rolling die is a closed system of its own that really has nothing to do with the greater game system.

This is distinct from other kinds of “unpredictable” or “uncertain” events. In chess, for example, players have some limit to the number of turns they can look ahead. Beyond that point, the events that occur are indeed unpredictable for that player. However, players can and do learn to look further and further down the possibility tree as they get better at the game. Part of the skill of chess is being able to explore that ever-increasing possibility space and come out with more predictive ability.

So while chess does have unpredictability, it does not have randomness. All games must have some kind of unpredictability in order to function, but randomness isn’t the only way to achieve that. Chess’s source of unpredictability – a highly complex game state – is unlike a random source in that it can slowly be chipped away at and understood.

 

Types of Randomness

Randomness can be separated into two categories: input randomness, and output randomness.

  • Output randomness – when we think of randomness in games, we’re usually referring to this. Output randomness is noise injected between the player’s decision and the outcome. Examples would be the dice roll combat in Risk or Memoir ’44, or the random number generation combat in X-Com or FTL. I will refer to systems that do not have this type of randomness as “deterministic”.
  • Input randomness – this type of randomness informs the player before he makes his decision. Typical examples of input randomness would be map generation in Civilization or Rogue, or face-up tiles or cards in a worker placement game like Puerto Rico or Agricola. (People often use the term “procedural generation” to refer to this kind of randomness in digital games.) This article will not focus on this type of randomness, but it’s important to know the distinction.

Interestingly, while these two types are certainly distinct enough from each other to warrant the classifications, they do technically exist on a continuum. Without going into much detail on it, it should be noted that irresponsible use of input randomness – where the player has very little time to respond to the new information, or where the game generates problems of wildly varying difficulty match to match – causes similar problems as output randomness.

 

The Strategy Game Learning Engine

Strategy games are engines that allow us to understand them. We play a game, we win or lose, and we make connections. “Oh, I see!” we say as we figure out some element of how the system works. For evolutionary reasons, we find this process enriching and entertaining. This is the “essential fun” of strategy games (largely the premise of Raph Koster’s book, A Theory of Fun for Game Design).

Let’s break down the process further.

fig110

  • Informing the Player – The player takes a look at the game state, trying to figure out what move to make. He is informed by his “skill” database – the collective total observations about the system and how it works that he’s made up until now.
  • Deciding the Move – A move is chosen, and the action is taken. As a result, the game state is changed. Alternatively, this could be “deciding the strategy” – a series of moves that collectively adds up to a larger strategic gambit.
  • Feedback in Outcome – Over the course of the rest of the game, the system responds to this input. A series of events take place after that decision, including the final win/loss event; all of which serve as feedback for the player, highlighting some causal relationship between them. Feedback also comes following a strategy, or at the end of a game.
  • Recording Skill – The player observes and records this cause-effect relationship and records it to his database. The player can then use that skill to make moves in the future. (Notably, this moment is where the essential “fun” of strategy games comes from, but it of course relies on the rest of the machine to function.)

As a player plays a game, over many matches, he builds to this “skill folder” and becomes a stronger player. In a shallow game, there might not be very many of these moments, whereas a very deep game can continue delivering these moments for decades if not lifetimes. This is generally why it’s considered a good quality for games to be strategically “deep”.

How do we achieve that depth? Well, the first way, which all game designers already understand, is emergent complexity. In order to create complexity, we design our games so that they generate complex emerging situations throughout play. A bishop, knight and rook against three pawns and a queen is not inherently complex; there’s a very small amount of data there. However, unleash these two forces on each other on a chessboard, and the amount of possible situations that could emerge is huge.

 

Complexity Effectiveness

The second method for achieving depth is, as far as I can tell, not understood by most designers today. This method involves being aware of complexity effectiveness: the amount of correlation between a state, and the history of past states.

A strategy game only has a finite number of states throughout a match. From what I can find, it seems that the average number of moves in a chess game is somewhere around 40, for example. A real-time game doesn’t have discrete “turns” per se, but there’s still a finite number of meaningful states, no matter how you divide it up.

If your game is a continuous series of events that lead deterministically from one to the other, then you are maximizing the amount of unique situations that can occur. I think this idea is counter-intuitive to many, who think that random events occurring somewhere in there must increase the amount of unique situations. However, the opposite is actually the case.

Having a system be entirely deterministic causes your emergent complexity to be maximally effective. This is because each emerging situation is given the maximum amount of contextual nuance by all of the events that came before and after it.

In the deterministic game, the current game state has ties to every part of the entire timeline. Because of that, it is being pulled into a more complex and more unique shape. What this is illustrating is the way that context, when deterministically related, provides meaning to a game state.

Of course, even highly random games do have some deterministic elements that do provide some context to game states. For instance, in a game like Summoner Wars (a turn-based wargame involving dice roll combat), the health of your summoner and the positions of units are both relatively deterministic and do provide some context for game states.

However, the vast majority of contextual information in a game no longer has meaning. I attacked your unit, and I rolled the dice. It came up as a “miss”, and then next turn you killed that unit. That event – you killing my unit – is not deterministically linked anymore to the actions I took beforehand. What happened was that I took an action, then something random happened, and then you took an action. The tie has been severed, and we can no longer use my move as contextual nuance for our current game state. Your game is now no longer “A, therefore B, therefore C“. Instead, it is now “A, then B, then C“.

The most significant bit of feedback is the goal-state. Once a match has ended, that win/loss condition sends a charge backwards through the course of events, revealing a positive or negative charge for every event that led to it. This move was somewhat good because it led to this, which led to that, which led to this, which led to that, which led to my win.

This is not to say that when a player wins, all his moves were good moves. However, it does provide an anchor point that informs every other move. Of course, moves are made in an attempt to get the player as close as possible to the win state. Once the match ends, we can now see how and why each of those moves was effective. (Because of this, players can get a lot of the same kind of fun out of watching a replay and analyzing it as they can from playing the game.)

Overall, after playing a deterministic game, a player is left looking at a coherent strategic picture that has been painted over the axis of time. Alternatively, the non-deterministic game could perhaps be considered more like a number of incomplete pictures. In this way, the deterministic game maximizes its complexity effectiveness, and the non-deterministic game does not. The non-deterministic game is adding complexity, whereas the deterministic game is multiplying it.

 

Imagined Depth

Output randomness does not increase the depth of a game. How could it? There is nothing to explore in a dice-roll. We all know that the odds are 1/6 for any face coming up. There is literally nothing else to know or explore.

What it actually does is obscure the outcome. You may have played perfectly, and still lost. The game has now sent you off on a wild goose chase, thinking about where you must have messed up, when in fact your play wasn’t the problem; dice rolls were.

Because of that wild goose chase, the game seems more complex than it is. The game provides unreliable feedback, and only after playing many, many games will it become clear which feedback you should ignore. Essentially, random games delay learning – the essential fun part of games – by injecting false signals into the engine. It’s a super-cheap way to create the appearance of depth, which is why it’s incredibly tempting for game designers.

Humans are pattern-seeking animals. We see figures in the clouds, we see images in the static, and we see conspiracy where there’s only coincidence. The reason is due to the fact that it’s evolutionarily favorable to think this way. The same quality that causes a person to think he saw a ghost in some rustling bushes is the quality that causes a person to think he saw a lion in some rustling bushes. And over time, those who thought they saw a lion were the ones who escaped when there actually was a lion. Those were the people who passed their genes along to us.

Rorschach_blot_01

For this reason and others, we’re now both cursed and blessed with seeing patterns everywhere we look, and game designers have been exploiting this in us for as long as games have existed.

Gambling machines have always relied on psychological tricks to exploit us into playing them. In order for anyone to actually want to play something as vapid as slots or roulette, some degree of self-deception has to take place. On some level, the player has to feel like he is responsible if he wins. Otherwise, how can they be invested at all? From ancient religious superstition (the Gods are angry at me!) to their more modern counterparts, like “blowing on the dice”, kissing “lucky” items, or other self-deceptions such as the gambler’s fallacy, we find ways to attribute meaning to events that are actually pure noise.

Serious players of highly random strategy games tend to be skeptical that this same trick could be working on them when they play their Summoner Wars and their Hearthstones. But why? If players are able to perform this trick on themselves in a system that has no strategy at all, it seems very easy to believe that such tricks would work on a smaller percentage of the overall system. In fact, baking random elements into a strategy game makes it all the easier to conflate noise and strategy feedback, because some of what happens in the game really is strategic and deterministic!

In these games, there is the actual skill of the game, but then there is also an additional “phantom skill” amount, which makes the game seem vastly deeper than it is. In actuality, most players probably have the system close to solved somewhat quickly, and the randomness is the deciding factor.

 

Counter-Arguments

I’ve been arguing this position for a few years now, and over time I’ve encountered a number of counter arguments that I’d like to address.

“Output randomness is just input randomness for the next turn.” – Game designer and blogger DanC of the Lost Garden has said this to me numerous times in response to my positions. Basically he’s arguing that there is no actual difference between output randomness and input randomness.

This position has two major flaws. One is that it seems unaware of the possibility of a larger strategic picture that could be providing tons of complexity effectiveness that otherwise you’re losing out on.

The other major flaw is that even if it’s actually input randomness for the next turn, that’s what I call “unfair input randomness”. It’s up so close in your face that you don’t have time to respond to it. You now have a significantly different game state than you did a second ago, and there’s no discernible reason for it. On some games, you might play optimally, but get put into this position and lose anyway. On other games, you don’t get put into that position because the dice rolls go your way. Input randomness, when put up close enough to the player so that he can’t plan around it, is basically output randomness. Feedback is being artificially delayed.

Ironically, I agree with Dan’s sentiment that there’s no significant difference between output randomness and input-randomness-for-the-next-turn, although I think they’re equally bad.

To really drive the point home, imagine a scenario where you have a character who has a “to-hit” dice roll against a tough monster. He swings, and he misses! Well, that’s ok, it’s just input randomness for the next turn, after all! He tries to attack again, and misses again! At this point, you may already have lost, and it wasn’t because of any decision you made.

 

“Some games need output randomness to work.”

If you were to just rip the dice rolls out of Risk, it definitely wouldn’t work.

This simply means that they are shallow games. It’s understandable, because creating a coherent system that is deep is very, very hard to do. However, this is not a defense of randomness; more an indication of a weak design.

 

“If there’s randomness, then it’s all about risk management.”

A favorite of poker players. The idea behind this argument is that having random elements adds a “factoring in your odds” element to the game. You have to weigh the odds of outcome A happening against the odds of outcome B against the benefit of outcome A and the benefit of outcome B, and that makes games more interesting. Essentially, it’s combining odds and valuation.

This kind of risk management is not unique to random games. In any game that you haven’t solved, really every move you make is to some degree a risk that you must manage. In chess, there could be two major strategies – strategy A and strategy B. You might figure that A is more likely to work than B, but B has a bigger payoff than A, for instance. Randomness isn’t necessary.

As to the “calculating odds” aspect of this, determining odds is never interesting, especially not when you’re talking about something like counting cards in poker. Calculating odds in a deterministic system might be harder to do, but it would certainly be far more interesting due to all of the variables at play in a good, dynamic strategy game.

 

“Randomness doesn’t matter – just do the best you can!”

The argument goes something like, “if you care about randomness, you care too much about winning. Just have fun!”

This argument is not actually a defense of randomness in strategy games; rather, it is a defense of randomness in toys. Strategy games have a win/loss condition. If you are telling us to ignore that in FTL, then you are saying that FTL is a toy and that’s why randomness is OK.

 

“Players with a wider skill range can compete against each other.”

If a grandmaster and a newbie play chess against each other, the result won’t be interesting or fulfilling for either party. That much is true! This argument suggests that the answer to that is to throw in some randomness.

Of course, that’s throwing the baby out with the bathwater. You’ve now severely damaged your game for the sake of presenting people with the illusion of more-similar skill levels. The real answer to this problem is good matchmaking.

 

“Randomness makes a game more like real life.”

To quickly counter this argument, let’s simply assume that there is a set of values for strategy games which we can separate from the set of values for a simulator.

 

“Games with randomness still have skill to them!”

True, and I haven’t argued otherwise. The issue is that on a practical level, you will be able to actually explore less of that space in your lifetime, since so many of the games are essentially wasted on false random outcomes.

 

Other Feedback Distortions

I should also note a few types of output randomness that are not usually regarded as such, but function so similarly that they have many or all of the same pitfalls.

 

Simultaneous Action – Trying to guess what the opponent will do in RPS, for example, is effectively random. In fact, that’s why we use it to decide who has to go take out the trash – we consider it fair, because it’s random. The whole reason people agree to use RPS as the determining factor for who will take out the trash is because they know that there is nothing that they or their opponent can do to increase their chances. (Sure, there’s some study that says people are slightly more likely to play rock. But did your opponent read that study, or not? You’re now back to square one.)

Execution – Execution in games is a matter of “can”, not “should”. Can you press this sequence of buttons before my jump kick hits you in the face? Execution is still slightly better than randomness probably, due to the fact that you can at least get better at it. However, inside of a single match, it’s basically the exact same thing. The complex chemicals, nerves, muscles and tissues that stand between “what you wanted to do” and “whether your body actually makes the desired input” have tons of room for error. When you choose to make the input for your Dragon Punch, will it actually work? It’s effectively random.

 

Conclusion

Our collective perspective on randomness in game design really hasn’t budged much in 4,000 years. It’s time that we really gave this question some serious thought.

I’m not arguing that there is no place for any kind of randomness in game design. In fact, I argue strongly in favor of well-balanced, low-variance input randomness in multiplayer games. And single player games require input randomness.

However, output randomness in all its forms is to be avoided. The only time you should use randomness of that kind is if you’re making a gambling machine, or if you’re insecure about the depth of your system.

 

  • Channing Jones

    I pretty much agree with this except that I think that a small amount of randomness for a strategy game is ok.

    The reason is that it gives it a bit of connection to real life. In real life you never have situations like chess where everything happens deterministically.

  • Rickard Elimää

    Good article, and understanding the difference between ingoing randomness and outgoing randomness is important. I kind of missed that in “Uncertainty in Games” by Greg Costikyan.

  • CraigStern

    This is a good article, and I almost agree with it in its entirety. However, I do find this one small part objectionable:

    “Output randomness does not increase the depth of a game. How could it? There is nothing to explore in a dice-roll. We all know that the odds are 1/6 for any face coming up. There is literally nothing else to know or explore.”

    Obviously, the range of numbers the die might happen to roll is not in and of itself tactically interesting. What is interesting is the possibility that your action could fail. Risk management in a purely deterministic system entails weighing the risk of each available move vis-a-vis your opponent’s potential responses; if your actions can fail, however, you effectively have to explore multiple possible outcomes to each action, each with its own attendant range of responses by your opponent. On a turn-to-turn scale, it enlarges the possibility space considerably.

    I think this is basically a rephrasing of your argument with DanC, however, in that randomized results increase the possibility space on any given turn by serving up unpredictability going forward into the next turn. And I do, by-and-large, agree with your objections to doing this on the grounds of fairness–at least, within the context of a traditional system where all abilities rely on randomized results.

    But consider: what about a game where the player is in control of whether to either make moves relying on randomized results, or moves which are perfectly deterministic? (See e.g. Julian Gollop’s Chaos, or Telepath Tactics.) In a game like that, the deterministic move is the safe move, and the randomized-results-dependent move is an optional gambit. There, if the player chooses the non-deterministic move and it fails, I think it’s reaching a bit to call it unfair–the player had the option of a deterministic move, but freely chose non-determinism in the hope of an added reward. The player does not need an opportunity to react to a failed gambit in order to render the results fair.

    As long as everything is built atop a deterministic skeleton, and non-deterministic moves are limited enough that the player has a full range of tactical options without being forced to resort to non-deterministic abilities, I’d argue that many of the disadvantages of non-determinism become fairly negligible.

    (I discuss this stuff in greater detail here, if you’re curious: http://sinisterdesign.net/unpredictability-and-control-in-turn-based-combat-an-examination/)

  • Evan Jensen

    This is an excellent article. However, I do think some elaboration is necessary on a few points.

    The first is that your diagram regarding the player’s move, outcome, and acquisition of skill/learning is simplified in a way that perhaps obscures part of the significance of what you’re trying to explain here.

    Players engaged in a strategy game should (almost universally) adopt a particular strategy in a particular match. Even those who do not do this consciously are still engaged in this process of following a strategy, but often composed of a gestalt of their whims rather than a focused set of choices intended to effect particular outcomes in the game.

    It is the feedback of the strategy in the context of the outcome of the game that provides the data that allows the player to modify their strategy or generate a new, different strategy in a subsequent match. A strategy that just fails utterly will be discarded as just a bad strategy. A strategy that had some success and some failure might deserve additional testing or modification in another match. And one which decisively wins is worth testing again to see if it performs well against different opposition, or under different circumstances.

    As a result, the feedback process is not actually about a particular move. But instead relates to a specific STRATEGY. And by strategy I mean the game theory definition where a strategy is a set of instructions/plans/principles which completely describes the player’s actions in the game.

    The reason this distinction is significant is that randomness drastically multiplies the number of repetitions that must be performed before the player has enough data to effectively evaluate whether a strategy is good or bad. For a particular move, even in a context with randomness, it is generally possible to calculate expected value in advance. But this isn’t really a strategic thought process so much as a mathematical calculation and value-maximization. Retroactively evaluating whether that move was good or bad is only really significant regarding whether it contributed to your larger strategy, or detracted from it. The same move might be advantageous for one strategy and disadvantageous for another, regardless of whether it was beneficial or detrimental in the context of a particular match.

    This problem of learning about the game, and whether a particular choice was good or not, is much worse and far more pernicious for a complete strategy than for a particular move. In the context of an individual choice, or an individual move, that choice can be evaluated based on its immediate consequences or payoff, including randomness and probabilities. For a single move the calculus, including randomness, can be feasible. But a strategy must be tried many orders of magnitude more times before the strategy can be effectively evaluated as being good or bad.

    And, because strategies can only be compared to other strategies, there is a strong tendency for “unstable” strategies to actually proliferate in a game that is less deterministic. Unstable strategies are (intentional) strategies that rely on chance, rather than trying to maximize probabilities and utility. For example, a strategy that guarantees a win about 20% of the time, and pretty much ensures a loss 80% of the time. Unstable strategies absolutely do not work in deterministic games. However, as games become less deterministic, unstable strategies begin to become more and more viable, and even appealing to more players.

    The reason why this is a problem is that highly volatile games are inherently unstrategic. In fact, you could even argue they are inherently low-skill games. But, winning in such games still feels good (at least to many players). Which creates a perverse incentive for the game designer to exploit the “slot machine” effect in a pseudo-skilled environment which is actually quite unstable. Instability makes it more likely that new players, or bad players, will win. Historically speaking, most of the players in strategy games are really not very good at those games. But losing horribly every match is not fun, and threatens to push players away to play something else.

    This encourages developers to “fudge” their deterministic, planning, real-strategy gameplay in order to increase their broad appeal. The result is a feedback loop where the player wins and attributes their win to their choices and their input. This reinforces their selected strategy, since they have confirmed it works. Even if the result was largely determined based on chance, it obviously succeeded, and the logical move is to try it again in a different match to try a different context. Unstable strategies therefore have an enduring presence, even in a supposedly “strategy” game.

    Like Call of Duty did for shooters, RTS games are trying to “trick” its playerbase into the completely baseless belief that they are skilled. However because of the difference in genre, the mechanism is different. FPS games did it with killstreaks to cause snowballing after highly random initial encounters. RTS games do it by injecting randomness into the results of player strategies, making it take much more time for strong strategies to be conclusively identified, and to ensure that even bad players with weak strategies will (at least sometimes) win.

  • Jason Hinchliffe

    I’m sorry but this blog was deeply questionable. That hurts to say, because I wanted to like it, and it makes some interesting arguments, but you seem to fail to understand your critics arguments. You also make some assumptions which I simply find arrogant, such as calculating odds being a non-interesting activity. Randomness as you define it is silly. A result a person can never know is not interesting I agree, but that’s also almost never the case. Your failure to look at probability curves and ranges, and consider contingencies and convexity in regards to payouts in imperfectly predicable situations demonstrates your own tenuous understanding of the subject. In fact, the fact that you think dice are random just shows you have a dubious understanding what true randomness is.

    Let me explain. A pair of die are not random. They can only produce numbers between 2 and 12. I will never roll a 13. I can also know how likely each number is to be rolled, and therefore although any individual outcome is not perfectly predictable, I can make long term plans around this output as I can know perfectly what the range of POSSIBLE outcomes is. Any Settlers player will tell you that although the 8 isn’t a guarantee, you’re wise to get a piece on it. It will PROBABLY help.

    Frankly, I don’t think you have a problem with randomness so much as you have a problem with it being used lazily in game design to obfuscate results and avoid boring determinism without increasing complexity.

    PS. Understanding systems is fun, but your evolutionary argument is bunk. The evolutionary pressure is actually that people are hardwired to work under uncertainty, and that’s why we enjoy risk and randomness so much. It’s a natural part of who we are.

  • A pair of die *are* random. What you mean to say is that they have an unequal distribution. That doesn’t mean they’re not random. Sure, you can know that 7 is your best bet. Is that interesting to you?

  • It’s a good point, overall. At the “atomic” level, I think my argument gets it right, but I’ll make the quick point that a strategy is basically a larger arc that gets its own feedback, like a move does.

  • Numerous people made this “it’s like real life and therefore it’s OK” type of argument. There are other counter arguments, but for now, let’s assume that there is some value to strategy games that can be separated from the value of simulators.

  • Jason Hinchliffe

    Is the game I’m playing just me betting on the outcome of a die roll? If so, then no, and please find me a better game to play. However if it’s an actual system where this input can affect numerous possibilities within the game, and I can plan and mitigate for undesirable results, then yes it very much can be.

  • So the reason it’s interseting is *not* the randomness then. Thus the question becomes: Does the randomness make it more interesting (and if so, how exactly)? Or does it only make it *seem* more interesting at first sight?

  • Jako

    Adding randomness perhaps makes a game less strategic, but not necessarily less enjoyable.

    A game designer should take into account skill level variation among players, and make is so that even less skilled players, and even losing players, get something out of the game.
    Randomness is one way to achieve that. It gives losing players a chance to have small wins along the game, at the expense of players playing more wisely.

    It might appear as “unfair” to the smart player, but actually it’s everyone’s interest that all players enjoy the game.
    The amount of “smart players” at any given game is limited, and a smart player needs another player for the next round.

    Small, random wins are one way to ensure a steady inflow of players to the game.

  • “If you were to just rip the dice rolls out of Risk, it definitely wouldn’t work.”

    Uh, Little Wars.

  • Daniel Slawson

    Great article, thanks for writing it!

    I differ a bit in that I think well-balanced, low-variance *output* randomness can be appropriate in both single and multiplayer games (with the understanding that a deterministic system is clearly best for some games).

    I completely agree that perfect play should never result in a loss because of the interference of randomness. In fact, I think that is a good, ironclad rule for game design generally. Too many designers rely on high amounts of randomness.

    That being said, I think it is possible to include small amounts of randomness/variance in ways that still allow for perfect play to result in a win. This is not trivial to achieve, but possible when the result of a random roll does not push perfect play into a loss state, but instead changes what constitutes perfect play for a given decision.

    Let me illustrate using an example: say a game features a duel between two players, each controlling a character with various skills. Let’s say there’s a skill that deals between 7 and 9 damage, rather than a flat 8 every time. For sake of argument, let’s assume player 1 gets lucky and rolls 9 on that skill every time, and player 2 is unlucky and always rolls 7.

    If player 2 has deterministic choices she can make that offsets her lower damage, she could still win with perfect play (even with imperfect play). Her decreased damage output might mean that she can’t win in an all-out offensive rush against player 1: in this case, that option no longer represents perfect play. In order to achieve perfect play in her circumstances, she might make use of choke points, high ground, deterministic abilities, etc. that might more than compensate for the damage difference, and still allow her a win.

    In other words, the difference of 2 damage every time that skill is used could be compensated for (if the game’s mechanics allow). If that skill were deterministic and did 8 damage every time, the game would be that much easier to predict and create strategies for. The skill having a very small, calculated amount of variance actually creates more potential game states, without necessarily disrupting perfect play.

  • Oh yeah, I forgot about Little Wars which everyone loves to play and definitely isn’t solved! 😀

    But yeah, they also have different rules, so the point doesn’t really work. Here are the rules to Little Wars if you’re interested. http://gardenwargaming.com/wargame/LW39.html

  • CraigStern

    “Her decreased damage output might mean that she can’t win in an all-out
    offensive rush against player 1: in this case, that option no longer
    represents perfect play.”

    The problem with this example is that Player 2 will not have this knowledge in advance–since we’re talking about damage variance, a 7, 8, or 9 is each equally likely on any given attack. You’re defining “perfect play” here in a way that has the benefit of hindsight, not in a way that a player can be reasonably expected to predict at the time they are actually playing.

  • Daniel Slawson

    Thanks for the response! Yes, you’re totally correct. But predicting the exact roll isn’t necessary, if they are still able to play perfectly and win after the roll has been made.

    If both players have a reasonable amount of hit points, their strategies will change over time if one of them notices their HP is notably lower than their opponent’s because of this roll variance (perfect play for a player in that circumstance would probably be to retreat). The one scenario where this variance is critical is if a player has HP low enough that they could die, but might not depending on the roll. It’s up the players to use the options available to them to avoid those scenarios if they aren’t in their favor, and take the choices that would still allow them to win despite the small roll variance.

    If the skill has no variance, it’s easier for the player who moves second to plan to retreat after X rounds, since it’s easier to predict in advance the damage she can expect to take. But with slight variance, this prediction is harder to make: she may have to retreat after 4 rounds or 6, for example. More potential game states that make prediction more difficult.

  • To be fair Little Wars does have flung projectiles at the pieces which is non-deterministic, and measuring with string can be “fuzzy”, but the point is Risk without the dice rolls would not definitely not work, double negative intended. Imagine a deterministic game of Risk where taking a territory with X units will require attacking it with Y, where the relation between X and Y is known. Imagine getting rid of the cards in favor of a fixed resource collection making rewards predictable. I think the game would need a little massaging, but I don’t think it wouldn’t work. In fact it might elevate the game to a strategy level similar to Go.

  • quakerinabasement

    After reading this article, I thought of Strat-O-Matic Baseball. (On the odd chance that anyone is not familiar, the game uses actual baseball player statistics and dice to determine the outcome of a pitcher-batter matchup.) How does a game like this fit into your thinking?

  • Avi

    I agree with the vast majority of this and entirely with the conclusion. I appreciated the reference to FTL, and I think there’s an interesting discussion to be had on whether the procedurally designed perma-death mechanic in rogue-likes results in overall lazy game designs. Similar pointless randomness happens throughout the genre.

    My other comment concerns your counter-argument to randomness makes it like real life. This is something I’ve thought a great deal about in games, and I actually think your argument in this instance is weaker than it could be. The much more concise argument is that real-life randomness can be responded to through randomized improvisation. Take the following examples:

    A fight in your apartment breaks out. As you flee from your much larger assailant, you fall over an end table. From the floor, you spot the metal ashtray that had fallen under your couch. You pick it up and bash it into the assailant’s face as he comes around the couch. In the game version of this, the random event of the “crit fail” (falling over the end table) might happen. However there will not be sufficient breadth of sporadic interactions to adequately respond. If your sword is knocked from your hand, you cannot throw sand into the eyes of your opponent as you run to pick it up.

    The second example concerns a larger scale strategy. You’re a military commander at a small remote outpost consisting of twenty soldiers. You’ve gotten intel on the enemy setting up camp to the East, presumably to attack in the next 24 hours. As you’re developing a suitable strategy, a large storm occurs. This requires redirecting your resources to locking down equipment and keeping your troops out of range of lightning strikes in the open area. However you also know that the enemy camp is down a severe slope if they’ve placed it on the Eastern side. Their camp is being flooded with water and dirt right now. You send two groups of four soldiers out to flank their camp while they’re distracted by the storm. The enemy tents were being washed away when the soldiers arrived, and only one of your soldiers comes back wounded. They sent the enemy scattering, killing most of them. The game will simulate the storm, and you’ll take percentage hits to crucial variables. There will be no or limited simulation of enemy distraction, and the ability to turn the situation to your advantage will be negligible.

    The crucial element of unforeseen circumstances in real life is that they create new opportunities. This is inherently impossible in games. If the circumstances were to create new opportunities, those opportunities would have to be programmed specifically for that set of circumstances. That would mean the circumstances could no longer be random. Random circumstances will result in limited opportunities for player response at best, if at all.

  • Rob Seater

    Ho do you feel about games with hidden information but no random elements? For example:
    – Dungeon Twister (feel very skill based)
    – Battleship / Salvo (feels very luck based)
    – any deterministic game with simultaneous action selection

    They are technically deterministic, but their optimal strategies are mixed strategies requiring the players to inject randomness into the system in order to play well. Second guessing your opponent and identifying the correct mixed strategy are both strategic, but in a very real sense the players have turned a deterministic game into a random one.

  • In general, I’m a big fan of hidden information, BUT it has to be a few “steps out” from “activation”. In other words, I’m happy with how the new plants come into play in Power Grid, but I’m not AT ALL happy with combat in the card game Yomi (or Rock Paper Scissors, if you prefer).

    And again – your “technically deterministic” – I sort of talked about this in the beginning of my article. You know what else is technically deterministic? A dice roll! So forget about that. The question is whether a player could or is even supposed to be able to figure out, to somewhat solve, the problem or not. In the case of Battleship guesses and Yomi combats, the answer is “no, never”. It’s basically as random as a dice roll.

  • Masoud

    Thank you for this great article!. I just have a short point: all the argument of the article is based on the premise that “…we find this process enriching and entertaining. This is the “essential fun” of strategy games”. If this is the case, then I agree with everything else. However, I am not sure about this premise, or better should I say, I am not sure how much “essential” is this part of the game for making it fun. I think that there are other imprtant parameters in a game for making it fun, and a game is more “fun”, if the designer can maximize the weighted average of these parameters for the intended group of the game. One of such other parameters in my opinion can be feeling the tenseness of waiting for the result of an unpredictable action. shortening the amount of time between the action and the result increases the tenseness, and unpredictability is necessary for it too. This can be achieved in very good way with random outputs. Again, for including such a mechanic we should know the audiance of the game, and know how much different parameters of making a game fun are weighted for them.

  • Anonymous

    I see your point, but you’ve made several mistakes.
    1. Odds are not the same as probability, but that’s less of a big deal.
    2. You assume that RPS is equivalent to a RNG but that plain isn’t true. First of all, have you heard of a Prisoner’s Dilemma? RPS is a bit simpler, admittedly, but some of the ideas there still apply, and you did say that simultaneous picking=RNG.
    3. Your conclusion totally destroys your argument. You admit something is true which you argued made for a stupid game above.
    4. Most of your counterarguments are against Straw Men. That is, you oversimplify many of the arguments in favor of having RNGs.

  • Secret Library

    another case of weird output randomness is street fighter 2 controls, which many people (mistakenly) assume to be deterministic. there is a dice roll that determines the amount of frames you have to complete a dragon punch (8-15, according to david sirlin):

    http://www.sirlin.net/sf-hdr/street-fighter-hd-remix-design-overview

  • Hi Keith!

    Great blog post! I just stumbled upon your article after I just finished my blog post about random password generators for non-techies. If I didn’t already finish my article, I would’ve totally included some of your info in my post! It’s going to be published tomorrow, but I’d love to hear your thoughts! Here’s a link: http://blog.dashlane.com/how-random-password-generators-work/ It won’t be available until 8/9/16 at 8am EST. I look forward to your feedback! 🙂 -Malaika

  • Spam Keith Burgun | Lead Designer at Dinofarm Games
    Author of Game Design Theory

  • DukeZhou

    There is clearly a place for randomness is strategy games, with the caveat that it must be applied properly and the game it is applied to must “have legs”. Monopoly, Risk, and more recently, Catan, validate this position.

    That said, it was a loss in Risk in middle school, brought on by an extended series of terrible rolls in a single, decisive battle over Kamchatka, which led me to believe Risk was “broken” in the sense of this excellent article. While I could rationalize the event–the Spanish Armada sunk after all–the experience set me on a trajectory to address the problem of randomness in strategy games, and thus I was delighted to find this manifesto, which I wholeheartedly support. [Again the caveat that I almost always won at Risk–had I won every time it’s unlikely my pals would have wanted to replay as much as they did. The flaw of randomness provided a utility function in that play itself was the point; victory was merely a fringe benefit.]

    Experience leads me to believe that the underlying problem may actually reside with the sheer difficulty of crafting purely deterministic games.

    Part of this difficulty may derive from the astonishing rarity of fundamental game mechanics, particularly mechanics that may be understood as combinatorial, in that what we’re really talking about form the standpoint of wide adoption are combinatorial “playgames” such as Chess, Checkers, Tic-tac-toe and Go–deterministic games are wholly logical and mathematical, thus their fundamental forms are always abstract, and the abstract forms of deterministic games seem to be the most widely played.

    When we include extensions of these Ur games, Othello/Reversi, Connect Four, Pente/Gomoku, etc., it’s still a very short list. [Note: I speak from a design as opposed to a mathematical perspective in categorizing the second group of games as extensions. Go and Reversi are mathematically distinct, but it’s difficult to see Reversi as something other than a product of Go.]

    Essentially we have a small group of core mechanics–placement, movement, connection and takeaway, with the latter two emphasized in the sense that they allow the victory conditions. To wit, we don’t categorize as “movement” or “placement” games, but rather as “connection” or “capture/takeaway” games. These mechanics can be combined and expressed in different forms–Go involves capture by surrounding, Checkers involves capture by jumping, Chess involves capture via intersecting; Pente is a combination of connection and capture, in this case by flanking, which is itself a variant of surrounding. I can think of other methods of applying takeaway in partisan contexts, or methods of applying connection in dimensions that exceed the geometrical dimensions of the gameboard, but I’m hard pressed to think of such examples in widely adopted games of pure skill. [Note: “bump” and “push” do seem quite promising, per the critical reception of recent games such as Auro and Arimaa.]

    This may indicate a “shibui” element in that deterministic games that are widely adopted must be simple enough to engage players, complex enough to defy solution by the players, and, most importantly, compelling enough in the sense of emergent complexity to captivate the mind. This constitutes, essentially, the application of aesthetics to algorithms.

    Great art is hard enough to produce. Great art that is mechanical and mathematically non-trivial seems to be much more difficult: we have museums filled with great sculptures and paintings, libraries of great music and literature, but only a handful of deterministic games that may be said to be widely played.

    Returning to personal experience, the first obstacle I encountered when I formally began to design such a game, is that the mathematics are absolutely unforgiving, and a profoundly more challenging problem the development of new mechanics and aesthetic component.

    For me, arriving at a small set of novel, fundamental game mechanics that did not always result in a win for one player or a draw was a project not of months or years, but a decade. Even then, the mechanics still required years of testing to validate, with the full awareness that ultimate validation is measured in centuries and millennia.

    From a purely economic perspective, it’s less risky to “cheat” by utilizing randomness in the design of strategy games because it profoundly reduces the difficulty of the process.

    That said, I suspect that Chess, Go and Tic-tac-toe will continue to be played so long as humans persist as a species, thus I think your own endeavors in utilizing more complex strategy game mechanics, typically associated with random gameplay elements, in purely deterministic contexts is not only worthwhile, but highly commendable.