What about Chess is Random?
How strange quirks in our decision making adds some random chance to every game.
The last Medium piece I wrote was about Card Games (following the traditional ‘CCG’ format) and how too much randomness within player’s decks could effectively curtail the strategic value of the games. One strange result of the article and the discussions that arose was that a few people, for whatever reason, seemed to believe that I wanted to remove all randomness from CCG-style games (which was odd considering I believe I specifically mentioned that some randomness is often necessary) — and there were more than a few people who jokingly recommended Chess to me, as it is a non-random game.
I didn’t take any of this personally — because I think Chess is a pretty great game that I’ve neglected, and it gave me a chance to mention a theory that I have, and that other people have commonly told me ‘is crazy’, until I get down to it… and that is, ‘there is some randomness in Chess’.
Obviously there are no random systems within Chess’ gameplay. No decks of cards, no dice, no random number generators. But we do use our brains to play it — and our brains may well have some pseudo-random properties. It is in the irrationality of our brains that Chess may experience some somewhat randomly determined situations.
How about an example?
Imagine that you have two equally proficient Chess players (as much as two players can be equally proficient) named Jessica and David. They are playing via an online Chess program that renders them anonymous, and they have never played each other before. As Jessica and David start playing, they of course react to the other’s moves — but their long term strategies are not clear to the other until certain milestones are met. Let’s say that Jessica, having no information about David aside from his moves, chose to start pursuing strategy A, or, as we’ll call it “The Rotating Cuckoo Clock of Doom” (can you tell I don’t play Chess yet?). David similarly can only use Jessica’s moves as an indicator of what she’s planning, and has pursued a strategy aptly named “The Grasshopper Likes Listening to Kanye West” for his end game strategy.
Now, how did they determine their strategies? An aspect of the choice may be a an arbitrarily made decision, based on their own personal preference, or based off a ballpark guess of what the opponent’s strategy is, considering their opening moves. All of these could be considered relatively arbitrary. Someone flipping a mental coin could be considered random enough, personal preference is information that is external to the current game, and their opponent cannot account for, and basing a strategy off early moves is always going to be fairly unreliable (if it wasn’t — the strategy would be easily counterable due to its ability to be easily detected and dealt with, and therefore not very good against experienced players, right?). Whether David or Jessica can correctly guess their opponent’s long term strategy from early game moves is likely to be fairly random. If they find the needle in the haystack, nice! But if they don’t — well bad luck, they may need more time to counter their opponent.
But let’s say that, unfortunately for David, Jessica’s “Clock of Doom” actually fares very well against his own “Hip Hop Hopper” (I’m so sorry) strategy. What if, by the time he’s got strong wind of where the game’s headed, he’s already traded a piece that he may desire to counter Jessica’s strategy. While he may have traded it for something worthwhile (that he ascertained at the time of trade — he didn’t necessarily blunder), there are now permanent effects on the rest of the game. Now, had either players ‘felt’ like pursuing a different strategy from the onset, or arbitrarily reacted differently to the early game — the game would be in a completely different state.
David may still win the game — his early game disadvantage may not necessarily cause him to lose the game, hell, Jessica may underestimate him as a result of her early advantage and he may turn it right around… but it doesn’t change the fact that pseudo-randomness has affected the game state in a game oft considered non-random, and it has lead to one player having an advantage over the other at this point in time. This is not necessarily a bad thing, but it is a true thing.
Note that I’m not saying that this makes the winner of a chess match arbitrary, or random. As much as any game can be, Chess is decided based on skills, like applying tactics and forward thinking. But there is certainly a light sprinkling of randomness on top of every game, which makes each game unique and engaging.
Is this really random though?
Now, some of you may be sticklers for the definition of randomness. Is ‘the way I feel right now’ actually random? How about choices that I make that seem random on the surface level, and may be seated in my subconsciousness? Is my brain’s output pseudo-random? That’s quite a philosophical question concerning how the brain works that we may not be able to answer now. I would argue that within the scope of the game, the small quirks of our brains and how those quirks may lead us to make one choice over another should be considered a ‘random’ variable. Or at least, ‘arbitrary’, if definitions are bugging you.
It would appear that, at least on some base level, all games operate under some basic level of ‘randomness’, even if there are no randomised components (dice, decks of cards) within the game, purely by virtue of being played by beings who can factor many different odd quirks into their decision making. If you believe that humans can make a decision within a game that is not entirely based on the game state (e.g. a hunch where they have incomplete information) — then any game where a decision like this has been made has had some level of ‘arbitrariness’ added too it.
So how can we measure this — are there games where players act more arbitrarily than others? Absolutely — and this is a huge issue for many game designs.
The Decision Making process.
How do players make decisions?
I would split it into 4 steps:
- Consider all options.
- Remove as many ‘bad’ options as possible.
- Evaluate and rank left over options.
- Pick the ‘best’ option.
Note that this is usually a fairly subconscious process until the last two (and even the third step often has subconscious components to it). We rarely consider every single move possible (in a game of chess there are probably hundreds of possible moves at any time aside from the first and last few turns), and we often can recognise moves that won’t achieve anything and dismiss them very quickly.
The third step is often where people get bogged down in thought, and as far as evaluation goes, the metric depends on not only the game, but the player. If I like playing aggressively, I will probably evaluate the “goodness” of aggressive moves higher than defensive moves. Vice versa if I like playing defensively. Once the moves have been given a subconscious ranking (or conscious, if you’re really having trouble tossing them up), players will often pick the best one… unless they arbitrarily (or is it?) pick the runner up in an attempt to throw off their opponent.
Picking the best move or its runner up may be arbitrary or not. If the player feels they understand their opponent, they will likely be very certain in picking one or the other, but for players who are uncertain of how well they’ve gotten in their opponents’ heads, it could be very arbitrary. Especially if a player is having a lot of difficulty deciding (and therefore slowing down the game with long thinking periods — called ‘analysis paralysis’). maybe it’s actually better to decide which move to use with a coin toss, it’s quicker and may have the same effectiveness. At that point the outcome of the game would be indisputably affected by randomness.
Get to the game design!
The interesting thing is — sometimes when you get feedback for a game, players will say it’s ‘too random’, and I’ve had this feedback from games with no random elements, like Chess. And, sometimes that’s the player’s way of saying, ‘decision making was too arbitrary’. An example is here on reddit. This is also exactly what I faulted Yomi for, comparing it to Rock, Paper, Scissors in my Yomi vs. BattleCON piece. Now, if they lost the entire game based on bad die rolls — that’s likely what they’re referring to, but in games with no random components, or if you look at the game experience and there didn’t seem to be big upset moments with dice, this might be what you’re looking for.
Just as randomness removes your agency by introducing chances of failure, or limiting what options players have, players can feel that they have no agency when they simply do not have any reason to take one action over another. Instead of ‘too few’ choices, they have too many.
This can be an issue with any of the steps detailed above. In the ‘consider all options’ step, the player likely doesn’t understand the game well enough to comprehend all the actions that they can take. Unless your game is an extremely complicated mess, this should not be an issue if players understand the game (and if your game is too complicated, I’m sure players will tell you that directly). You may need to give your players aides to help them remember options, or reduce the complexity of the game.
In the second step, where players try to eliminate ‘bad’ actions, if players cannot assign quick and easy evaluations to their options, and there are a number of them (more than say, 4-5?), this can feel ‘too random’. Most games rely on the majority of options being immediately handwaived to ensure players don’t get overloaded. Make sure your players can narrow down to 4-5 or less options quickly. If they are presented with more options than that, make sure there are ‘bad’ choices that they can ignore. Give them information to help them discern their options.
In the third step, players may falter if they cannot assign more detailed evaluations to their options, they may get caught here in analysis paralysis. They may just ‘give up’ on their decision and pick the ‘best’ action they can without really thinking any of them through. This is also where players will start to try to react to other players’ possible future actions, in an attempt to ‘read’ their next move and outsmart them. Now, if you cannot easily evaluate your own options, how on earth will you evaluate your opponent’s options?! You need to give players information to help them discern clear differences between their moves and those moves’ effectiveness . A player should also be able to discern (what they consider to be) the most likely, or the two most likely actions their opponent will take.
This is where it gets really hard. There is no easy format for ‘making a set of strategic options where the player can anticipate their opponent’s actions and outsmart them’. It depends on your game. One thing to note though, is, it can be too easy to decide on a move to use. If a player’s best move is always obvious to them, then there is really no choice to be made. It’s like Tic-Tac-Toe where your best move is obvious, and so every game is relatively deterministic.
So you don’t want one move to stand out too far — or it’s a no-brainer, but you also don’t want them to be too similar, otherwise it’s a coin toss. Ideally in 5 moves, a couple stick out as the best, but the others also have practical applications depending on what the opponent does, or could do. Think of traits like ‘benefit gained’, and ‘reliability’. You may get the best benefit from Action #5, but it may require your opponent to make a certain move to work. Action #3 may benefit the player no matter what, but its benefits might be the smallest, making the more powerful options more appealing. Players can then choose and try to outplay their opponents.
Now I hope all of the tips have shown a trend. Giving your players more useful information aids decision making. Make their options clear. As always though, don’t overload them with useless information, in the hopes that it will be useful.
- All games operate at some base randomness, because humans are not deterministic, and will occasionally make semi-arbitrary decisions.
- Some games will operate at higher levels of base randomness than others, and that is often caused by higher player feelings that the decision making is arbitrary or, ‘random’.
- Give your players enough information to make decisions, but not so much that it overloads them, or your game becomes deterministic as one action always trumps the others.
- Always try to ensure player options are clear in how they differ, and ensure that players can subconsciously eliminate the bulk of their possible actions, to reduce the options they are weighing to a manageable number.
- Actions will often differ in the advantage they give, and in their reliability when considering the opponent’s possible future actions. These are two very basic traits that can be contrasted to make multiple options appealing.