I thought I’d attempt something a little different this week, and write about a subject that often crops up at the games table, randomness and luck. I have put together my own theories about these two things, come up with my own definitions, and given some examples; but whatever I think, as long as I get you thinking about them, then I’ve achieved my purpose and will be happy with that.
Randomness and luck are terms often associated with tabletop games, but what do the actually mean?
Randomness – ‘The quality or state of lacking a pattern or principle of organization; unpredictability.’
Luck – ‘Success or failure apparently brought by chance rather than through one’s own actions.’ Or, ‘Chance considered as a force that causes good or bad things to happen.’
That’s the Oxford Living Dictionary definitions; let’s examine things a little closer.
As it’s a solid concept, let’s take a look at randomness first. Randomness in games is a physically created ‘thing’, unlike luck, which is illusory – the same random result can be good luck for one person but bad luck for another. Randomness is the introduction of specified results in an unpredictable manner – For example a die contains a known range, but one cannot predict the number that will be rolled. Likewise, consider a normal deck of playing cards – we know its contents, but once shuffled we cannot predict the order in which they will be drawn.
The majority of games use randomness to some extent – Gloomhaven has its attack modifier cards, Pandemic has its player and infection decks, and innumerable games use dice; all great ways to introduce a degree of randomness into the game, but why?
Imagine playing Pandemic, but instead of shuffling the infection deck at the start, put it in alphabetical order. Whenever an epidemic occurs, once again place the discard back on top of the infection deck in alphabetical order, A at the top. You have now taken away the random draw of infections, the players know exactly what will be coming out, and when – it makes the game predictable.
Playing Pandemic in this manner results in an easy game that is boring to play – the tension and excitement that the randomness brings to the game has been removed. It becomes a simple puzzle, with very little replayability.
The degree of randomness introduced into a game is relatively easy to control, after all, it’s just probability. Consider dice: In a Role-playing game you may roll a 6 sided die (D6) to see if you hit that big bad troll bearing down on you. A 1 is a fumble; 2 or 3 a miss; 4 or 5 a hit; 6 a critical hit. So, the probability of hitting the troll is 50%, and in every 6 rolls to hit the likelihood is you will do 1 critical hit and 1 fumble – on the whole that’s not great for the game, or the player. Who wants to keep fumbling every 6th roll, or wiping out each monster you meet with a quick critical or two?
Now, change the die to a D20; 1-fumble; 2-10 miss; 11-19 hit; 20 critical. You still have a 50% chance to hit, but the odds of either a fumble or a critical have been dramatically reduced – from roughly 33% down to 10%.
Exactly the same thing can be done using a deck of cards. Let’s say a deck has 20 cards, and within this there is one that causes a monster to appear, so there’s a 5% chance you’ll draw that card. If it contained two monster cards, then that has doubled the chance to 10% – easy isn’t it!
Unlike a die though, the chance of drawing the monster card changes dependant upon whether the drawn card is returned to the deck or a discard pile. Let’s take our deck of 20 cards containing two monsters. First draw you have a 2 in 20, or 10% chance of drawing a monster. If this card then gets returned to the pack then our chances of drawing a monster card next are the same as before – or are they? Well yes, if the pack is shuffled; If the returned card is just placed on the bottom though, that’s a different matter, and it’s exactly the same as placing it in a discard pile, at least until it comes back round to the top!
As it’s very unusual in games to just place the card on the bottom of the draw deck (it does occur in some games, but I’m not going to go into it here), let’s use a discard pile. So, the first card we draw is not a monster, and we place the card in the discard. We now have a greater chance of drawing a monster card – a 2 in 19 chance, or approx. 10.5%. If we’ve drawn 10 cards without seeing a monster we are now at a 2 in 10, or 20% chance of drawing one next.
This is all very good maths, but what does it mean for the players?
First off, if the players are aware of the deck’s contents, then they can plan their strategies based upon the chance of drawing a certain card, or not. As the deck grows smaller, then the risk increases. Likewise, if both monster cards are drawn straight away, then the players know they are safe until the deck runs out.
Secondly, it adds tension to the game, especially in a game where other constraints and pressures are being applied to the players over time. Going back to our monster deck again – The players have being investigating the dungeon for 12 turns now – they are yet to discover a monster, but have triggered several traps and other encounters, which has left them battered and bruised. They have the end in sight, just another two turns, but they know there is a good chance of discovering a monster (2 in 8, or 25%), so the tension is increased as they move towards their goal.
Knowing the deck’s contents is a big advantage to a player with the ability to count and remember cards, so what difference does it make if you play with an unknown deck of cards?
In terms of what and when things will happen – absolutely nothing, they are random events whether the deck is known or not. Where it makes the difference is in being able to strategise based upon the chance of something happening, and with an unknown deck this becomes impossible.
Playing a game for the first time I never look through a deck of cards, I like to be totally in the dark on what to expect; every time a card gets drawn it becomes an exciting event, a leap into the unknown. Of course this leads to a slower and more cautious style of play, but I find it a great way to learn the intricacies of a game, as well as getting that induced tension right from the start of the game. Unfortunately it’s short lived; after playing a game once or twice, it’s inevitable that you learn what is included within the deck, and even subconsciously you start planning for certain cards turning up.
Obviously, there’s a lot more to including randomness in a game, and creating an exciting experience, than that – and it’s when you start combining random elements that things start to get a little more complicated, and for the designer, I would imagine it becomes a lot more difficult to balance.
The amount of random elements included within the gameplay is an important design issue – Too few can see the game become predictable and unexciting; too many and the game can become a mess of uncontrollable events, which puts the player right off, especially if these events are linked.
Linked randomness is a nightmare! This is where one random event affects another random event – Using Pandemic once again; You draw from the player deck and reveal an epidemic, this then causes you to draw from the bottom of the infection deck and place 3 disease cubes there. Drawing the epidemic was a random event, which then leads to the drawing of another random card, this time from the infection deck, the result of which could be game ending!
This is an example of linked randomness in a well-balanced game – an epidemic only occurs within a set criterion; if you are playing the standard game with 5 epidemics, it works out that you should draw a single epidemic within roughly each batch of 10 cards. So the player has some knowledge of the event. The worst that can happen is 2 epidemics in a row, when one is the last card of one batch and the next card the first. But then the player is safe in the knowledge that another shouldn’t come up in the next 9+ cards.
To make this a bad example of linked randomness, just shuffle the epidemic cards in to the player deck. The player now has no knowledge of when the event will occur, and the game goes out of balance because the revealing of the epidemic cards is out of control. It would be possible to reveal all epidemic cards within the first 5 draws, making the game impossible; likewise, they may all be at the end of the deck, and the game becomes extremely easy. By controlling where they are placed within the decks balances the game, and having them randomised within that limit makes the game exciting.
Once you get in to a greater number of linked random events, control over the game play starts to run away, and it takes a very well designed game to control everything.
Before I move on, here are some of my definitions –
- Uncontrolled Random Event – A random event where the possible result is totally unknown. For example: Out of a deck of 20 cards, 10 are drawn without looking at them, and used for the game. The players have no idea what is included within that deck, and so cannot form a strategy based on the possible outcomes of a drawn card.
- Controlled Random Event – A random event where the possible result can be surmised by probability. For example: Pandemic’s player deck where it is known that, in a standard game, there will be and epidemic card turn up in every 10 cards drawn. The players can then form a strategy taking this into account. A die roll could be considered a controlled random event because the result of a roll can be guessed using probability, and the greater number of dice rolled, the more controlled the event becomes. However, dice rolling can also be considered an uncontrolled event – it depends upon how the decision to roll the dice is made. If it is forced by the game, then it could be considered uncontrolled, as the player can’t plan for the result. But if the player is given a choice whether to roll a die or not, then the decision is made upon the chances of getting a positive result, and is controlled.
- Linked Random Event – A random event is where the possible result causes, or affects, another random event. These maybe controlled or uncontrolled events. For example: A die is rolled and upon a certain result a random card is drawn from a deck of cards.
Next, let’s consider a typical turn during ‘This War of Mine’, a game that places you as a survivor during an ‘unending war’. The game highlights a variety of random events and uses them to produce what could be described as a ‘survivors misery’, which I have no doubt was intentional. This is a hard game to beat, and I think the reasons why will become obvious as we go along.
- Random event number 1. Firstly you draw an event card – thirteen of the included eighteen event cards are used in a game, including the chapter cards. All are pretty negative and widely diverse, meaning it is difficult to plan for, other than the fact you can guarantee that cold will increase, so you need to be thinking about offsetting this.
- Next you take your three actions per character, if they have three actions to take that is. It is at this point you need to try and plan for what may happen in the rest of the turn; taking into account the event card just revealed, your characters weakness’ in spirit, and your ability to feed supply them all with food, water, and any medical needs they require. Some of the actions that can be taken also result in a random event taking place – Random Event Number 2 – Nearly all the events that can be taken as actions are controlled random events, such as die rolls or cards the flipping of cards, which are randomly placed, but should be known by anyone who has previously looked through the deck or played the game before.
The actions taken here, ‘water and hunger’, are not random events, however, having the food and water to give to the characters, often depends upon the outcome of random events.
This sees the players assigning characters to various tasks: Sleeping, guard duty, and scavenging. None of these are random events.
- Random Event Number 3 – The players choose a location to scavenge in. I know what you’re thinking, if the players choose the location then surely it isn’t random? In effect what is happening here is that the players are being given a degree of control over the random events that will occur during the scavenging. It lists the results of rolling the die for special findings; It gives the player an idea of what to expect – closed doors, rubble, basement entrances; It lists the journal entries for exploration and rare findings; The position of the location on the board also sets the number of unknown. So, this would be classified as a linked random event as it directly influences the results of later random events, and it is a controlled event as it gives the players an idea of what to expect so that they may be able to plan ahead.
- Players then choose equipment, taking into account the location they have chosen.
- Noise marker is set.
- The unknown deck is prepared and exploration begins. To prepare the deck 10, 12, or 14 cards are drawn from the exploration deck to form the unknown deck. Random Event number 4 – The cards in the unknown deck are drawn and resolved one at a time. Though the drawing of the cards themselves is an uncontrolled event, as the player has no idea what to expect, there is a certain amount of control over inbred into the system. Exploration can be ended on the turn of most cards, and some cards give the player control over reducing the deck down, or influencing the result of another card. Exploration in itself can lead to a number of other random events occurring, such as die rolls for noise, findings, or journal entries, but due to the number of possibilities I’m not going into them all here. Though there are two worth following through, as they usually occur every turn, and that’s findings, and the result of being too noisy! Random Event number 5 During the resolution of the unknown deck players may be told to draw and resolve a findings card. This is both a controlled, and an uncontrolled, event. Controlled because a die is rolled and the special findings list on the location consulted, the location that was chosen earlier by the players. Uncontrolled because the findings card also contains random items listed upon it, and the player has no idea what these may be until it is drawn. Random Event number 6 – When a card directs the player to roll for noise, a die is rolled and the result compared to where the noise marker is. If the result is less than or equal to the current noise, then a card is drawn from the residents deck. I would consider this die roll to be controlled, as the player can make choices through the various cards drawn from the unknown deck – It’s very much a ‘push your luck’ mechanism. Random Event number7 – Drawing a ‘residents’ card. This is a controlled event, as there are ten cards in the deck to start with (other, more difficult ones are added as the game progresses), as anyone who has previously played the game should have a fair idea of what is contained within the deck, and the majority of them get removed after being drawn.
- Finally, the players choose which items to return to the shelter with, as they can’t carry them all. Though I haven’t highlighted it here, this has elements of a linked random event. Which items to take back is a decision based upon the essential needs of the survivors (food and water if they’re running short), and what they may require dependant upon random events happening later – the drawing of a fate card.
- The players choose weapons for the guard.
- Random Event number 8 – A card is drawn and resolved from the night raids deck. This is a controlled random event in the same vein as drawing a ‘residents’ card above.
- Crime Wave – 2 cards are added to the night raids and residents deck, in any combination. This is adding a degree of control to those random events when a card is drawn.
- The party returns, bringing with them their findings.
- Meds and bandages are assigned to those characters in need.
- Random Event number 9 – Draw and resolve a Fate card. This is an uncontrolled event, as the drawn card is shuffled back into the deck once resolved. It is also a linked event, both to the random choice of characters during set-up (a random event I didn’t include here), and to the items the characters have in storage as spirit resolution often depends upon items such as coffee or cigarettes being present. These items are usually found by resolving one of the previous random events during the scavenging phase, such as a findings card.
- Random Event number 10 – 2 ‘Narrative Action’ cards are drawn, read, and one is chosen whilst the other is shuffled back into the deck. This is a difficult one to define as controlled or uncontrolled, as it has elements of both. On one hand, because the unused card is shuffled back into the deck, it could be considered uncontrolled. On the other, because the players get a choice of which one to pick, you could argue it has elements of controlled. Personally I would classify it as uncontrolled as you have no idea what is going to be draw next.
As you can see, during a typical turn you are likely to encounter at least 10 random events, that’s a lot! In my mind it’s far too many, and the game just isn’t balanced.
Imagine play testing – every outcome of every random event needs to be considered in how it affects the game, and when random events are uncontrolled this becomes more difficult to do. Then, take into account the number of linked events, for example one where you may require a certain item, but the random events leading you to gain that item have not occurred. It becomes a game that, rather than placing the win/lose condition upon how well the players played their part, it places it upon how ‘lucky’ they were during the random events. And that brings us nicely around to the next part…
I said earlier that luck is something that is illusory, what did I mean by that?
Luck is something that exists only in our own mind, it has no substance, and the concept of luck is different for everyone.
What one person believes is lucky may be perceived differently by others – Person ‘A’ finds a £10 on the floor, picks it up and thinks, ‘that was lucky’. Person ‘B’ on the other hand, may have picked it up thinking, ‘a £10, what were the chances of that?’ Unfortunately, person ‘C’ realises he’s dropped a £10 and considers the reason for this happening!
Interestingly, this reveals three different types of person, and each would approach ‘This War of Mine’ differently.
Person ‘A’ would proceed through the game cursing his luck when events take a turn for the worst, and then riding that ‘lucky streak’ when things went well. They will be up and down throughout the game, shrieking with delight when they roll that fourth straight 10. This is the excitable person happy to win or lose, just enjoying the experience, after-all, it’s just down to luck.
Person ‘B’ is the calculating kind of person, and they would be analysing every turn of a card or throw of a die. They would realise that the game is nigh on impossible to predict and beat consistently – it would be put away on a shelf, never to see the light of day again.
Person ‘C’ would play the game several times, losing badly in a variety of ways. They would sit, contemplating the rules; are they playing it correctly, have they missed something obvious? They would then try again, determined to beat the game, until finally they would sell it on E-bay, believing the game is just to advanced for them!
Getting back on track… Playing tabletop games seems to create a much greater dependence upon luck than, let’s say, a video game.
When playing a board game a player advances around a corner only to meet a very large and ugly Orc. The appearance of this great beast is due to a card being drawn from a random deck of encounters. The player considers himself unlucky, especially as he’s just about to be handed his bits on a plate!
Now consider the same thing happening during a video game – the player turns the corner, sees the great hulking thing, and proceeds to hammer on his buttons until he’s re-spawned at his last save point. Being unlucky never enters the players head, it’s just part of the game, and he faces whatever is thrown at him.
Why should the same event provoke a feeling of ‘unluckiness’ in one instance but not the other? I believe it is because the mechanisms used to create the event are visible when playing a boardgame, whilst they are hidden when playing a video game, though the actual programming to create the random monster is probably as simple as drawing a card.
A similar thing occurs playing role-playing games. When a monster turns up in front of the group, they don’t consider themselves unlucky, and that’s because they don’t view the event that caused it to appear. Is the monster a planned encounter, or is it a random wandering monster created on the roll of a die – the players will probably never know.
A run of good luck may see the player dancing for joy, claiming they are, ‘in the zone’. But all a run like this is, is probability working itself out. Roll a D6, 6 times and in theory you should expect to see one of each number, does this ever happen? No, but if you increase the number of rolls, say to 60, then the chances are you will see close to 10 of each number, and the more often you roll it, the more even the spread.
So, when you first pick up that die and start rolling, you may get four 6’s in the first six rolls – good luck? Really? No, because you can bet your life that you’ll go through a period soon after of not being able to roll a 6 for love nor money. I would say the real luck is in knowing when to trust your instincts, and get out when the going is good!
With all this about randomness and luck in our mind, how does it affect our playing of games?
Random events crop up in the majority of games, and understanding how they work can help to beat the game, or at least give you a fighting chance. If you’re aware of the odds that a certain result will occur during a random event, then you can base your decisions based upon that knowledge – when to take a risk, and when to bail out. This is certainly the way you should approach controlled random events, as they are often controlled for a reason, and if you can work out that reason, then you should be able to use it to your advantage.
With uncontrolled random events, it’s a case of grin and bear it; there is nothing you can do to predict the result, so expect and plan for the worst, and if possible try to bypass them.
Be aware of linked random events, and look for chain reactions – Pandemic is the perfect example of this: An epidemic card cropping up leads to a random draw from the bottom of the infection deck, which then has the discard pile shuffled and placed on top, leading to a controlled random event when the infection draw is made. You can plan for this, as this is an obvious linked event, but others may not be so apparent.
Remember there really is no such thing as luck; it’s just a label we use when probability is running a little wonky! Bad luck will often follow good luck, that’s life, look past it and work the odds.
Finally, a word about ‘luck’ mitigation. Games that include random events, especially when these are harsh on the player, often have a way of mitigating the more severe results – be it with power ups, special actions or equipment a character can gain, or chances to increase the number of dice rolled – One should always be on the lookout for these, and use them to your advantage. If one character can roll a greater number of dice when carrying out a ‘Lore’ test, then make sure you use that character to do just that. Beware the games that have severe random events, but leave the players with nothing in their defence, you’ll get frustrated, and you’ll lose!
Thanks for reading, I’d be interested in hearing your views on luck and randomness; do you agree or disagree with my theories? Please feel free to comment below…