Player agency over dice randomness

As I explained last time, I'm a mechanically-focused gamer. I enjoy engaging with various game systems to figure out in what ways they work, and in what ways they don't. Last time I covered the difference rolling two dice instead of one can have on game feel, and this time I would like to talk about how a few systems handle player agency over dice randomness. By that I mean various mechanics and resources the players can use to influence their dice rolls, especially the important ones.

The problem - roll and pray

At least once every session a player will come across a situation that is very important to them - perhaps they are down to one last hit point and need to kill the enemy before they themselves are killed, or the character is preparing for a hack that is a culmination of the adventure and their paycheck depends on it. Those dramatic moments where the players wish there was anything they could do to improve the odds, but alas - they can only roll and pray.

I really don't like such situations of being powerless and not having some mechanic one can leverage to improve those odds. Luckily, there are a number of options that various systems have implemented.

Rerolls

The simplest approach to addressing the issue is letting the players have a reroll handy. Those often are a resource shared either between the party or given to the characters individually. Stars Without Number has an Expert class that can reroll one non-combat roll each scene, Savage Worlds gives players a few Bennys per session they can use as a reroll, Godbound has a Word of Luck that is all about manipulating rolls and causing rerolls in various ways.

In general, rerolls are a good option when the characters are somewhat competent at a task and should be able to normally succeed at it. It's a bit less useful when the character is trying to tackle something they are woefully unprepared for - if it's unlikely they will succeed, they will likely still fail with a reroll.

Improving the odds

Another approach that can be taken is improving the odds of the roll before it is made. Chronicles of Darkness gives you more dice to roll if you have the proper gear, and you can further improve your odds by spending Willpower - a limited, refreshable resource. The system also features a number of ways to fudge the dice math to imrove the chance of an exceptional success by using 9-again or 8-again rules. Star Trek Adventures makes you accrue Momentum by performing tasks you're good as so you can buy more dice for a roll when it's really needed. CoD, STA and Stars Without Number give the option of other characters assisting the primary character performing a roll - usually a success by the assistant gives a small boon to the primary roll.

Overall, improving the odds can be especially useful when the character isn't as competent - even a small boon to an unskilled character usually translates to much better odds than adding a simple reroll. They also usually are simpler to justify narratively - you have a clear explanation of why you're getting that +2 to the roll.

Success, but at a cost

Success in a role playing game is often seen as binary - you either succeed, or you don't. Maybe you also have a botch or an exceptional success, but either way - the line is drawn hard. However, there are a few systems where there is a gradient to the failure.

In The Veil (or Powered by the Apocalypse in general), you almost always have two or three gradients of a success. You can outright fail, succeed completely, or in the middle - succeed at a cost. Perhaps you don't get everything you wanted, or there is some complication that occurred as a result of your actions. Star Trek Adventures gives players an option to Succeed At a Cost instead of accepting a failed loss. This allows the task to be successful, but some complication will arise as a result of that roll.

This option is useful for fudging the numbers a little - if the player is just slightly off from succeeding. It's always useful to have it as an option to allow the player to decide whether their task is important enough to warrant the extra cost.

Controlled botches

Somewhat related to the previous section, although distinct enough to warrant its own. In a lot of games you will find rules for botching a roll - failing so badly it's causing some trouble. World of Darkness up to Chronicles of Darkness had a rule where rolling 1s meant subtracting from successes and causing botches if their amount got too big for example. While fun in their own right at times, it's perhaps best when this option is left to the player - letting them decide when they want to be discovered, mislead and so on to create a more interesting game.

Chronicles of Darkness is perhaps a good example of how to give players agency over their botches. When a roll is a failure, the player can opt to make it a dramatic failure to get a beat (a point of XP basically). Outside of attempting a roll with only a chance die, this is the only way to botch in the system, giving the players agency over when it occurs, and rewarding them for allowing it to happen (hopefully giving the GM a chance to make the game more interesting as a result).

Trumping the roll

Some games allow the characters to outright trump the roll with some usually high-end powers or skills. Stars Without Number have some Psychic and True AI powers that roll back an entire turn or a single roll, or go as far as dictating an entire turn and all of its rolls completely (max level True AI power). The same system also features a Warrior class that can auto-succeed at one combat roll per scene, or negate an incoming attack completely. Godbound has one Gift that once per character's entire life they can succeed at any undertaking, albeit only at a basic level.

Those powers are neat, but they need to be carefully balanced not to throw the entire game off balance.

Conclusions

There are many ways RPG systems have figured out how to allow the players to influence their rolls. It's good to have at least once option available to the players, and multiple of these examples can be used simultaneously without feeling like they're overlapping too much.

Scifi media as a blueprint for scifi elements in your game

Recently, my group started playing a game of Stars Without Number. The setting of that game is very light on the backstory and the sector you play in will be a giant sandbox you fill in yourself. The game is very conducive to a lot of scifi - you could play it as Firefly, Battlestar Galactica, Cowboy Bebop, Altered Carbon, Fading Suns, etc. What you decide to fill your sector with is up to you.

This gives the GM and the players a lot of flexibility as to how to create the setting. The vanilla Stars Without Number assumes present history happened, humanity built its own FLT ships in the near future, expanded outwards into the far reaches of space, everything collapsed 600 years ago due to The Scream, and now the worlds are connecting up again. This means you can get very varied levels of technology, very varied types of societies, and very specific feels for each solar system.

Swan Song for example decided to build a few planets and organisations around the current or past cultures of Earth. You had the Sunbeam Multistellar megacorporation that was heavily steeped in cowboy culture and mannerisms. The Hoveydan Caliphate was influenced by the cultures of the middle east. Sigrid is viking and Norse through and through.

In our game of Blackstar, we stumbled upon our gimmick for the sector by chance.

Scifi as a blueprint

So here's the idea - in the modern world, a lot of things are influenced by what came before. Scifi has inspired modern technology, the US government was built on the ideas of a classical Greek ideas and the Roman Republic, and even a lot of our fiction draws from the history - talking about emperors, feudal societies, pirates, etc. A lot of things are built on top of blueprints that came before.

So the core conceit for our game is that since the modern history we live in today did happen in the Stars Without Number universe, it's quite possible a lot of modern books, novels, films, or culture in general was passed down over generations, being remixed and remade over and over. Because of that, those stories have become ingrained into the society - you can reference the Alien movie and describe something as a xenomorph just as casually as you can reference the Odyssey and call something a siren or a cyclops.

With the scifi proliferation that exists in the universe, it would then be natural for people and societies to build things based on those concepts. A xenomorph is not only a creature of legends, but a blueprint for a gene engineered killer machine. Starship Troopers is a series of stories from a past, but also an idea on which to build a society. Lightsabres are mythical weapons, but they have also been made real with futuristic technology.

In other words - the scifi contemporary to us the players formed a set of blueprints for the game setting to build around. While the events of that scifi didn't happen (you did not have the God Emperor of Mankind undertake the Great Crusade against the xenos), but the ideas from them did influence the development of the future timeline, and they did inspire various entities to turn some of those ideas into reality with the future tech.

Heck, even Stars Without Number's supplement Suns of Gold refers to space merchants that rule over primitive societies with their future tech as a "kurtzer" in reference to Kurtz from Heart of Darkness. Kurtz went from being a character in a 1899 novella to a template of behaviour that is recognised in the universe of the game in year 3200.

Because of this you could have things like one of the PCs, as a True AI in a Synth shell, looking like Roy Batty, a replicant from Bladerunner. Obviously, whoever designed that shell took inspiration from the old fiction and based the shell on the designs from ancient Earth. Or you could have predators from the Predator franchise appearing as a society - some gene engineer obviously modified some rich big game hunter and gave them powerful weapons so they could go across the galaxy and hunt the most dangerous prey.

This sort of approach to filling a scifi universe gives the GM and the players an excuse to draw from any number of sources without having to reconcile why you would suddenly have stormtroopers shooting xenomorphs - neither of those franchises did happen, but the ideas from those works of fiction did take place. If Picard can quote Shakespeare, he should also be able to quote Star Wars after all ;).

Conclusions

In a sandbox scifi game that branches off our real world timeline like Stars Without Number, it's not unreasonable to use modern scifi as a blueprint for elements that can be found in the game. With sufficiently advanced technology, it would be possible to copy ideas and concept from such works of fiction and make them real in the game's world. This would allow the GM and the players to build on top of such scifi and take elements from them as needed without worrying about keeping the world consistent with those works of fiction.

The Doylist choices in role play games

In the culmination of a recent game of Stars Without Number an interesting trope sneaked up on us. We were fighting the cartoonish villain Thanatosis and her Hungry Boys in a Mad Max style dust thunderstorm. (episode end spoiler alert) After having her tank blown open and surviving a sniper rifle shot to the back of her head, she decided to retreat away from battle and ride away into the dust storm cursing the name of one of the PCs.

Our characters could've pursued her and put her down without too much trouble, and at least one PC did want to do that, but as players we decided to let her go. On one hand, it was because our characters had other things to attend to, on the other hand, it made more sense from the narrative perspective to have her as a new nemesis that may one day return to fight our characters.

Thinking and talking it over later, that's how I stumbled upon the "Watsonian vs. Doylist" trope.

Watsonian vs. Doylist trope

The Watsonian vs. Doylist trope related to the dichotomy between the character in a story and the author of that story, or in RPG space - the player character and the player. It refers to Sherlock Holmes and two types of commentary you could have on the events of the book - the perspective of the in-universe Dr. Watson, explaining things as he understands them, or from the out-of-universe perspective of the book's author, Arthur Conan Doyle, explaining things as the author of the book.

An example of this trope is trying to answer the question of "why are there so many human-like aliens in Star Trek?" Watsonian answer is that an ancient humanoid race seeded the galaxy with all of those aliens. The Doylist answer is that making human-like aliens is cheaper for the show and allows the audience to understand the characters easier. Both sides answer the same question in a relevant manner, but give incompatibly different reasonings.

The Watsonian choices

In terms of role play games, we often focus on playing our characters in a Watsonian way - we are very attached to our character, want to play them "optimally" and make sure we don't leave any hanging threads the GM could try to weaponize against us later. This is basic human nature - we are very loss-averse, and since we associate our characters as being a representations of us, we don't want to see them die.

This is a fine approach to take, but at the same time, it makes the characters act too neatly and perfectly. They don't make too many mistakes, they don't act impulsively against their better judgement, etc.

One approach I've seen this issue addressed was the Limit Break mechanic in Exalted. This is a mechanic that helps the characters act like the classical tragic hero of myth they are meant to emulate. Basically, the characters accumulate emotional baggage over time until they snap and take the self-destructive course of action - they might blindly charge into battle to their possible death, lash out at their allies straining that relationship, or may be frozen with indecision at the climax of a conflict.

The Doylist choices

On the other end of the spectrum, we would have the idea of playing the character in a Doylist way - making decisions that tell the better story, but may go against what's best for the character in question. This would mean letting some antagonists live so they may become a nemesis for the character, or deciding to go on a drunken bender and break the law so the player could be cased with the challenge of escaping the police or having to cover something up.

A good example of taking the Doylist choice comes from Roll Play Swan Song, an actual play game of Stars Without Number. (spoiler alert for Week 1 and a game-long side-plot) In the first session of the game, one of the PCs, Higgins, ran into an old contact, Randy, that helped them get off the planet with an unknown parcel for their mission. In exchange, Randy wanted to get a ride off the planet - easy enough, basically a free favour. However, Higgins' player decided it would be more interesting to betray Randy at the last moment, shooting him in the head on the landing pad, in order "not to leave any lose threads".

Because of that action, the character has antagonised a criminal to whom Randy owed some money. This would later mean they would have a bounty hunter come after their heads. When the players would return to the planet later to help save some civilians, Higgins would be recognised by the military, making him have to flee and possibly leaving a lot of people to die. In the final episode of the game, Randy would also make a return with a cybernetically fixed head only to snipe Higgins dead as a revenge (luckily, Piani was there to psi-heal his exploded head back up).

All of this happened because the player decided to take the Doylist choice (the player even confirmed the choice was deliberately made to make the game more interesting). The game and the podcast were better for it, even if the character had to suffer because of it.

I think I saw the Doylist approach in the actual game books only twice. Chronicles of Darkness give the player the choice when their characters botch their rolls in exchange for XP.

The only diegetic example of this approach could perhaps be found in Exalted: The Fair Folk. In the world of Exalted, there exist entities known as Rakshasa, The Fair Folk. They are being made out of the primordial chaos from outside of the world that assume humanoid shapes to weave narratives to strengthen themselves. Essentially, they act out stories and conflict like actors (meaning playing Rakshasa is the player playing a character playing a character). Rakshasa become more powerful the more connections they have, meaning they seek to cultivate a web of nemesis, rivals, lovers, underlings and so on. This means for them the Watsonian and Doylist choices are one and the same - the players want to make mistakes because their characters want to make mistakes because that creates a more compelling narrative for them.

But let's reel this back a bit from this crazy meta-example ;).

An agreement between the player and the GM

Obviously, making the Doylist choice in a game leaves the PC and thus the player more exposed to the GM and their narrative. It would therefore be good to discuss this topic ahead of time and develop a mutual understanding between the players and the GM. The GMs shouldn't take those vulnerabilities and possible nemesis and just use them to screw over the player willy-nilly. Just as the player made a wink to the GM while letting that villain go, so should the GM at some point in the future wink back to the player.

If you want to quantify this relationship in some sense, perhaps each time the player makes such conscious Doylist choice, they would get some point or token. In exchange, the GM would get to keep that plot thread or character that would later come back "to take revenge" on that character (in a more narrative sense, there may be no actual killing involved). However, the player could cash those tokens in at any time to get something significant they want. It could be that they would escape being killed in some situation, or be able to get some artefact that would normally be out of their reach, or the like. It would have to be something significant enough to make the Doylist choice worth it, although not as significant as to make it some sort of wish-granting engine ( ;) ). You'd be basically cashing in "I'm not doing the adventure that would've enabled me to get this thing" in exchange for an "IOU of an adventure surrounding this plot at some point in the future".

If the GM brings back that same villain back and they are out for revenge, or the situation bites the players in the ass because of their past mistake, the debt is settled. If that character or plot would come back later again, it shouldn't be any more dangerous than any other adventure would be - you don't necessarily want to have to deal with some crazy person plotting to murder you every other session just because you let them live once.

On the other hand, if the player's overcome the obstacle and they get the upper hand, there is nothing stopping the player from taking another Doylist choice, earning another token and letting the GM keep the villain once more.

Also worth noting is that the "revenge" part of the deal can come in at any point really - if the GM wants the villain to try shooting the PC in the back right after they have been spared by the player, that's fine. If it worked for Dragonball (Goku sparing Frieza only to be shot in the back), it can work for your game too ;).

Conclusions

Players are often inclined to make Watsonian choices to do what's best for their characters and avoid leaving any lose threats least they be used against them. However, sometimes making a Doylist choice would make the story and the character more interesting.

It's good for the GM and the players to discuss this topic and come to some mutual understanding of how they would want to use this approach in their games. The players would be more encouraged to make Doylist choices if they knew the GM would handle that in a responsible manner.