These are some preliminary thoughts on like… making duels interesting in RPGs, following discussion in the @gamerslikeus tabletop server.
I’m looking to videogames for inspiration, specifically, fighting games.
In fighting games, generally the situation is that both players have a lot of different options available to them, and generally every option is risky, i.e. it can be countered - but to counter a move, you have to correctly read or predict what your opponent is trying to do to use the appropriate counter. A successful counter might put you in a place to press an advantage, for example doing a combo that allows no real response, or having ‘frame advantage’ meaning you can do your attack and hitstun the opponent before they can do theirs.
In order to avoid being countered, you try to be unpredictable, not doing the same thing over and over, or the opponent will learn the pattern. This might involve unexpectedly waiting a bit of time so your opponent tries to block too early, then hitting them in the recovery, or switching from one sequence of moves to a different sequence of moves. (I thought that was generally called a ‘mixup’, but Jackie @baeddel told me a mixup is something more specific, basically doing something that requires them to change their counter e.g. going high then low; you can still be predictable even with a mixup.)
A lot of fighting games is the mind game, a very intimate kind of thing where you’re trying to get a sense of what kinds of actions your opponent likes to do, and what they’re likely to do next - and they are of course doing the same. It seems like a pretty natural basis for a way to represent intense, emotional duels in a tabletop RPG, such as in a pirate game.
However, a lot of that doesn’t necessarily translate to RPGs. For example, correctly reading your opponent’s action isn’t really something you can have an analogue for in RPGs, because the difficulty comes from timing - you have a certain number of frames to read the telegraph and input an appropriate response.
Tabletop games generally do not have this same sense of real-time, high-speed decision-making, and also, we want to abstract away things like our ability to react in a fraction of a second, because ‘our characters’ are different people than we are, doing things we couldn’t do.
The dragon in the room
Dungeons & Dragons is the original and most popular roleplaying system, and it has a combat system that makes a large part of the game (at least, under ‘standard’ play assumptions). How this is presented has varied over the years, but at least since 3rd edition, there’s been an emphasis on a kind of tactical combat system. Outside of D&D, a whole lot lot of tabletop games use the same basic paradigm. So let’s investigate D&D…
D&D’s turn-based system is designed for battles between small groups of combatants, rather than individuals. At the beginning of a fight, every character gets an ‘initiative’ roll, which determines the order that people act each ‘round’. Every round, you go through every player in turn, and they can make (depending on their build) one or more actions, and move.
In D&D initiative-based combat, you’re basically always doing what fighting games call a 'trade’ - you both get your hit in, and both take damage. There are very few means to counter what an opponent is trying to do.
That said, unlike fighting games, there’s a random chance to fail, abstracting that ability to counter (by blocking, dodging, having strong armour) to a probability.
So on a purely tactical level (ignoring that this is also supposed to be expressing a story), there’s two kinds of ‘metagame’: firstly, you build your character before the fight to make sure they have the greatest chance to hit, the best chance to passively avoid being hit, the most useful spells and abilities. Then, in-game, it’s more a matter of resource management than predicting and countering. The ‘players’ are generally expected to win: the question is how much it will cost them to do so.
Focusing on 5th edition, which has less of the balance issues that plagued 3rd edition (I’m not as familiar with 4th), each class has a certain amount of resources that refresh with either a ‘short rest’ or a ‘long rest’. For a wizard, they are spells; for a fighter, they’re special combat 'maneuvers’ that are more effective than a regular attack. In addition to these options, there are unlimited-use options - a spellcaster’s cantrips, a fighter’s basic attack.
Some classes in 5e D&D, such as the wizard or cleric, refresh primarily on long rests (as written, the nightly sleep). Others, such as the fighter or warlock, refresh primarily on short rests (sitting down in the dungeon for an hour). Hit points, which prevent you dying, can also be recovered in these rests; you have ‘hit dice’ you can spend during rests, which refresh on long rests. Depending on how frequently they get to rest, one or other kind of class may have more resources. But either way, you only have so many to spend before you have to rest.
So in D&D, the main tactical decisions you’re making are things like: who is the highest priority target, which of my abilities would most effectively negate their ability to harm us and bring us closer to winning, and especially, is it worth spending my limited resources now, or should I save them for later?
In some ways, D&D is more akin to a character action game such as Bayonetta or Devil May Cry than a fighting game. The players are much more capable than their enemies, although there are multiple enemies and they can still be hurt; the question, though, is not usually ‘will the heroes win?’ but ‘how will they win?’.
In a character action game, the complexity is there to allow a kind of artistic expression; you don’t just make the enemies dead but you do so in a very cool way. People compete in competitions to make the coolest, most skillful combat video, juggling the enemies for a long time and chaining together complex series of movies that look cool. In the old days, this would be on VHS tapes! Here’s a more recent one for DMC3:
Now of course D&D doesn’t give you nearly so many options to engage, nor take nearly as much skill to execute, but it’s still a matter of picking your attacks, spells etc. in part to express what kind of person your character is. There’s a bunch of enemies to defeat, and a bunch of ways to do it, each of which says something.
That’s fun and all, but the question came up that it’s not particularly exciting when you want to have a duel with only two characters.
The standard context for a duel is a fencing match of course - part of the inspiration for this post was a pirate game. But let’s also consider what TVTropes calls the Shapeshifter Showdown, and folklorists apparently call a transformation chase. Of which one of the best known examples would be the one in The Sword in the Stone:
D&D has options for shapeshifting magic in spells like shapechange and polymorph, but note the rhythm of this fight: the spellcasters are constantly coming up with counters to what their opponent is doing, and victory arrives when one of them finds a form that can’t be countered.
Likewise in a duel between two people - part of the fun comes from seeing them pull out new stratagems to surprise each other (and us). For example, while I have mixed feelings about The Princess Bride in general, this duel is justifiably iconic (in large part because it’s quite homoerotic, like all the best duels):
Inigo and the ‘Dread Pirate Roberts’ are at first mostly just doing Mornington Cresent-style bullshitting while ‘flynning’ their swords against each other, but even though the techniques are fake, there’s definitely a back and forth, as each one reveals more of their skill and the other matches it or counters it. And the payoff - ‘I am also not left handed’ - is amazing.
This duel reminds me a little of an experience I had playing Absolver last night. I was playing against someone who was clearly quite a lot more experienced than me, and because I’ve barely started playing the game, making a bunch of mistakes (like relying on my not-exactly-great dodging skills rather than using the block button to counter light attacks). I could tell my opponent was kind of going easy on me, backing off and letting me heal to extend the fight. When I started catching on to some of the tricks and playing better, they would quickly use the ‘thumbs up’ emoji. Eventually, I started to actually beat them sometimes (though I suspect they were still going easy on me).
There’s no chat in Absolver, and there’s a difficult line to walk between being encouraging and being condescending, but this player did it really well, and it was an amazing experience that I feel like, taught me a lot.
It reminds me of the approach taken in many parts of SchwaAkari’s Dark Souls 3 videos, where she attempts - far more than trying to win in the most efficient way possible - to give the people she invades a unique, memorable experience, and encourage them to explore aspects of the game they haven’t considered. Some of her duels end up in her being so charmed by the person she’s invading and an intense, satisfying duel that she doesn’t kill them (as the invader is ‘supposed’ to in Dark Souls) but gives them an item and exchange gestures of appreciation.
One of my favourite Akari fights comes quite late, in her 11th video.
If the timecode didn’t work, skip to about 17:28, and keep watching after Akari dies; the full thing’s about 10 minutes in total. This fight is entirely unscripted and viewed exclusively from the player camera, but it’s way more thrilling than a whole lot of choreographed fight scenes and you can feel how much both players are enjoying it.
That’s one thing I really like about the trope of an expert duel, how intimate it is. A visual novel that takes that to extreme is Heaven Will Be Mine by @worstgirlsgames where the line between mecha fights and making out with your opponent is almost invisible. (HWBM is incredible, and also quite hard to describe… the ‘fights’, insofar as they’re even fights, are very abstract!) Another one doing the ‘fightsex’ thing is an old nsfw @porpentine game called Mutant Heat.
While the explicit sexuality wouldn’t really be appropriate for most tables I play at, I’d love to capture that kind of thing, that kind of emotional intensity and engagement.
Of course, it doesn’t have to be like that. Another favourite duel is the agni kai between Zuko and Zhao early in Avatar: The Last Airbender. I think that was the point when the show absolutely won me over. Unfortunately, I don’t seem to be able to find it on Youtube! I recall reading they had a particularly skilled and renowned animator working on this episode, and it shows with the incredibly fluid and detailed animations.
Anyway, if you watch this episode, while this is an example where Zuko and Zhao absolutely hate each others guts rather than where they’re really into each other, it tells a very neat story. First, Zuko is on the offensive, but Zhao confidently and smarmily blocks his attacks. Zuko redoubles his effort, making more complicated moves, but Zhao still deflects them. Then, Zhao comes after Zuko, and Zuko deflects the hits, but he is forced back with each hit. Zuko’s on tilt! Iroh in the background shouts to remember his basics and ‘break his root!’ and the camera focuses on their feet to emphasise the point.
But Zuko is knocked down, Zhao goes to kill him. Then, suddenly Zuko does this amazing breakdancing sweep thing and knocks Zhao’s feet out from under him. We see Zuko plant his feet firmly and smile, and he soon takes offbalance Zhao down - but refuses to kill him. As Zuko walks away, Zhao (of course) attacks, but now Iroh intervenes, easily negating the attack and dropping some absolutely sick burns about Zhao’s dishonourable behaviour in defeat compared to the exiled Zuko (who is, we know, obsessed with honour). In just over a minute, we’ve learned a whole bunch about how firebending works, about the notion of honour in the Fire Nation, and especially about the characters of Zuko, Zhao and Iroh.
There’s loads of great fights in Avatar, but that’s one of my favourites.
Fights where someone has something to protect are also very good. Early in Seirei no Moribito (released overseas as Moribito: Guardian of the Spirit), Balsa has sworn to protect Chagum, but refuses to kill for reasons we learn only later on in the season. She’s being chased by a bunch of samurai who think she’s kidnapped Chagum, and has to fight them off with her spear… (skip to 3:18 if you don’t want to be spoiled on Balsa’s backstory!).
Seirei no Moribito provides an excellent illustration that ‘who dies’ is not necessarily the most interesting stakes for a fight scene. Balsa’s a brilliant spearwoman, probably a better warrior than all the samurai individually, but because she’s not willing to kill them, she’s at a disadvantage - and then her spear breaks! The question is only partly ‘will Balsa die’, but also, ‘will she pushed to the point that she must choose between compromising her principles and killing someone, and fulfilling her goal of protecting Chagum?’
In a later episode, Balsa is pursued by a spearman who is determined to fight her to the death, and he harasses and threatens her and those she’s travelling with when she refuses. After they duel, Balsa becomes convinced that she’s killed him, and, it’s messy… I won’t say what happens next!
Anyway the point is: duels are really cool and more importantly, expressive of characterisation, of the connection between people, all kinds of stuff. And PvP action games, especially fighting games, are similarly expressive. How can we make the mechanics side of an RPG into a useful tool for expressing all the kinds of story we’ve seen above?
So what would happen if you attempted to do that kind of thing in a D&D style combat system?
Mostly, it wouldn’t be as exciting: you’d take turns to attack, maybe using a variety of different attacks, but in most cases, it would be a one way thing - by which I mean, there’s very little you can do in D&D that would cause you to have to respond to something.
That is, ‘stand next to your enemy and attack’ is basically what’s termed a “degenerate strategy”, as in, it’s clearly better than all other strategies so the other options are irrelevant. If your opponent tries to attack and move away, you get an AoO. If they use a Disengage action, you move up and attack, since there’s nobody keeping you there!. So really there’s no reason - unless there’s some kind of terrain advantage to be had - to do anything but stand there hitting each other.
There are exceptions, but you kind of have to build for them. In 3.5e there’s an infamous category of Spiked Chain builds, that allows you to constantly knock enemies prone when they attempt to close with you. With the right splatbooks, this may even let you move away! Still, most duels, you’re not going to have a reason to be moving back and forth, pushing for advantage, doing acrobatics or whatever. Indeed, this usually has the risk of a failed roll.
D&D does have a ‘Dodge’ action, but it’s only a good idea if you have something else threatening your opponent; it’s like a stall tactic in pokémon (we’ll come to pokémon later!). The best outcome of the Dodge action is that you are in the same position you were at the start of the round before. In pokémon, this might be worth it if, say, your enemy has the Toxic condition, or you want to ‘scout’ what moves they have. In D&D, there’s little in the way of ‘damage over time’ effects, and it’s rare that you don’t have enough information to decide what to do. The only exception where ‘dodge’ is useful in 1v1 is if you’re expecting your opponent to use a limited resource, like a spell.
Wizard duels have a similar problem. There’s an amusing strip in OotS where two high-level (3.5e) Clerics have a save-or-lose spell duel:
There are a few exceptions in D&D to the ‘no meaningful counters’ rule. Arcane spellcasters have the option of the counterspell spell, which allows them to spend an equivalent spell slot (min 3rd level) to try to negate an enemy’s spell. That said, unlike fighting games, this rarely puts them in a more advantageous position (e.g. with frame advantage), since the cost of counterspelling is the same as the cost of casting the spell in the first place. At best, if you have more nth-level spell slots left than the enemy, you can exhaust their last one and have a spell slot advantage?
There’s also the use of mobility options such as Fly or Spider Climb. If an enemy has no effective ranged attacks and you do, you can cast Fly and fly out of range and bombard them without them being able to close. This, however, does not tend to make a fight more exciting; it either lets you stomp someone since they can’t touch you, or you have the same ‘just spam attacks’ problem, this time in midair.
That said, this is a very ‘rules as written’ look at what you can do. You can try to make it more interesting through how you describe your actions - the ‘fiction’ as opposed to the closed world of the rules. For example, while mechanically you might be rolling 1d20+attack bonus against your opponent’s armour class, you have the opportunity to say you’re using some clever sword technique, and if you roll a miss, your opponent came up with an equally clever counter.
In the above comic, rather than having the clerics just stand there, the players could have described how Redcloak and the nameless cleric are weaving the threads of magic to cut off each others’ spells, or something.
The role of the fiction
There’s still an issue here. Vincent Baker, creator of Apocalypse World, wrote a post where he views RPGs in terms of arrows pointing back and forth between the fiction, and the ‘real world stuff’ - dice rolls, numbers on character sheets, that kind of thing. He describes three resolution systems, including one where the resolution is entirely handled by the mechanical side - and points to the lack of ‘rightward pointing arrows’, i.e. the fiction has no effect on the mechanics, which is to say, you’re using the same resolution system in every situation.
This means that your descriptions tend to ‘feel’ interchangeable, and therefore, somewhat meaningless. In D&D, the main ‘rightwards-pointing arrow’ is your range, which determines whether you can make a melee attack, ranged attack, etc. At best, at the DM’s discretion, fictional circumstances might give you Advantage or Disadvantage on your attack roll.
Apocalypse World, and its descendent games, seems to be essentially designed to answer that issue. Every mechanical effect in AW is a ‘move’ triggered by some specific event in the fiction, and typically the result is the injection of complications into the fiction in addition to purely mechanical consequences such as losing hit points (though they may also do this with ‘harm’).
In the first edition, there were three combat related moves: for when you’re attempting to force someone to do something by threat of violence (’go aggro’), for when you’re attempting to take something (including abstract things like ‘the high ground’) by force with resistance (‘seize by force’), and for when you’re doing anything in a dangerous situation (‘act under fire’).
In the 2nd edition of Apocalypse World, they make this a little more complex, introducing various tactical moves such as laying down covering fire. They also make explicit provision for duels with the new ‘single combat’ move. This is entirely focused on what fighting games call trades: both parties take harm, but with a good roll it’s possible to do more harm and negate some.
This makes the fights into a game of Markov-chain chicken, where (because the dice rolling system in apoc world favours intermediate outcome over hard success and failures) you will both take severe injuries. The fight only ends when both parties are willing to break it off, or one side surrenders or dies; the question is how far you’re willing to take it for whatever you want to achieve. I think of the pilot duel at the end of Porco Rosso, where the protagonist and a rival pilot wreck each others planes then beat the crap out of each other on the ground until both are exhausted.
I heard an example of such a duel in the podcast Tabletop Hell. It was devastating for the character who got in the duel - but it didn’t really involve a lot of decision making, just: fight, or surrender? It’s very suitable for a gritty, post-apocalyptic game where every victory comes at a cost. Ultimately, the player character won, but was severely injured.
But if we want something a little less grim, and perhaps trying to capture that interplay of counters, prediction and mind games that fighting games have, what could we do?
Simultaneous resolution games
Most tabletop roleplaying games - and indeed, most board games - proceed with one player having a turn, when they can decide how to act, and then another player getting to choose their action in response to the updated situation. This paradigm goes back to some of the oldest board games like Chess and Go, and can clearly support all kinds of complexity!
Indeed, though I don’t really play them enough to be able to comment with much authority, trading card games like Magic, Netrunner or Yu-Gi-Oh support some of the metagame features of fighting games we’ve seen, like playing counters to particular cards or strategies. In part, that’s because strategies take several turns to complete. There’s a huge variety of carefully analysed chess openings, and even more so, in Go there are joseki which are established sequences of ‘optimal’ play for both sides, that can be deployed against each other at different places on the board.
There is another option, though, which I find pretty interesting: simultaneous resolution. In this situation, both players decide their action in advance of the turn, without knowing what the opponent is going to do.
The classic example of such a game is of course Rock Paper Scissors, and all the many variants such as Rock Paper Scissors Lizard Spock or even RPS-101 which add more options. This is about the simplest simultaneous-resolution game, and is inherently balanced. It’s not all that exciting, however: it ‘feels’ identical no matter what you’re doing, and there’s no real way for the fiction to enter into it except in very crude ways (e.g. you play multiple games, and the fiction gives one player or another some free wins).
Another famous example of such a ‘game’ is more of a thought experiment for developing game theory: the Prisoner’s Dilemma, and its iterated variant. There, the two players pick whether to ‘trust’ or ‘betray’ each other; ‘both trust’ is better than ‘both betray’, but for each player, ‘betray’ individually beats ‘trust’ whatever the opponent picks; so in a single game, the worst outcome, ‘both betray’, tends to happen. A lot of writing has been put towards winning strategies in ‘iterated’ version, where you can respond to your opponent’s previous choices, and how this corresponds to models of society. Which strategy ‘wins’ depends on what other strategies are in play.
While kind of interesting, this is tangential to our situation. We’re looking at a kind of ‘zero sum game’: there will generally be a winner and a loser, and the fun comes from the struggle to determine who is who.
In board games, Diplomacy is maybe the most famous simultaneous-resolution game? I have barely played Diplomacy, but it’s mainly a game of negotiation and lying. Between each round, you discuss with people, promise alliances, etc., and then secretly record your action, such as ‘invade another nation with this army’. Invasions generally only succeed with an alliance of nations attacking, so to get ahead you need to mislead people as to your intentions, and correctly judge who’s lying to you. Interesting, but not all that relevant to our 1v1 duel game.
OK, those aside: the existing examples I’m going to talk about are Burning Wheel and competitive Pokémon. Although they’re turn-based, I think they both capture quite a few of the things I was talking about in fighting games!
In Pokémon battles, each player brings six pokémon to the game, each of which has its own hitpoint total before it faints and is removed from play. Turns are resolved simultaneously; you can either pick one of the four moves your pokémon knows, or switch to a different pokémon.
Actions aren’t resolved perfectly simultaneously. Rather, each pokémon has a ‘speed’ stat, depending on the species of pokémon, how much the player has invested in speed EVs, and certain items such as a Choice Scarf or stat-boosting moves. Each turn, the players pick their actions, then the pokémons’ moves are resolved in order of speed. Certain moves such as Protect and Quick Attack resolve at a higher ‘priority’, and switching pokémon has the highest priority.
There is a lot of prediction in pokémon. For example, suppose you know your opponent has a super effective move that could severely damage your current pokémon, or burn it and severely reduce its ability to attack, or hit it with toxic poison that would make it ineffective as a stall pokémon… the opposing pokémon is a counter or check to yours. A sensible thing to do would be to switch your pokémon to a different one that resists that predicted move. But, your opponent knows you might do that - and if they assume you’ll switch, then they should instead pick a move that will hit the pokémon they expect to come in. But that move might do little damage to the pokémon you currently have in.
(In this randoms game, I figured my opponent (@baeddel! <3) would switch out, since Victreebel can hit Suicune super effectively. So I clicked something else - maybe Sludge Bomb - to cover whatever might come in better that would resist Grass!)
So, if you think they’re going to try to cover a switch, you might stay in and grab a ‘free’ attack instead; in this way, you gain the momentum, a somewhat vague concept parallel to sente in Go (at least, under one definition); your opponent is forced to respond to you, rather than the other way round.
(Instead, she stayed in and hit me with ice beam. it wasn’t devastating but I lost what could have been a big advantage!)
A lot of high level pokémon play (and I am very definitely not a high level player lol) is seizing and holding momentum. For example, switch in a hard counter to a pokémon, forcing the opponent to switch out; then you can do something that forces their next move and so on. Sometimes, for example, you might double switch - bring in a pokémon to force a pokémon out, then immediately switch it for a different pokémon that can counter whatever you expect your opponent to switch in to counter yours!
When you knock out a pokémon, your opponent has the opportunity to ‘freely’ switch in any of their pokémon (without risk of being hit on the turn they switch in), meaning they can switch in a fast but fragile pokémon to ‘revenge kill’ the one that just killed their pokémon. In this way, they have a chance to take back the momentum.
I feel like any 1v1 simultaneous-resolution game with a ‘speed stat’ would probably end up feeling a lot like Pokémon, but of course Pokémon can only support that much complexity because it’s on a computer! A game tends to last for tens of turns, but that’s fine because the computer is doing all the calculations for you so the turns go by quickly.
There were attempts to make a Pokémon tabletop game in Pokémon Tabletop Adventures/United; I believe they used D&D’s D20 system as their starting point, and tended to move away from exactly replicating the mechanics of the game. In particularly, they abandon the simultaneous resolution aspect in favour of an initiative system like D&D (though without rolling).
Pokémon has a huge variety of situational effects, status effects, stat boost multipliers, items etc., plus a very complicated graph of damage types. Although the basic structure is simple, it would be very difficult to run a manual game of Pokémon using the game rules!
The Burning Wheel system is better known for other aspects such as its ‘beliefs’ system, the ‘say yes or roll the dice’ principle, and other influential things. But it is probably the main example in a tabletop RPG I know that has the simultaneous-resolution system I’ve been describing.
The Burning Wheel system is one of the most mechanically complicated RPG systems I’ve ever encountered - to the point that it’s one of two different fight resolution systems, used only for really important battles when a lot is at stake.
Burning Wheel has a very complicated system for resolving fights with simultaneous resolution, taking into account different actions taking up different amounts of time, the effect of spacing and all kinds of other factors. This is actually rather different from the general philosophy of Burning Wheel, which encourages you to roll once for the outcome of a situation and then ‘let it ride’.
Burning Wheel actually has three such simultaneous resolution systems: apart from physical combat, the other one is the ‘duel of wits’ simulating an intense debate where you try to convince a person or audience. The example given in the rulebook is an argument about whether a clan of Dwarves should join other nations in going to war, or stay out of it. A huge part of Burning Wheel is establishing characters’ intent in a scene, i.e. what is at stake.
Let’s go over these systems in turn. They’re quite similar, and I won’t try to list all the nuances, just explain how they relate to the concepts we’ve been discussing above.
Both systems revolve around a series of ‘volleys’, in which players script their actions several turns in advance. This necessarily requires a meaningful choice of actions. There’s a huge list of actions:
Strike, Great Strike, Block & Strike, Lock & Strike, Avoid, Block, Counterstrike, Assess, Change Stance, Charge/Tackle, Draw Weapon, Get UP, Lock, Push, Physical Action, Beat, Disarm, Feint, Throw Person, Aim, Fire Gun/Crossbow, Nook and Draw, Release Bow, Snapshot, Throw Weapon, Cast Spell, Drop Spell, Command Spirit, Sing/Howl/Pray, Command, Intimidate, Fall Prone, Run Screaming, Stand & Drool, Swoon
So that’s fun. There’s also an entirely different subsystem for approaching an enemy from a distance, and having a ranged battle:
Close, Sneak In, Flank, Charge, Maintain Distance, Hold Position, Withdraw, Sneak Out, Fall Back, Retreat, Fall Prone, Run Screaming, Stand & Drool, Swoon
The Duel of Wits ones are not quite as long, but still pretty varied.
Avoid the Topic, Dismiss, Feint, Incite, Obfuscate, Point, Rebuttal, Cast Spell, Command Spirit, Drop Spell, Sing/Howl/Pray, Fall Prone, Run Screaming, Stand & Drool, Swoon
In all three cases, you script out three ‘volleys’ in advance. In the Duel of Wits, everyone gets one action per volley; in the combat rules, it’s more complicated. That said, in the Duel of Wits, when you get to that volley, you have to express how you’re making the point, avoiding the topic, dismissing their point etc.! You can’t just pick an action.
The resolution method is determined according to a table: e.g. if I pick Avoid the Topic, and my opponent is making a Point, we roll against each other. The players have a variety of options for how they make this roll depending on the action, according to Burning Wheel’s standard resolution system. Generally, for the Duel of Wits, the outcome will be to subtract some points from the ‘body of argument’, or negate an opposing move.
There’s very much a question of prediction and countering here. For example, one example paragraph in the rules says:
My opponent’s been playing pretty conservatively. I’ll bet he thinks I’m going to come on strong in the first volley. Rather than script a Point, I’ll put in a Feint to blast by his predictable Rebuttal. If I’m right and he scripts that Rebuttal, my Feint will ignore it and I’ll get a free attack against his body of argument. If not, and he scripts a Point or Dismiss, I’ll be in trouble.
So there’s a lot of parallels here with stuff we’ve discussed above. Doing a Point against a Point would be a trade; otherwise, you play the metagame. For example, if you Dismiss too early, you can get punished hard for it since you skip the next turn, so it’s designed to decisively end a duel of wits.
The main drawback of the Burning Wheel system is how incredibly complicated it is to learn, and how much effort it takes to run! I’ve played one game of Burning Wheel at university. In that game (before I transitioned and everything) my industrialist character was arguing with someone protesting the injuries caused by my character’s machinery… frankly, he should not have won. Still, it was a blast to play, but it took up most of the session and it took most of my effort to follow the rules and think up what my character’s arguments would be - rather than strategise, predict, counter etc.! I picked options more or less at random.
Consider what it takes to resolve one volley:
- we decide our actions in advance
- we read out our actions, and consult the table to work out what we’re supposed to roll. we pick which skills we’re going to use (whether we’re going to use e.g. Persuasion, Stentorious Debate, Susasion, Poisonous Platitudes… that’s just a sample, BW has a lot of possible skills!)
- we do some acting, make our short speeches (or describe them) according to the action and skill chosen
- different actions have different complications; e.g. making a Rebuttal requires you to split dice between attack and defence. Most likely, both players end up rolling.
- to resolve a roll, they need to go through the full procedure of choosing the relevant skill that their character has, suggesting Fields of Related Knowledge to get bonus dice, rolling and adding up successes
- then you probably have to do some maths. If it’s an opposed roll, work out the margin of success. either way, probably subtract something from the body of argument.
You need to do this two more times before you even get to make another tactical decision!
We never played a combat in that game of BW at uni; it wasn’t that kind of campaign. But the combat system is, if anything, even more complicated. The number of actions you have per three-volley exchange depends on your stats, there’s a whole separate thing for working out who’s at the ideal range for attacking if you have weapons of different lengths (six different length classes), various stances, and all kinds of small bonuses and penalties depending on what happens each round.
It is probably the closest thing to simulating all the nuances of a fighting game or action game I’ve ever seen in an RPG. It’s also… well, maybe it works at the table once you’re familiar with it, but it sounds absolutely exhausting! I think at that point I’d want to write a computer program to handle all the interactions, so the players can focus on the strategy.
Is there a middle ground, between the exhaustive complexity of Burning Wheel and the ‘roll to hit repeatedly’ of D&D?
Monkish Combat in the Arena of Promotion
@baeddel told me about a system described in the 2nd issue(!!!) of Dragon magazine, all the way back in 1976, which used a simultaneous resolution system. You script six moves in advance, and different blocks are effective against different kinds of hits. The long scripting could be changed if you succeed on a random roll. Various moves advance forwards and backwards; there’s striking range and kicking range. Apparently it’s based on the ‘En Guard’ rules, though I don’t know where those are.
Resolution is not super clear. I believe it’s not perfectly simultaneous, but does depend on the opponent’s move; the monk with more HP moves first, but if the other monk is blocking, their attack may be negated.
I don’t know how fun this game would be to play, and I haven’t made any attempt to check for degenerate strategies.
Fight! the Fighting Game RPG
This one came up in a random /tg/ thread I found on google. I have no idea if it’s good or not, but the blurb suggests it does some of the stuff I’ve been talking about!
A gesture towards designing a system
This is going to be just a sketch, because this post is already very long and it’s quite late at night.
What we want: a system that produces dramatic, emotionally intense duels, with reversals of fortune, and a fair amount of strategy beyond ‘attack repeatedly’. I also enjoy simultaneous resolution so let’s try and build it within that kind of framework.
We don’t want to build the system just by listing ‘things that people could do in a fight’, but rather we want a functional prediction game.
My first thought is to try building the actions for such a system in an iterative way. I’m not exactly sure it will ‘work’. I’m taking a little inspiration from the priority system used in foil and sabre fencing, but not trying to recreate it exactly.
Suppose you have two actions: attack and parry/dodge. If you both attack, it’s a trade and you both take damage. If you both try to parry, nothing happens. if one person attacks and the other person parries, then the parrier negates the damage, and gets some advantage - perhaps a free attack that can’t be dodged.
This system sucks of course. It’s a game of chicken: parrying beats attacking, but if you keep parrying, the fight will never end; but whoever breaks the chain of parries and attacks first gets punished for it, then it’s back to square one.
Let’s add another action: a feint. If you feint and the other person parries, then you get an attack that can’t be parried. But if you feint and the other person attacks, you are of course not parrying it!
So now we’ve reinvented rock paper scissors. Feint beats parry, attack beats feint, parry beats attack, with the winner of the exchange dealing damage. If you both choose ‘attack’, it’s a trade; otherwise if you both pick the same nothing happens.
That 'works’ - rock paper scissors is a balanced game! - but as we said above, it’s not very interesting. Still, from this starting triangle, we want to add more complexity.
Here’s some possible avenues to expand:
- different types of attack: high, low, etc.? and you have to say which you’re guarding
- stances - taking more risks to be more aggressive, or playing it safe. we want to make sure we favour aggressive playstyles over ‘turtling’!
- telegraphs - maybe a scouting move which gives you a chance to gain information on your opponents intent in the next clash?
- a range system: moving forwards and backwards to range out attacks (e.g. an attack that moves you closer, an attack that moves you further), different options at different ranges, the equivalent of a fencing flèche to rapidly close with an opponent
- strong moves that require some kind of setup, so high risk high reward?
- some kind of system for dealing with emotional state and tilt?
- damage to weapons and disarming people - a very common trope in anime to add unexpected complications to a fight!
- plenty of room to pause the fight - for breath - and speak to their opponents.
- feedback, positive and negative. positive feedback helps someone move an advantage towards victory. negative feedback helps a player who’s fallen behind catch up. both kinds are necessary: ideally we want a kind of fleeting advantage, so that a player can definitively win an exchange, but it will ultimately reset so they can’t just totally sweep their opponent without difficulty.
- an injury system that applies penalties would be an example of a strong positive feedback system. winning a fight even when injured is a cool trope, but any such system would have to be handled very carefully - Jackie often reminds me there’s a reason hit points are so ubiquitous!
- also injury systems tend to be much less granular that hit points: you are injured in a certain way, or you’re not, whereas hitpoints have as many different possible states as there are hitpoints. so injury systems are less flexible.
- footwork: how secure you are in your stance, how quickly you can move, that kind of thing. maybe the advantage/disadvantage system could involve disrupting their footing?
One absolutely classic scene in a lot of media is where the fight is won and the hero has the enemy at their mercy, and has the choice of whether or not to kill them. We don’t necessarily want this in every fight, but that’s obviously a very expressive moment - think of Zuko refusing to kill Zhao, and by contrast, his own father burning him at the end of their abusive, cruel duel! (As in the Zuko fight, often, media will try to have its cake and eat it, with that cliché where the hero shows mercy, then the defeated enemy gets up and attacks them, but the hero - or perhaps someone else - immediately dispatches them.)
But anyway, one of the possible outcomes of the fight should be that the enemy is defeated, but not dead.
In D&D, this is managed through the abstraction of hitpoints. Characters do not take actual injuries (except for minor cuts) until they hit zero hitpoints; rather than a degree of injury, hitpoints represent the character’s like, ‘energy’ to stay in the fight.
Of course, there’s also the classic Kurosawa samurai duel, where they exchange decisive blows and we don’t see who wins for a few moments as the injury like… takes effect or something lol. That probably wouldn’t be covered so well by the system we’re designing - we’re looking for a prolonged fight, with lots of back and forth, advantage shifting etc.
What this system is currently missing is ‘rightward pointing arrows’ - all the suggestions above are self-contained, or rather, ‘leftward pointing arrows’; you make a mechanical choice and it affects the fiction. So that’s another aspect to think about.
Good game design is hard; I’ll come back to this. I’m thinking of building a system around my setting Let us loot the chambers of God’s heart! (which I’m going to post about another time! it’s not ready for public exposure yet lol), and this should be a component of it.
(Huge huge thanks for Jackie @baeddel helping me write this post and all the conversations with her that inspired it! I am gay)