Problem-Solving Combat: Example in Cunning Knave

Spoilers for my players!

I've run three sessions of my first draft of Cunning Knave, and they've gone quite well. Only in the third session did we end up in an unambiguous combat situation, so I had an opportunity to see how the Problem-Solving Combat rules felt in play. I want to provide a summary here to give some concrete examples of what I'm talking about in Problem-Solving Combat: Breaking the Cage of "Roll Initiative", which includes part of the rules used here.

Imperfect memory has altered some things and some rough edges are smoothed over - don't take this as an exact log of events, but a combination dramatized play report and idealized example of Problem Solving Combat using Cunning Knave, with some commentary about how I adjudicated things on the fly.

I wrap up with a continuation of my thoughts on this style of combat from my initial post, which I think doesn't quite stand on itself enough to get my point across. I may follow this up with a third post on the topic, focusing on advice and procedures for generating and evolving interesting combat situations, for players to sink their problem-solving teeth into. And I intend to eventually release Cunning Knave into the wild as well.

The Situation

The Battlefield: 

Room 13 of The Sky-Blind Spire, essentially a large room with two broad staircases with landings at the top, bottom, and middle.

The Foes: 

4 Shadow Puppet soldiers from Through Ultan's Door Issue 1.

HD 1, AC 11(Asc), polearm or scimitar, MV12 MR12, Special: Surprise on 1-4 in dim light, dispelled by Light spell. "They will melt away if they suffer damage. They will drag fresh corpses to [area in TUD]."

A few things to note here for conversion to Cunning Knave. 1 HD means just one good hit will take it out. 12 Morale means they won't be dissuaded. And since they want corpses, I judge that means a downed PC will be captured instead of killed outright (yes, it does say corpses, but "captured alive" is so much more interesting than the alternative).

I'm using TUD as a kind of intrusion into SBS, so I roll a D20 to determine which room the door to Zyan had opened in, and hence from where they were approaching - Room 8, which was just on the other side of the door at the bottom landing.

The Party and some of their Relevant Implements: 

  • Jules (spear, torch)
  • Garmayne (dagger, net, smoke bombs)
  • Ingram (warhammer)
  • Sybil (buckler, lantern)
  • Adelaide (longsword, caltrops)

All items were randomly rolled. None have much armor to speak of (Armor Defenses of 11 or 12). And keep in mind that the PCs get taken out of action with only one "hit", and are then in danger of death if the foes wish to kill them and the survivors can't prevent it.


I had rolled an encounter near the end of the prior session, so I kept it in my pocket. At the beginning of this session, we picked up with them taking a look at the details of the room they were in.

I considered what the initial challenge of this combat would be. The obvious choice was to take advantage of the creatures' shadowy nature, particularly in this room which I had already described was rippling with strange shadows. So I chose for them to essentially have a surprise round, getting the drop on the party. That could be pretty brutal though, so I decided to pull the first punch and have only one of the 4 Shadows take part in the initial ambush. Besides, revealing the others mid-battle would be fun, and embody the kind of shifting situation I want from this style of combat.

Surprise Round

I asked what kind of investigations everyone was doing: While Garmayne, Ingram, and Sybil inspected the statues on the upper landing, Jules and Adelaide go to peer down a doorway on the middle landing that they haven't been down yet.

Since the Puppets were emerging from the lower landing, Jules and Adelaide would be in danger first. I asked Jules for a Wisdom roll to determine how he would be able to react to a surprise shadow-spear-thrust.

He rolled a 15, which in vanilla Knave means a failure - a Save needs to roll over 15 to succeed. However, in Cunning Knave, this activates a new rule - the opportunity for a Complicated Success. I really enjoy the concept of partial successes from PbtA games, but I have found it fatiguing to keep them interesting when they are often the most likely result - but 1 in 20 makes it a rare thing, which I think and hope makes it more exciting, like an interactive Critical. So the way it works in CK is that, normally the attempt fails, but either the player or the GM can propose a cost or complication that manifests as a result of the action, and if the other person agrees, it does succeed.

Since Jules' player necessarily didn't know what was happening in order to propose a complication, I gave him two options - either take an attack and be able to alert the rest of the party, or focus on being able to avoid the hit. The player chose to alert, so the attack still occurred - luckily Jules succeeded an Armor Roll meaning that it didn't threaten Resolve; he was still in action, and the party was able to notice what was going on before anyone else was threatened. This triggered the next (and first proper) Combat Round.

Round One


Something I want to note about the format of this post is that I naturally default to a sort of bullet list style, but that is primarily for organizing the presentation of the information. In play, it was much more of an organic discussion about potential courses of action. This is important because I didn't want them to think in discrete terms of what their individual character could do within the next 6 seconds or whatever.

First Step: Describe Situation

The party was still in the positions they mentioned earlier, and everyone was now able to see the shadow of a soldier, cast on the wall of the lower stairs, spear outstretched toward Jules and now pulling back for another strike. I answered some clarifying questions, like "If this is a real shadow, where does it look like whatever is casting it is standing?" to which the answer was "Somewhere on the lower stairs," but of course the Shadow is just a shadow, which I kept to myself.

I also described the voices in the room which has previously enticed them into completing the Ritual, saying "Interlopers!" At first this made them think they had angered them by inspecting the statues in the room, but there's actually something else going on. I let them stew in their unease.

Second Step: Plan Actions

"What does everyone do?" Some folks spoke up, declaring what they'd like to do, and then I made sure to ask anyone who didn't what they would be doing. It went something like this:
  • Ingram: Spoke up first - they wanted to run down the stairs and tackle the area where I described a shadow-caster would be, hopefully getting a hit with their warhammer in.
  • Garmayne: Toss his net at the same place? I brought up the potential kerfuffle if Ingram was also rushing that location, to which the player doubled down, I think primarily out of amusement.
  • Sybil: Go down to the middle landing to help defend, while telling the voices "We're just here to help!"
  • Adelaide: Pull Jules away from the Shadow.
  • Jules: Counterattack from a new defensible position.


Third Step: Resolve Actions

Knowing what everyone was intending to do - which I did feel wasn't too much to keep track of, thankfully, I was able to determine a rough order for these actions to take place in.
  1. Ingram: Rushing down the stairs and pouncing at thin air, definitely a Dexterity roll just to: Failure - hitting nothing where they were expecting something solid, they tumble down the rest of the stairs, landing in a heap near the door at the bottom.
  2. Garmayne: Throwing the net, roll Dexterity: Failure - And I can't resist, so, passing through empty space, it lands at the bottom, right on top of Ingram.
  3. Adelaide: Her action seemed obvious and A Good Idea, so I didn't call for a roll there, it just happened; Jules was pulled into a less dangerous place, now not in danger of being immediately attacked again.
  4. Sybil: Likewise, now in position to defend Jules against the Shadow, which would grant advantage to an Armor roll in the case that the Shadow attacks.
  5. Jules: Lunging back with his spear - I keep this to myself, but Strength isn't going to have an effect on something insubstantial, so hitting the Shadows is definitely a Dexterity task: Success - In thrusting his spear toward the shadow, he felt nothing connect with it - but his spear's shadow (cast by the lights both he and Sybil are holding) is also thrust into the body of the Shadow, immediately dissipating it. Cheers from the players.

Round Two

Throughout this, I'm describing and responding to questions about how the shadows seem to be moving and interacting with their light-sources, eventually confirming that they don't seem to be cast from their lightsources, but are somehow moving independently from them.

First Step: Describe Situation

The three other Shadows emerge from the door at the bottom, sliding across the walls and flanking the prone, tangled figure of Ingram. One raises a wicked scimitar to strike at the helpless Ingram; A second scimitar-and-shield and a spear-wielding Shadow make to climb the stairs toward the others. Sybil shields Jules next to Adelaide on the middle landing. Garmayne surveys the situation from the top landing.

The voices say "Shadows from Between... begone!" Ahh... the players think... they aren't talking to them, these creatures are unexpected arrivals from a place they've seen referenced in Tintardinal's writings. They start to think about asking the voices for help, but don't follow up on it.

Second Step: Plan Actions

  • Ingram: Will be struggling to get free of the net; if not successful, they won't be able to dodge the incoming strike.
  • Garmayne: Intrigued by the shadow physics at play, takes out a mirror and will move down to the middle landing where the party's light sources are, to try focusing light at the incoming Shadows.
  • Sybil: Continue shielding Jules
  • Jules: Ready to strike at the advancing shadows
  • Adelaide: Also curious about how darkness would affect the shadows, will attempt to bait one of the Shadows to following her up the stairs and into the pitch dark hall at the top. Perhaps it can't harm them if it can't distinguish itself from the dark?


Third Step: Resolve Actions

  1. Ingram: To get free of the net, roll Dexterity: Fail! The shadow-scimitar comes down and cuts cold and deep; Having a Resolve of 1, Ingram is out of action, and in danger of being killed, if the Shadows wish!
  2. Garmayne: Using Optical Science on the Shadows is a great idea; no roll necessary. However, I judge these Shadows being magical, even focused with the mirror, mundane light is insufficient to damage it. However, it will distract it like a swarm of insects, granting advantage to Jules' attack.
  3. Sybil: No need to roll to continue shielding Jules, granting advantage on an Armor roll in the case that he is attacked.
  4. Jules: Seeing the Shadow distracted, strikes out at the Shadow, roll Dexterity with Advantage: Failed! But while they don't strike true, the combined assault does prevent it from counterattacking, and pushes it back into the corner of the lower landing.
  5. Adelaide: Takes a measured retreat, keeping the other Shadow focused on her as she lures it up the stairs. No roll needed.

Round Three

A lot of Dexterity rolls going on, but it seemed like the best representation for attempts to attack these creatures, given that it needs to be the shadow of party's weapons that need to strike true, and coordinating that with the light sources is definitely not Strength. I suppose I could have granted an Intelligence roll if someone had described some clever way to do that involving a more calculated approach.

First Step: Describe Situation

Having struck down Ingram, the Shadow sheathes its scimitar and begins dragging him through the door! The cornered Shadow remains distracted while Garmayne laser pointers it and Jules harries it with his spear's shadow, and the third continues advancing toward Adelaide up the stairs.

Second Step: Plan Actions

With Ingram under immediate unknown threat as they're dragged away, the party switches up tactics.
  • Adelaide: Wants to go down and retrieve Ingram, so takes a stand against this Shadow instead of luring it further.
  • Garmayne: Since the light doesn't seem to be damaging, tries another tack - will take out a smoke bomb and throw it at the cornered Shadow. Perhaps it will scatter the light of whatever is actually casting this Shadow? Recognizes it's a long shot, but willing to take the risk.
  • Jules & Sybil: Wait to see if they can strike again after it reacts to the smoke.
  • Ingram: Is unconcious


Third Step: Resolve Actions

  1. Garmayne: No need to roll to throw the smoke bomb, it explodes into a plume, obscuring the cornered Shadow from sight... but suddenly a shadow-scimitar comes lashing out! It seems unaffected by the smoke, instead just obscuring it from sight. Roll to Dexterity to dodge out of the way: Failed! Roll Armor: Failed! The wicked blade slices through his gambeson and Garmayne falls to the floor...
  2. Jules: Roll Wisdom to notice something: Success! Somehow, the Shadow is slinking up along the ceiling out of the smoke, you have a chance to react; He drops his torch, so the shadow of his spear is cast on the ceiling as well, roll Dexterity to strike: Success! It evaporates.
  3. Adelaide: There's an opportunity to strike the other Shadow as it ascends toward you, roll Dexterity: Success! The shadow of her longsword slices it in two, the halves curling away into nothingness. I judge that she can also hustle down the stairs to be ready to address the one dragging Ingram.
  4. Sybil: Get in place to defend Adelaide.

The rest of the fight saw Adelaide throwing her caltrops at the Shadow dragging Ingram away, which succeeds in enraging it into fighting them instead, and it was summarily dispatched. After some swift treatment, Ingram and Garmayne were back in shape.

In real life, it took a few more rounds, but that was primarily due to me messing up the attack rolls, having them roll to beat the default 15 instead of the Shadows' AC (11), so I elided a few more failed attacks from this summary.

Playtest Commentary and Further Thoughts on Problem-Solving Combat

As a test of what I want out of this combat system (the AC hiccup aside), I feel like it was a complete success. It produced teamwork, evoked creative thinking about the stuff the PCs had, generated clever tactics, and shifts in strategy as the situation evolved.

D&D-like combat systems can certainly be portrayed with this kind of interesting detail retroactively, and players might even ask to have an off-menu action or interesting plan adjudicated on the fly during combat. But this system lets the players actually engage with it in those concrete details in any way they like, and generate these specific actions and plans as the act of combat itself.

Where D&D-likes abstract the characters' specific in-world actions during a combat rounds, assuming multiple exchanges of blows and hence discouraging GMs from allowing "called shots" despite player tendencies to want to describe specific attacks, Problem-Solving Combat is all called shots; the player needs an approach that works in their present context.

This also has the benefit of bringing all of the different Abilities into play where it makes sense. One of the players was wracking their brain to try to bringing their very high Strength to bear, and I didn't even need to tell them a direct brute-force approach wasn't much use against these foes. Not one Strength roll was made. A fight with corporeal foes would be completely different.

Some newer OSR systems emphasize speed of their combat rules, so that more of the game time is spent doing interesting problem solving outside of combat. Instead, this system is an attempt to make combat as interesting a problem to solve as the rest of an adventure.  It definitely takes longer to resolve a round of combat, but I would argue the time spent in combat is far more fulfilling in this regime. And it could take fewer rounds overall - particularly with Cunning Knave's tighter health and damage rules, and the advice that each round should create a significant shift in the combat situation. If the Shadows didn't have max Morale, I would probably have had them retreat after the second loss.

And this isn't intended to encourage combat to happen; it's still very deadly, which should discourage unnecessary combat as much as any other OSR system. If Jules had failed his Armor Roll, the second roll of the session, he would have been out of action for the entire fight. And if the Shadows' description didn't mention they abducted victims, Ingram and possibly Garmayne would be dead (though soon replaced with a new PC, of course).

It certainly requires more GM brain-simulation, free-form orchestration, and arbitration than D&D-like combat systems. The tide of battle and lives of the PCs is as much in the hands of the GM as it is in the dice's. There were some instances (elided from the example) where I could have judged that a Shadow struck back as a result of a failed attempt to attack them, which I didn't take. So the GM has a significant amount of control if they wish to tweak how potentially deadly an encounter is - but it's still not just fiat - a die roll is involved in anything that threatens life, so that decision is still disclaimed, with all the benefits of impartial arbitration.

Furthermore, I might argue that the work the GM puts into this style of combat isn't much to manage than the crunchier iterations of D&D-like combat, since it's utilizing common sense rather than a whole set of particular rules and their exceptions.

I will also say that this experiment utilized a very rules-light framework, and there  is probably more work to be done to create as satisfying a regime within, say, BX, or anything heavier. Here I will mention a post at the Spells and Steel blog which presents some good house rules for at least the initiative/turn order side of things. On a related note, Group Initiative was brought up as a remedy to issues I expressed in my prior post, which definitely helps with some of the planning-related issues. I used it in my QBX game to decent effect. But it's not the silver bullet for the more endemic issues that I hope I was able to elucidate here.

Anyhow, again:

This is what I've found works for me, to evoke the kind of engagement I want with and from my players, compared to how I've seen and played other game systems run by other GMs.  If it sounds good to you too, cool! See how you can integrate it into your games, and look forward to the indeterminate future when I release some public form of Cunning Knave.

I encourage you to comment below, rather than elsewhere.

Long live the Blogosphere!


  1. I'm quite interested in this idea, but it strikes me that a lot of the tactical depth in this example springs from the unique qualities of the shadow puppets.

    Have you tried this system with more mundane enemies? If the enemies were orcs instead of shadowmen, then a lot of the unusual tactics your players employed wouldn't apply.

    1. I agree they are pretty unique opponents; I didn't really need to work to identify the interesting challenges for this encounter. As mentioned, I intend to follow this up with advice on how more mundane situations can be spiced up, but in case that never comes, there is really no shortage of content on that topic out there already - you can just ignore any rules-related bits they adhere to and see it more holistically.

    2. And of course, if there's nothing to present an interesting challenge from a given opponent or situation, you can always shift it to a different type of encounter - a social challenge (goblins bickering among themselves where you might be able to steer their conversation into infighting), more of an obstacle (an obviously strong orc blockade, guards just telling you to fuck off), etc.

  2. This is a really cool idea. I've had similar feelings about making combat more like the dynamic problem solving space of normal adventuring but have never been sure how to develop it. I really like what you have shared here, but I have some questions about he particulars.

    How do you differentiate weapons? Normally in Knave, there is a trade-off between higher damage dice and the number of slots and/or hands they require. But if successful attacks mostly do 1 resolve in damage, that doesn't work.

    And in your examples you describe characters both making dexterity checks to try to avoid attacks, and armor checks if they don't. Normally in Knave, armor is the all-in-one damage avoidance measure. Is this how you differentiate weapons, in place of a damage roll making an armor check against the damage potential of the weapon?

    I'm definitely interested to see more details about your particular rules hack.

    1. Thanks for the interest! These are good questions. Briefly:

      Weapons are only differentiated fictionally. Damage is only in amount of Resolve, and most normal physical attacks will only deal "1 damage". If the right weapon for the job is used, or some plan would obviously cause more impact, the referee should increase the damage dealt.

      I'm currently deciding how to handle Armor item values, but I did just recently add this to the
      combat section, which may still change:

      Note that Rolling Armor should be taken literally. As long as there is a reasonable chance within their plan, other contingency rolls could come into play before a PC is actually struck. As a common example, dodging with a Dexterity roll, if they were free and able to anticipate the attack.

      Happy to continue discussion on this if you have feedback!

    2. (Though larger weapons should still take up more Item Slots, and you can only wield held items that makes sense at a time.)

  3. I like this. This is something that's been bubbling around in my headspace, but I've yet to sit down and figure out how it would work at table. Already I've switched to framing my combats as 'here's the situation, what do you all do' instead of 'your turn, now their turn, now MY turn.'

    This is really helpful, thanks for the detailed example. I am a little worried about speed and keeping track of things. While my PbtA combats in the past have gone swimmingly, in other games I've played that used simultaneous combat systems I've had trouble keeping track of who's attacking who and how and when.

    Probably, the more rules-lite the game, the easier this is. But I've yet to find a simple game that works for me, so I feel like my combat satisfaction is at the whims of whatever game I'm playing at the time.

    1. I feel that, I'm definitely relying on the simplicity of Knave (and even simpler with the HP>Resolve and a few other bits) in order to handle running the game overall, but I think it's worth being able to shift that effort where it feels more interesting and salient to have a deeper dimension of detail.

    2. And I don't think I mind if combat takes longer than in other systems, because the quality of that time is very different. Here it's player ingenuity all the way down, and that's the good stuff. And still, it can be uncommon that you end up in a necessary fight - this one actually was primarily due to the nature of the opponents.

  4. Awesome stuff! I really like it and have been thinking abou something similar.
    How do you plan to deal with hirelings? I see a lot of potential of integrating them into maneuvers ("Bib and Bob give me a boost so I can jump behind the phalanx", etc) but if they all get individual actions it might slow combat down at the later stages.

    1. Yeah, good question and concern. I like your suggestion though; they can be "used" by the PCs just like they use weapons.

  5. Thank you for this. This example crystallised in my head an idea that's been kicking around, which is that combat is (or should be) a continuation of OSR problem-solving, and that each round is a discrete challenge presented to the PCs. Up till now, I'd been fiddling with mechanical hacks for combat, rather than systematising my approach to it as a puzzle. This is what makes initiative-less combat, imo.

    It inspired me to type up some procedural ideas about the elements of a combat puzzle.

    1. Nice! I skimmed this post and it seems really good, but I’m going to stop myself from reading too deeply to keep my own thoughts on the idea intact for the third post.

  6. This is great!

    It’s also exactly how combat plays out in games like Fate Accelerated, and its something that i miss VERY MUCH in my OSR games. This includes the fact that weapons are largely irrelevant (concerning the damage they do).

    Like so many things in B/X and its derivatives (I’d include Knave in that) combat is one of the many sub-systems, or micro-games, that make up the larger game. Yet we rarely consider swapping it out the way we do the crummy Thief skills for a 2d6 table, or LoTFP’s skill pips, or replacing encumbrance rules with an item slot system.

    By treating combat as a series of potential skill checks, the way you do any other situation, puzzle, or trap, it really opens up the way That Players can approach it. And when the focus is taken off of “roll to-hit so that I can inflict HP damage” I think you start to see more creative play: creative maneuvers to give a comrade advantage on a roll, screaming insults as a way of reducing an enemy’s Resolve, etc.

    On weapons: perhaps give them tags to differentiate them. Daggers are “Good in tight spots”, hammers and maces “crush bones”, etc. possibly giving a mechanical boost under specific conditions, or giving the wielded more options -in confined quarters your magic dagger gives you a boost to intimidate foes, rather than a bonus to hit.

    I really loved this!


Be kind, approach disagreements with curiosity, enrich the conversation!