Mass combat and whole-dungeon thinking

Mr Mike Shea , perhaps better known as Sly Flourish, has a wonderful blog and, indeed, a wonderful pen name. SF follows the philosophy of "The Lazy Dungeon Master". This philosophy can be put into a simple equation: If the quality of your game = the quality of your prep x the time spent prepping, you can up the quality of your game by upping the quality of your prep, here meaning its efficiency. If you can work out how to condense four hours of planning into just one hour by focussing on the right prep, then you can instead do two hours of prep, thus earning two hours of quality time to yourself and magically doubling the quality of your game into the bargain.

I explain this only because I think more people ought to know about Mr Flourish and his insights. The actual reason I bring up Sly and the way of the lazy DM is because of a recent post on his blog about running big battles in D&D. This seems to be an eternal question - there certainly seems to be at least one thread about it on the D&D Reddit-sphere every time I gaze into that particular abyss. And no-one seems to have found a solution, at least not a satisfactory one. The most common answer seems to be "Don't. At least, don't mechanise the battle. Instead, prep a series of encounters for the players that showcase different parts of the battle." And this is the advice that Sly "Mike" Flourish offers, quoting the Explorer's Guide to Wildemount. But the interesting thing is that I think Sly and the quote are actually slightly at cross-purposes. And I think explaining how can help explain why this question is so intractable, and how we should approach it to design a system - yes, a mechanical system - to make battles/wars doable in D&D (or a favoured rpg of your choice).

🕸

Here's the quote:

Since D&D is primarily a game about a small group of characters going on adventures alone, it can be difficult to simulate massive battles using D&D combat rules. Because of this, it's generally best to keep the characters away from mass battles. However, huge conflicts with thousands of combatants are a cornerstone of epic fantasy, and your players might be disappointed if your war campaign doesn't have at least one climactic battle.

To solve this problem, you can break down your mass combat into manageable chunks. Find a significant location that the characters can either defend or conquer with minimal reinforcements, like an overrun citadel. Then,
have the major battle proceed in waves that guide the characters from one cinematic encounter to another. You can think of these encounters like rooms in a dungeon; some rooms have multiple doors that the characters can choose from, while others only have a single passage.

Now, SF agrees with this, saying to keep the spotlight on the characters' role in the battle. But, where the Explorer's Guide says to guide players from one encounter to the next, SF says to let them be self-directed, and choose how they want to help out with the war effort. He talks about characters being in the war room, the place where battle strategy is made. His version of involving the characters in a battle/war (he does admittedly seem to hop back and forth) seems to be to involve them in the strategy of the war, deciding how it is prosecuted, rather than dragging them from scene to scene with the occasional choose-your-own-adventure style fork in the road.

It might be obvious from my language that I think Sly's is the better of the two approaches. But it's worth looking at why. It's not just that choice = good, so more choice = better. That's part of it, but that's only the surface level. The real point, the real meat of the matter, is about what makes this problem so tricky in the first place - making a battle feel like a battle, or a war feel like a war.

🕸

When we want to create a new mode of gameplay, or a new level on which gameplay functions (either could aptly describe modding massed battles into an rpg not designed for them), we want to try to capture the feel of a particular genre or mood in fiction, or of how we imagine it feels to engage with a particular environment or set of actions. Usually GMs, being game designers by nature, will look to incorporate this into the game by creating a rules framework - that's how different modes are usually accommodated into a game, so it makes sense that that's how that would proceed. For instance, in D&D, you have one game mode - combat - that mechanically dominates almost everything in the game. But there are other rules frameworks to capture other styles of scene, such as rules for running chases, carousing and, in older editions, fleshed-out rules for wilderness/dungeon expeditions, realm management and so on.

But it's important to think about the criteria for success for systems that create a particular game mode. Those criteria can largely be summed up as the following: "Does playing with the rules for x capture the feeling of doing x?" D&D's combat system, depending on how it is implemented at the table, captures the feeling of swinging at your enemy, having the blow connect well or glancing away, doing little damage, getting in a lucky hit etc. Compare Tunnels & Trolls' combat, which is abstract - it simulates a melee about as well as D&D (well, probably), but rolling dice and counting up for your side doesn't have the visceral feel of the to-hit and damage rolls of D&D. And this is why the spell breaks when D&D combat is overly fiddly to run, when the pace drops etc. - because it no longer feels like a rough, tense, gritty fight. By the same logic, dungeon-exploration rules that emphasize the dwindling of resources as characters push deeper into the unknown are successful, and serve to create good gameplay, while 5e's carousing rules are just a bit rubbish, because, while they may effectively simulate the results of a night of drunken revelry, they don't evoke the feeling of revelling drunkenly.

Even though they don't represent particular systems or sets of rules, you can see the alternate approaches of SF and the Explorer's Guide in this light, and why the former has a better idea of what makes a good battle. Sly's approach of handing the characters information about the situation and letting them form their own plans to help out feels more like helping to win a battle. Or rather, it feels more like there's a battle going on, and the characters are engaging with it, because they get an open-ended choice about how to affect the overall situation. On the alternative, the characters are just doing what they would be doing anyway, namely following threads between predefined scenes - their quest-giver just happens to be a general or something.

Kill a messenger? Sure. Sabotage a supply line? You got it. Mount a contrived defence of a position against oncoming hordes? Just us four? No other soldiers? Well, okay then. Was there meant to be a battle going on around here somewhere, or are we just running errands?

On the Explorer's Guide approach, the battle sits in the background as an excuse to set up scenes. It's not something with which the characters have to engage - they have their specific tasks to be getting on with - so it may as well not exist. Hence the feel of being part of a battle, of pulling your weight for the team, of being integral to the success of the war effort, falls flat. It's like if the GM told you that the Joker blew up some NPCs you never met, and is therefore a really bad dude, as opposed to your having to race to save Harvey Dent and Rachel Surname, or choose between them. Even if you actually manage to save both of them, the Joker's villainy will still strike home much more effectively; you've had to engage with his bad-dudery, planning around it and combatting it directly, so you'll form a more compelling relationship to him.

Analogously, an approach to large-scale battles where the players take planning into their own hands has them actually engage with the battle, thus making it feel like there's an actual battle going on. The massed ranks and huge charges will still largely be in the background, but it's a background that serves to frame the action, not just to provide an excuse for it. It's an "active backdrop"; it provides context, but context for action, not the sort of toothless context provided by random lore or exposition dumps.

🕸

I've glossed over the distinction between playing a war and playing a battle. The same thinking as above applies when playing a war, but in much broader strokes. The concept of a campaign, after all, comes from the campaign in wargaming, meaning the whole war as planned by the generals and pursued by an army. The campaign, the state of the war, was what dictated which battles you would play out on the tabletop. The roleplaying meaning has broadened to mean the overarching story (where it doesn't just mean the group of players or their characters).

But it's worth looking at that old meaning, if only because it speaks to the importance of control over the PCs' destinies; in a wargaming campaign, you have the option of how to commit your forces, how or whether to bring the enemy to battle, which territory to leave and which to contest to the last man, and so on. Because you play the army, you have control over the army. Now, the players won't always have control of their own army. But if you strip that control back completely to a choice between pre-defined scenes, or even an outright railroad, the game has drifted away from being about a war at all - it's now about a laundry list, with a war as a token motivation to go from scene A to scene B.

Don't pull a muscle striking a pose there milady

The dungeon analogy from the Explorer's Guide is also illuminating. It's true, you can get a lot of mileage out of thinking of an adventure - any adventure - as a dungeon. That's a rather conceptual insight I didn't expect from Wizards of the Coast, who, in my rather crabby and jaded opinion, often don't seem to give a lot of thought to how games run at the table, or how to train GMs to run them. Despite being a good bit of advice, however, it's still problematic. See, thinking about a scene or an adventure or a non-dungeon location as a dungeon should mean plotting out a sort of node-map of the different potential points characters can visit (/scenes they can trigger/crime scenes they can investigate etc.), and, of course, how to get between them. This is an especially good way, for instance, to plot an investigation, where the links that convey characters from one scene to another will be clues. What it should not mean is what it seems to be being used for in the quote above, which is building a railroad with the occasional optional scene.

While a dungeon layout makes a good heuristic for planning, the individual choices (do you go left or right?) aren't what makes it useful. As above, and as in an actual dungeon, the choice of whether to take the left path or the right path isn't what's interesting. What's interesting is interacting with the environment as a whole. The choice of which path to take is only interesting in the context of a plan to get to a certain place, where the whole layout is known. And at that point it becomes pretty much a non-choice, at least in the moment. The Legend of Zelda dungeons are, at their best, good examples of this principle. That is, they require thinking about the whole space, modifying it, and working out paths through it several steps ahead of actually coming to a fork.

So thinking about adventures, including battles or wars, in dungeon-like terms is good. We've just got to remember that what makes dungeons tick, and therefore what makes these adventures tick, is interacting with and planning for the whole space - whole-dungeon thinking, not room-to-room thinking.

That definitely needs a post of its own. Filing that away for later. Anyway.

🕸

There's one more aspect of this discussion I've been ignoring. I'll be brief, since I've already rambled on for long enough. Why dispense with rules? And why would we want rules for running battles in the first place?

The second question's all but answered already: We might want a framework of rules because that would give the players a handle on the whole situation, and therefore the ability to effectively plan and act within it. It's easier to do this sort of whole-dungeon (whole-battle?) thinking when we have a solid impression of how the moving parts work. We can't be sure if we're making good decisions (or if our decisions have any impact at all) if we have to settle for "Well, our soldiers there are pretty tough, but there's a chance they'll be overrun. We make it less likely if we bolster their fortification though." Should you bolster the fortifications? How urgent is it - should you prioritise it over other tasks? It's very difficult to tell, and therefore very difficult to weigh up plans involving bolstering the fortifications, or not.

I suspect that many GMs who grapple with this whole problem start by coming up with systems because they understand the importance of whole-dungeon thinking, and of giving players something to engage with that enables it. Most systems in a game serve to stand in for understanding that the characters have but the players lack (this isn't the only thing they do, but it's one of them). I can't see the wall I want to climb to assess whether I stand a chance of making it to the top, but if my character has +9 to climbing checks, and I know the target number I need, I can assess my chances, and factor those into plans involving the wall. Similarly, the players can accurately assess the likelihood of an enemy force overrunning their fortifications, bolstered and un-bolstered, if we put numbers on them, and put them into a system, probably involving dice.

Why would we not want to do this? Simply put, because then we're playing a different game. Now, I personally would be fine to play a Warhammer campaign interlinked with, for instance, a D&D campaign. But I appreciate that we'd not really be doing the same game. Large parts of the gameplay would focus on unnamed soldiers, rather than the individual heroes we signed up to spend time with. And it's even worse if the heroes don't have complete control over the army. Assuming that they're involved in the planning process and generally commanding small groups on the frontline (think the Three Hunters at Helm's Deep), the GM's basically going to be playing most of a wargame themselves in between the characters getting to step into the action occasionally (there are a few good "RPG horror stories" on Reddit lamenting such games). On top of that, having the players learn a whole wargaming-esque system that they then won't be using directly is a hefty price for whole-dungeon thinking.

The Explorer's Guide solution is to strip out the rules, and give up on the whole-dungeon thinking with it. Sly McFlourish modulates this to trying to have whole-dungeon thinking, but still without a rules framework. My suggestion: Use a simple framework.

🕸

Here's a freebie.

Risk is a brilliant game. It has a conflict-resolution system that is extremely versatile, simple, and actually doesn't involve any maths beyond seeing which number on a D6 is higher than another. Basically it involves pairing D6s, representing attacking and defending units, off against one another. The defender wins ties, but only rolls up to two dice; the attacker rolls up to three. The rest of the game is logistics, but this remarkably simple core sits under it all. It takes almost no time to resolve, and is easy and intuitive to learn. I propose using this to manage combat.

Here's what you do. Map out your battlefield into several segments, and chart how you can move between them. Place a number of units to represent the opposing forces. Multiple units can be in the same location. A unit represents the fighting strength of one squad of x number of humans - you decide how many. So a unit might represent 100 men, while another might represent one dragon, or five giants, or what have you. The conversion rate here needn't be especially scientific.

Allow the players to see all of this, and let them know exactly how the resolution system works (I'll let you look it up if you don't know already). It's simple enough that they shouldn't feel overfaced. If you're worried, contrive a low-stakes engagement as a practice to help get their eye in - this is about building systemic intuition after all. Then hand over the reigns, as befitting their station as quasi-autonomous heroes. Don't make them play out whole engagements in the system, but let them direct their forces. About one Risk roll per round of D&D combat (adjust for your system of choice) should be right. To run a whole war, just zoom out and use bigger numbers and distances.

Try to make clear to the players that they can go off the grid - the game hasn't shifted gears to playing Risk. They can lead charges, granting bonuses to their troops. +1 to the highest die rolled is a good one for this, as is letting them roll extra dice, or having destroyed units count for double. Alternatively, they can devise and carry out their own missions for specific advantages, which don't have to be primarily systematic. At that point it's up to you to adjudicate the effects of their plans, but with a robust system under your belt, this should be easy enough. Just remember not to marginalise or play down the contribution of these missions. Again, the new system hasn't supplanted the main game; there's no sense in which the characters should go through it to achieve their goals, or else not be playing correctly. It's just a way to facilitate their engagement with the battle, and thus to make it feel like a battle, without necessarily having to foreground it too much. It's an active backdrop.

Good. Now I have something to link to when I inevitably see this crop up again.