The Game Master and the game as servant

I’ve been having an interesting time with Draw Steel recently.

At the start of 2023, Dungeons & Dragons’ new Open Gaming License was released, and people got upset. The history of the OGL and what it has meant for RPGs as a scene is a very long one to enumerate here, and is honestly centered more on the d20 boom era than it is now, but, sufficeth to say – at the time, there was a lot of badwill towards WotC that emerged in the massive D&D fan scene, and an interest in moving to new RPGs. This represented a significant audience to capitalize on. Not an audience for indie RPGs per se, as some people optimistically predicted. But, rather, a group of people with a lot of interest in a game like D&D, with enough serial numbers filed off and the money not ending up back in WotC’s corporate pockets. Which, fair enough, honestly. I’m an advocate of ‘alternative acquisition methods’ in circumstances like that, but whatever works.

In response to this news, many projects were announced offering just that. The D&D 5e experience, but without the same evil corporation behind it, and perhaps with design decisions you favor. At the time, I called these efforts ‘Paizo hopefuls’, since this was the path to success that Pathfinder rode back in the reactionary response to D&D 4e (and, interestingly enough, another effort to tighten up the OGL at the time). I predicted, somewhat pessimistically, that these were mostly misreads of the situation – there wasn’t the same window of opportunity that Pathfinder had here, because everyone already knew the trick and multiple people were jockeying for the same position.

I would say, in terms of how many names were bandied around at the time that have fallen relatively silent since, I was somewhat right. But I definitely underestimated the impact of having a prior significant audience base to sell to. Both Critical Role and MCDM have made their own Paizo hopeful successes, selling a mechanical framework that promises to offer the D&D experience, and those games have seen relatively solid success. Respectively, those are Daggerheart and Draw Steel.

All that to say that a lot of my complaints regarding Draw Steel can be traced back to complaints I have with the gameplay of D&D, and in retrospect it is very predictable that I ended up with rather a lot of them. And, similarly, most answers to the question “why was this designed this way?” will end up tracing back to an element of D&D 5e, either in gameplay format or in play culture.

Draw Steel in particular takes a lot from 4e as well. Its combat system is well-elaborated, with clearly defined powers and movement on a grid and such like. It’s a pretty good combat system, as tactical games go. That’s what drew me to it, really, and on that front it hasn’t disappointed much. In terms of games I would recommend looking at just to get a taste of the combat, it’s on that list!

Alongside the combat formula, it takes the overarching inter-fight attrition format from 4e. Everyone comes with a set number of heals, called Recoveries for this game, that get spent when they heal between fights or are targeted with a healing power. Once they run out, they can’t do any more, until they take a proper rest. There is a notional push-your-luck system at play, too – the more fights you play in sequence, the stronger you are at the start of each fight. The idea there is that, once you run out of heals, you might still want to push through a fight or two, relying on alpha strikes to carry you a bit further before you have to rest.

That sort of setup works best when resting is a difficult decision. And the game does suggest that the GM impose a punishment on rests, in the vein of “time has advanced so the enemies are now doing a new bad thing,” so the attrition can come up at all. (Otherwise, it would be best to just rest after every single fight, and it would be nigh-impossible to blow your full heal budget in one fight. That’s a hazard of letting players set the pace for these things.) This gives some variance in how many fights can happen between each rest, but the expected amount is (to my understanding) roughly 4-6 Victories’ worth. Difficult fights give two Victories, and sometimes the GM might give them out for accomplishments that don’t take a fight at all, so that isn’t as much as it seems. If anything, it’s comparable to or only slightly longer than Lancer‘s 3-4 fights between each rest, and the fights aren’t as tough as Lancer‘s unless they’re the notably difficult ones. So, it works out as a pace.

The standard progression pace goes from levels 1 to 10, and costs you 15 Victories to earn a new level. (Victories were secretly just the XP mechanic, which makes sense.) If you follow the standard pace of rests, and don’t push your luck to the point of getting in the 7-8 range before resting (which is doable!) you end up hitting 3 rests each level, or 30 over the course of a full 1-10 campaign.

…Which is a pretty useless number in isolation. Neat, I suppose? Why am I bringing this up?

Well, it turns out, rests aren’t just resets. They’re long-term progress. (This is, of course, another finger on the scale towards incentivizing resting after every fight. Which would remove any combat tension. The GM will have to work mighty hard to ensure that’s properly disincentivized.) During each rest, you can work on a long-term downtime project, and get something from that! This covers crafting, training, building something useful in the world, getting some buffs that last until the next time you rest, and a few other options. This isn’t unheard-of by a long shot. The aforementioned Lancer does something much the same. This is just a bit longer-term, since campaigns are gonna be a lot longer. Plus, the mention of crafting caught my eye in particular. As with D&D, this game has magic items as a significant source of power, as well as build differentiation. In D&D proper, those are essentially handed out by the GM at their leisure. Having a player-facing crafting system gives me as a player the chance to work towards the things I specifically want for my build, rather than relying on fiat! That’s quite appreciated.

There are a few factors that determine how much progress you get on a given project each rest. The base roll is 2d10 plus a relevant stat, with stats ranging from -1 to +5. (It is technically possible to break this cap at the very endgame, reaching +6 in one stat.) At the start of the game, your highest stat will be +2. If you have a skill you can apply, subject to GM approval, you get another +2. If you don’t have the relevant language, you get a -4 penalty, and for some projects, you get a -2 even if you do have the relevant language. At level 1, in optimal conditions, you get 2d10 + 4, and at level 10 in optimal conditions you get 2d10 + 7. If you don’t have the language, something relatively common to encounter, you get 2d10 + 3 at level 10 in otherwise optimal conditions. Often, you can’t roll your best stat, or you might not have a skill that applies. Also, there’s a crit chance, but the chance is 3/100, which is depressingly rare, and that means over the course of 30 rests in a game, your expected value doesn’t even hit seeing it once.

That’s a lot to juggle, but the numbers are pretty tightly bound. Since a full statistical analysis here would be pretty overkill and also include a lot of assumptions anyway, I’ll just go with – if you take 2d10 + 4 as a good benchmark for the roll people expect as they try projects they’re better and worse at, you get around 15 points per rest. That’s a good number to work with. Over the course of a campaign’s 30 rests, that gets us a cool 450 points. Nice!

How much can that buy us?

…Well, certainly not that. This is the first downtime project in the book, and it sets a pretty stark impression in terms of numbers. If people forgo their own progression to contribute to someone else’s project, something they can do (but that makes it more likely they end up having to roll worse numbers, of course), then it would take the entire budget of 7 PCs working all campaign on this and nothing else to get it done, right near the end of the campaign. Frankly speaking, I don’t think running this game for 7 people would be a good idea! A lot of things would break first. So, that’s essentially unobtainable, per this setup.

They aren’t all so bad. This one is unobtainable without help for the new ability or the discount, but you can get a damage buff to 3 different abilities over the course of a campaign and still have 90 points to spare (which can’t get you much, but can get you some miscellaneous narrative benefits). It’ll take you 8 rests to get to around 120 points, so you should get the first buff in just before the final push of level 3.

You can get a handful of consumables for only 3 rests! That’s just one level. Though, they do get more costly as they hit higher-tier consumables – the beefier ones are 180 points for a set, or more.

Most importantly, here’s the big one. A magic item. You have slots for three of these things, and they scale your damage or health quite significantly for the first two. And they cost… 450 points. Your entire budget. Sneaking in at the last rest of the campaign, after everything is said and done. You would need three times that to fill up your slots properly – and then there are lesser magic items you might want to craft, starting at 150 points and going from there.

This economy is in shambles. What on earth is going on here?

Everything is made up and the points don’t matter

There are ways to punch up your numbers.

This is pretty notable! You spend one rest recruiting a guy, and for the rest of the game, they roll on a project you set for them. Not as well as you, if the stars align with their skill, language, and relevant stat, they only roll 2d10 + 3, but you probably pick them for a specific reason, so let’s be generous and still give them the 15 benchmark per rest. One rest spent at the end of level 3 gives you 21 more, another at the end of level 6 gives you 12 more, another at the end of level 9 gives you 3 more. 33 net gained rests – that more than doubles our output! Which means, by the very end of the game, you can hit 2 out of 3 magic items, and you can get your first one by… two-thirds of the way through level 7, at idealized pace. (And that has your first two guys optimized for the first project, which means you might need to spend an action to get new ones for the second one.) So, still near the end of the campaign. And, followers aren’t necessarily forced to be crafters. You can take them as an additional fighter in combat, and that’s pretty dang significant.

That all assumes, of course, the GM decides to hand out Renown consistently. As it says, “in most campaigns, the Directer sets the characters up to earn 1 Renown per level.” If the GM doesn’t do that? If they decide a given opportunity for glory was squandered, or even they just forget to? That’s a rough reduction, and even if this is followed to the letter, with the GM not exercising any of the power the game is offering to them, you go through most of the game without getting anywhere if you aim for the meaningful investment.

What else can you do? Well, the GM can just hand out bonus points as a loot reward, and some of the backgrounds give a starting budget of bonus points. They get these instead of a language, so it does reduce the numbers they’ll be able to roll during project rolls to compensate – each one is a -4 per applicable roll. I believe the lump sum works out better than taking the language, for these numbers, but it’s something.

And finally…

Turns out these numbers were totally off all along. Somehow, you’re expected 10-20 rests per level. Despite only fitting 3 into the number of fights necessary for an adventure. So, all that math was entirely wasted, and, if we average 10-20 to 15 per level, that’s 2,250 points over a campaign, before considering the Renown bonuses. Much more manageable. So, case closed, right?

Well, if you’re the GM in the game I’m in, no, case very much not closed. Specifically, with the reaction of “I would absolutely not expect people to just toss a bunch of rests in a row at the end of a level to meet quota, that sounds like exactly the sort of thing I was supposed to be nipping in the bud with narrative threats.”

And, it does say even in the snippet there. It’s the Director’s choice how that pacing works out. It’s their choice how many rests to allow before throwing a new problem at you to make you go out and deal with it again. It’s their choice if they give you bonus points, or Renown, or whatever else you need to make pace. The only amount the system guarantees you is what you can earn from the rests you need to do during an adventure.

And it doesn’t even do that, really. See, you can get the points – but actually working on a project isn’t that simple.

Here’s an item I’m planning on getting for my PC in that game. It’s a major magic item, its goal is 450 points, as expected. It’s got some keywords, it’s got a name, it has the stats you can use to roll on it. And it also has two other lines. A source and a prerequisite set of items.

You can’t start a project until you acquire those.

How do you do that? Ask the GM.

Is steel probably around? Yeah, I suppose. Crystals that resonate with psionic power? How rare are those? How obscure are the texts describing the item?

Ask the GM.

If you’re the GM – it’s entirely up to you. Want to hand them out for free, because they can probably find them? Want to make them bargain for it? Want to send them on a whole dungeon delve to get them? The power is in your hands, and the guidance is just “whatever you feel like.”

(And, of course – unless the players proactively tell you which specific items they’re interested in, you have to consider giving this out for every item on the list. And there’s a lot of them!)

At the start of the game, it’s plausible to not have the requirements for any of the long-term progression options. And if you don’t, you’re stuck investing in some miscellaneous temporary buffs or narrative bonuses, and losing progress that long-term is pretty tight unless the GM actively chooses to loosen it for you.

And then, halfway through the project, you can get hit with this:

This is a bit of tech I really like from The Treacherous Turn. Completion stops. At certain points in a progress bar, you hit a roadblock, and have to go acquire a piece of data to resolve it. In TTT, it fits the bill of being a rogue AI that thinks weirdly and needs esoteric data to perfectly predict and control the world. It also helps that you have a lot of progress bars to work on at a time, and the resource-management of those projects is the gameplay. Here, events are rolled on a table, and sometimes they’re free rolls (ie a small bonus to progress), but sometimes they serve as a completion stop proper. A thief came in and stole the project source – go on an adventure to hunt them down, and you can’t continue until you do. The GM is free to choose whatever effect they want, to – if they want to stop your project, they can.

In other words, the GM controls how fast the projects go, if you’re allowed to start them, if you’re allowed to continue them – the only thing that’s not explicitly under their control is what options exist to be conceptualized as projects in the first place. And that’s easily done implicitly – just don’t hand out the requirement in the first place.

If you take this subsystem at its base, without the GM handing anything to you, you can’t get anywhere meaningful. 450 points all campaign, and who knows when you’ll be allowed to even start. If the GM does hand anything to you, they have a lot of control to define however much they’re handing to you, when and how you can get anything out of it, and what you’re even allowed to get in the first place.

This subsystem has circled to right back where it started. If the GM feels like it, they give you the item you want as loot. If they don’t, they don’t. You have no control.

Which brings me to this quote from myself earlier in this very post. This was my initial assessment of the subsystem – excitement and relief, that I had a system to rely on instead of GM fiat. I only realized I’d been deceived once I saw it in action and ran the numbers myself.

Who calls the shots?

I like rules.

I like them for a lot of reasons. But, as a player, I like rules as a source of power. Rules are tangible, solid, and invocable. I can point at the text and say “this happens, because of that,” and be demonstrably either right or wrong. That’s a consistency I don’t get in real life – and it’s extra valuable in what is ultimately a social situation containing people I may or may not trust or know all that well. When I’m a player, and there’s a GM with the authority to declare whatever they want and veto whatever I propose, having a rulebook that I can cite to push their hand back to what is the correct structure of how things are supposed to work is nice. It means I’m not entirely powerless, you see what I mean?

That’s one of the reasons I enjoy tactical games. And, in its combat, Draw Steel is solid for that. Its rules are clear both playerside and enemyside, and I can respond precisely to everything the GM offers – including correcting them. It’s a solidly structured ruleset that defines how I can engage with its tools, and gives answers that the GM is at least expected to innately accept.

The crafting system, despite looking like that on first blush, is not that. It is designed on every level to ensure the GM maintains veto power, in a variety of forms – and, in fact, the default form is a veto, due to the pacing. The numbers simply do not work, unless the GM willingly allows for and opts into providing bonuses of one form or another.

That’s the core of the above rant. I expected the crafting system to be meaningfully structured and codified in a way that I could use without augmentation, and the GM could rely on leaving it to the players to do the same. The numbers do not work out like that, and that means I am unsatisfied. It does not do what I wanted or expected it to do.

…What about the fights, though?

Sure, I can say the GM is expected to just accept what the rules say for them. But that’s within the context of the fray. When they’re setting up a fight, what stops them from making a map with a pit the melee guys can’t cross over and enemies that sky kite and launch attacks from above? For that matter, what stops them from ignoring the encounter budgeting rules and just tossing so many enemies at us that we instalose?

…Not the actual encounter building rules, it turns out! The budgets are fairly sizable bands to work within, and the GM is fully suggested to change the numbers as they feel like to fit the result they want. This, of course, isn’t a unique problem here – Lancer does the same thing, as do many other tactical games. I’d go as far as to say that the current standard for games like this is to make encounter budgeting only a guideline and mapmaking even less than that. We’re still relatively new to mechanizing fight objectives as a consistent thing. I’ve talked about this on the blog before. It isn’t really news.

But, where does that leave the ‘answers the GM is expected to accept’? In the fight rules only? Pretty much, even the skill rolls are GM call on stat, validity of a given skill, and the meaning of a success or a consequence. The power of interpretation means there’s almost nowhere to go in Draw Steel where the GM isn’t calling the shots. It’s just the fights.

An acquaintance posited a hierarchy of authority for RPGs a while ago, which I think is useful to keep in mind here:

The social agreement to keep playing
The GM
The rules of the game
The players

Essentially, something below on the ladder can’t do anything to violate the demands of those above it. The agreement to sit down and play the game was more relevant to the conversation as it came up – essentially, if the GM ever does something that prompts people to no longer be cool with playing, or that would tangibly present further play being possible without this being an accepted-upon endpoint, they face retaliation. In any other case, however, they’re generally afforded to do whatever they want. In particular, they’re above the rules, but the other players aren’t.

This is why a GM changing dice rolls is “fudging” and a complex conversation, but another player changing dice rolls is “cheating” and there’s nothing controversial about condemning it.

In essence, it’s a pervasive principle. As long as it’s true, there’s nothing any rulebook could do to counteract it, right? If the rules say the GM can’t do this, the GM does it anyway and the cultural expectations are on their side. You can say that’s improper play, and disclaim it. As a game designer, I’m happy to leave it at that. But, unless you’re entrenched in your ways to the point that the GM deciding to cheat prompts an active protest to the point that it violates the top of the hierarchy, that’s not really going to mean much in play.

The thing I’m curious about, then – why was this a surprise to me? Why was it a letdown to have the GM calling every shot in this subsystem, when I can ignore how it muddles the combat mode much the same?

I’ve come to three conclusions – two specific, one general. I think they’re all compounding factors that led to this reaction in my thinking. And I think they can shine a light on how people, or at least people who think like me, will look at subsystems and form expectations of them. (And, perhaps, how to design a subsystem that communicates itself better.)

Firstly, and most personally galling – it just does not work when left alone. The numbers do not function with the amount of rests you will get through. The GM needs to interfere and give other rewards, to specifically make room for a massive number of bonus rests at the end of each level, something. If the project numbers were tuned to the amount of rests a campaign would get without the GM handing out any bennies, this post would not exist. This is in part one of the cultural things from its D&D legacy – handing out bennies is strongly taken as given for this subsystem. It’s assumed the GM will be handing out more than the system produces as a default, and here and there it comments a vague suggested rate, if the GM happens to read that section closely enough. In essence, the GM’s job is to fix the numbers, and if they don’t, they remain broken. Rather than them tuning to be faster or slower from a baseline, they need to tune up to make it work at all.

I don’t know that there’s a lesson to be learned from this point. To be frank, this is the sort of thing I haven’t encountered before in any other game I have some respect for, and I’m just gobsmacked by it. I think the actual core of it is that the system wants the 10-20 downtime rolls per level to be the actual numerical baseline, but still wants them aesthetically tied to the rests you take during an adventure. Because those are very different rates, it then has to propose that you throw in a big pile at the very end to catch up, and hopes that will be intuitive enough for people. I do think sabotaging the pace of crafting was better than sabotaging the pace of combat rests, since the attrition is a lot more important there, but, still.

Secondly, there’s the core of how the crafting options are positioned in the system at all. The culture of D&D hands out loot relatively arbitrarily. It simultaneously ties it into numerical progression, provides average rates for how many magic items a player should have at any given time, and claims they’re an optional luxury that can be done without for a full game. Draw Steel makes much the same claim, and tells the GM it is not obligatory to give a player any magic items at all. (Let alone consider what magic items that player might specifically be keeping an eye on as worth giving, vs any other option on the list.) Crafting, then, is an option that seems to be available no matter what. Since magic items do mean increased damage output and health, they’re a boost that I would certainly classify as obligatory, given the choice between getting one and not. And, the GM controls when they hand it out, so the system seems to be presenting a choice you don’t otherwise have – when, in actuality, it’s still the GM’s choice, not yours.

The lesson I conclude from this is that the inclusion of the subsystem signaled and implied functionality that wasn’t there. Which brings me to the third conclusion, the general one: including a system signals to readers that it’s an option, one they can engage with as they would wish to. Including a mechanically detailed system is a much more significant signal on this front. And, projects do distinctly appear to be that! You have several progress bars to juggle as you pursue different things, a laundry list of goals to want, numbers to crunch to figure out what your roll bonus will be and if you need to invest further before heading onwards. Because there’s so many mechanical hooks to look at, as a player, I see something to conceptually orient myself around. That’s what mechanical detail expresses, on an aesthetic level, before you even get to the gameplay.

That’s one of the reasons why combat systems are the way they are. Or rather, there’s a feedback loop to it. Combat is conceptually difficult and complicated – if you want to model it, structured and detailed mechanics are very important. If you include those mechanics, players are drawn to them and centralize gameplay around it, in comparison to lighter subsystems. The mechanics themselves signal this with their presence, and, implicitly, their rigidity offers a reprieve from the GM calling all the shots. Within the domain of the mechanics, the ladder can flip, and now the GM is subordained to the rules, too. The players never get to be on top, but a temporary changing of the guard is still a very exciting prospect.

As designers, that’s something we can take advantage of! Something I had, honestly, taken largely for granted until I sat down to examine this impulse. Where there are rules, I will go, because I enjoy interacting with rules. That makes sense to me. But that’s not the only reason. Where there are rules, I can know the expectation is that this is going to happen. Where there are lots of rules, detailed rules, I can know the expectation is that this will be in focus, will be relevant to the point that all this detail was called for. In a system with as many combat rules as D&D, I know I have to be ready for a fight. That extends to Draw Steel. And within that fight, the rules are calling the shots.

If you present a mechanically detailed subsystem, this is how I’ll implicitly interpret it. When that’s a signal you want to send, there is aesthetic value in detailing those mechanics. It will communicate to your readers.

Draw Steel, ultimately, wants the GM to be in full control in the game and able to do whatever they want. It’s an inheritor of D&D, that’s really to be expected. The whole OSR movement sprang up out of a desire to preserve GM sovereignty over the encroachment of rules (that’s a somewhat reductive framing, but I’ll stand by it). The reason that it interests me, as a player, is that its combat doesn’t do that in the way that, say, Daggerheart‘s does. Within the domain of a fight, the game is not serving the Game Master, but the other way around.

But, that’s only within a fight, it turns out. Even among the other modes that look that way. And I think that’s a bit of a waste.

The weight of the world, or, On GMing

Have you ever GMed before?

…This is a blog post, I can’t hear your answer. I don’t know why you bothered.

To some degree, the GM is just a role like any other. Plenty of games have clearly defined roles of several types, several have no delineation and keep all players as equal, some have rotating roles, most keep them locked after initial assignment. The dynamic of one GM and many non-GM players is just one enduring arrangement among several, and it’d be an error many pieces fall into to claim that’s always how RPGs work.

But, that is what I want to talk about today.

I have GMed before. I’m doing it right now, in fact. Not right this minute, but, as in, there is a game I’m running in progress at the time of writing. It’s a fun experience! To anyone reading this who’s never tried it, I highly recommend giving it a whirl. There’s an odd stigma in places about it, but, it’s very unearned. GMing isn’t all that harder than being a player of any other sort, and, really, it’s a lot of fun getting to control a whole world instead of just one character.

There’s a meme, of sorts. Mostly in D&D spaces, but I’ve seen it in RPG spaces overall. The “forever GM”. The complaint is, one player always has to be the GM since everyone else they play with refuses to try it out. It’s a meme I find rather frustrating, for a few reasons. It’s not true to my experience at all, everyone I’ve played with has also been a GM for various games, for one. For another, it’s a self-fulfilling concept. The whole arrangement is predicated on GMing as an unfun chore, and, once it builds up that reputation, it makes people less inclined to want to try it out, meaning they keep pressuring the one person most willing to do it. I can’t count how many times I’ve seen someone interested in a new game, but decide immediately that running it is too hard because they just assume that GMing is too difficult, and so, they just don’t play the game at all. Let me tell you, just as a personal tip, all of my capacity to acquire consistent play groups and try out the GMed games I enjoy was by GMing myself for people who were also inclined to run their own things, and getting them excited about what I was running. I’ve gotten to play games I thought I maybe never would, by running them, and then one of the people I ran for ran it as well! It’s a much healthier arrangement than all the complaints would have you believe!

And, really. There aren’t that many more responsibilities the GM has than any other player. It’s not as much work as people act. Here, I can list them all out now, it’s really not much:

A GM is a narrator and tone-setter. The GM’s job is to set the mood of the places, events, and people that come up in play. Giving them all identity is important, but the real meaningful part is selling a tone with them. Boring or repetitive occurrences will fade out of the players’ thinking!

A lot of the skill for this comes from paying close attention to your players. What kinds of themes engage them? What parts of the premise of the game excite them? You’re the one trying to tie this all together into a coherent narrative, or at least something that feels that way. Be a writer!

Alongside that, a GM is a world simulator. They’re controlling everyone and everything that isn’t the PCs, after all. (Maybe there’s a few special side characters the players get to control. But besides that.) The GM has to make that all come together and make sense. Where is everyone? What are they doing? Who are the major power players, what do they want, and how are they making it happen? When the GM poses a scenario for the players to deal with, it has to make sense. When something happens, the players need an answer for how and why. When the players want to try something unexpected, the GM has to know what the results might be.

Together, those two roles define the primary responsibility of a GM – to present situations to the players. Hypothetically, an RPG occupies a world as detailed as the real one. Ideally, the GM’s head contains that world. However, in practice, you have to present specific chunks of it. Contexts to exist in, enemies to defeat, conceptual challenges to their ideals, etcetera. These scenarios have to make sense as an emergent part of the world, and they have to thematically fit into the growing narrative of play. Together, that’s GMing!

They also have to make those situations mean something. If there’s no challenge, the players will glide through it and not really think about it. As a result, a GM has to be a tactician and an opponent. You have to be prepared to beat your players down, to be blunt. Most obviously, this is true in games with combat as a focus – you’re gonna control a whole set of enemies, and they have to match the entire rest of the table, or it’s gonna get boring. But, even outside of games like that! Are you setting the difficulties they’re rolling against? Are you handing them an obstacle progress bar they have to work to get over? You need to be making sure you’re giving the other players a run for their money. A GM who doesn’t know how to outthink their players and put obstacles they’ll struggle with is a GM people aren’t going to enjoy, much, unless they’re really uninterested in the game part of the game. (And if they aren’t interested, that’s a problem in its own right.)

But, then again, a GM can’t play too hardball. For the most part, your players are supposed to be winning. Remember the narrator role! You’re here to weave things together into a coherent story. How many stories do you know that have their protagonists lose all the time? How many actually serious stories do you know like that? Plus, the players have to be satisfied with their time. And remember how much power you have – you control the entire rest of the world. If you used that, you could absolutely just make them lose by arbitrary decision. That means, when you control the world realistically and with thematic appropriateness, you have to actively not do anything that would mess with the players too badly. You can react to what they do, but they don’t have the same power. You’re not here to win. A GM is a jobber and a heel – you look bad and lose so the players look good.

And on that note, it’s important that you make them look good. The players and their characters are the stars of the show. So keep the spotlight there! If too many of the important actions in the story are made by characters the players have nothing to do with, or, worse, that the players want to stop but aren’t allowed to, that won’t feel great. If too much time is spent with NPCs talking to each other and being in focus, then the players are essentially just there while you tell them a story, and that doesn’t give them much autonomy at all. They might enjoy it, but they aren’t really playing so much as just going along with things. The NPCs should exist in relation to the players, and be there to bounce off of them. Make them challenge, make them disagree, of course, but don’t make them outshine, and don’t make them have nothing to say. The PCs are the primary characters – the job of the GM is to control all the secondary characters that frame and exist around them. You’re there to be a foil and a secondary support.

Which, of course, ties into the earlier goals. Those secondary characters need to thematically play off the interesting aspects of the PCs to tell a coherent narrative, and they need to be internally realized enough that their actions are a coherent part of the world. No having people do things just because it’s convenient – you’re here to make this feel real. (But it can’t be so real that the PCs don’t feel special.)

But, remember. You have to keep them on your toes. Support them, keep them in the spotlight, yes, but don’t let them feel certain in how it all works. As a GM, you’re a plotter and a deceiver – you can’t ever show your full hand. The players are going to get cocky since they know they’re the center of attention. They’re going to get confident, since the challenges you present to them are supposed to be overcome. If they spend all that time confident and smooth, it’ll get boring. Throw them some curveballs! You need to have a plan for what’s going on, and the players shouldn’t be able to see that plan until they’re through with it.

So, if they guess what you have in store, maybe add a bit more? Oh, but that might be unfair. The ideal is that your plan catches them by surprise, and then they work to adapt and overcome it, and nobody has to pull any punches. If that doesn’t happen, you’ll get varying opinions on how much you should change things up or how much you’re allowed to change things up – whatever happens, you’re the one who needs to make it work. If the players figure out who’s behind the plot to bomb the senate and install a monarchy, and then they stop it, and nothing surprising happened… well, you’ve failed, kind of, right?

Of course, the players have to be interested enough to theorize. They have to care and wonder about what they don’t know, or there’s no point in you plotting it out. If they just tune out all the politics of it, you’ll need to make it pressure them into action, so they do worry about it. They rely on the local democratic systems to form a community supply of resources for dungeon-delving, or something, and now they have less equipment next mission due to bitter sentiment. But don’t force them too much, that’s railroading. Your plan has to be able to go off without a hitch, and the players have to be able to never learn it ever happened, and they have to be able to find out everything about it and stop it well in advance. But, none of those should happen. You see what I mean?

In short, you’re the wall maintaining the fog of war. The players don’t get to see the whole world, and you have to be sure the bits they can’t see matter to them. Otherwise, why would they try to find them out?

And that’s it. Those are, broadly, the core skills and goals that go into running an RPG out of the box. Practice all of those, and you’ll be right where you need to be.

…Now, that said. Running a game out of the box isn’t the be-all end-all of what a GM does.

Sometimes games are bad. Sometimes games are mostly good, but have a few bad bits. Sometimes games are good, but have a few bits that don’t work specifically for the tastes of the people you’re playing with, even if, in the abstract, they serve a purpose. Or, a lot of the time, the game is actively relying on you to fill in blanks, make rulings where the rules don’t cover. It’s rare that a system will claim to cover everything relevant to play. Resolving the rest is where you come in. As a GM, you’re a game designer in your own right. You need to fix rules that don’t work for your group, add new rules that do work when things come up the game can’t cover, rebalance when certain options prove too strong or weak, and keep on the ball with the system enough that you catch these things before they get too bad.

But, you know, also bear in mind your other responsibilities, and the limits they put on what you do here before it’s unfair. If a player has a powerful tool that lets them beat challenges, and you nerf it, and then the players lose… remember that you’re also the one posing the challenges to them. You’re the one trying to beat them (but not actually beat them too hard). If the players get frustrated, because the person trying to stop them has the power to remove their tools, that’s fair, right? It’s easier to buff a player whose options are weak than to do the other way around, when you get down to it. Though, of course, don’t over-buff, you still have to put the pressure on them.

And, on the other side of the arrangement, the game isn’t all you’re running. An RPG is a social event, and that means organizing people. As a GM, you have a certain level of social power over the other players in this context, so you’re the best person to set schedules, handle absences, that sort of thing. You also get to decide who plays and who doesn’t, which matters if conflicts come up. You’re going to need to be good at conflict resolution, and at expressing your authority without alienating people. It’s an extra layer of social work, essentially, but it’s not too bad if you’re experienced dealing with and defusing uncomfortable social dynamics. If you aren’t, a lot of problems emerge, so, best to be prepared, just in case. If a player is a problem, you need to be ready to intervene, or everyone’s gonna suffer. Even if they’re your friend.

Ultimately, though, there’s one goal each of these roles serves. One point to GMing. You’re an entertainer. You’re here to ensure that everyone has fun. If that means bypassing or ignoring one of the aforementioned roles, that’s how it goes. Some tables don’t actually want a serious challenge, and that means you shouldn’t be that for them – and that, if the game would present that by default, you have to put on the secondary game designer cap and tune it down to be easier for them. If that means an extended character drama complete with silly voices that’s mostly just freeform and doesn’t touch on the system at all, that means that’s where the job of the GM is. If the players aren’t having fun, something is wrong, and you’re the person who’s there to fix that. All the roles I’ve mentioned above are just elements of how to keep the experience fun for the players.

That’s all! See? That’s not so bad, is it?

…Isn’t it?

So, here’s the thing. Initially, I did want this post to be what I outlined in the intro. A call to action that more people give GMing a try, because it’s a fun time and presenting it as a terrible fate sucks for everyone. I stand by that, still! I think it’s good for people to get the experience, and it’s a great way to gather new groups and play new games you otherwise might not get to.

But, you know…

I just outlined a lot of work, right?

A lot of contradictory work, to boot. These roles are filled with conflicts of interest. You have to balance enough to challenge the players, but not beat them outright because you have the power to just change the rules, but if that balanced challenge would be dishonest to the world you’re playing in it’s bad to do that, but maybe that means the world is wrong and the game takes precedence? It’s all muddled together, and kind of a mess.

The thing is, for most games, all of these are responsibilities of the GM to some degree, but not evenly. Games that claim to discard game balance won’t discard it entirely because the result is nonfunctional in play, but they will give an answer to the question of, hang on, which should I prioritize? Games with clear win-loss stakes that don’t just slam the play to a halt (like unmanaged PC death, for example) shift the GM away from the responsibility of making it so that the players win. They don’t lose it entirely, but they get a decent way there.

Plus, you’re not going to get people to agree on the importance of these goals, either. I imagine almost everyone reading this had a balking reaction to at least one of the ones I presented, and, to be entirely honest, so would I. Some of these goals I think are just bad, or undesirable, or, at minimum, not for my tastes. But they’re all goals that I’ve seen people express, or act on, in the process of GMing. They’re all present as semi-consistent expectations. They are all, for some subset of the audience, things they expect a GM to do. If you’ve spent any length of time GMing, I imagine you either know how to do the things I’ve listed, or pointedly don’t do them when you run, with the particular mix of the two depending on your tastes.

And… that’s kind of strange, right? Looking at all this. A long list of contradictory goals, a massive amount of mental bandwith eaten up by them, and a large amount of stress to juggle when you’re trying to ensure that you haven’t imbalanced your priorities too much for the players to have any fun. It’s a strange position to be in, and, when it’s all laid out like that, it makes perfect sense why it gets talked about like it’s such a dire fate.

Like, of course people are tired of being put in a role with a mile-long list of contradictory responsibilities. Right? I find it a lot harder to hold that against people than I did when I started workshopping this post. And there’s more dimensions of work, which I haven’t even touched on. These are all your responsibilities in play – but what about outside of it? Planning encounters (and drawing maps for them, so many maps) if your game has fights, building a mystery if your game has mysteries, writing a whole cast of characters in advance, organizing the game in the first place… there’s a lot of work beyond the game, too. And that all very much compounds on itself.

Plus, you barely even get to play the game!

The game and its play

One of the points I harp on every so often in this blog is, the GM is a player role. They’re a player like any other. And, from that, you might go – right, so, they’re playing the game. And, that is true! But it’s also not.

[Play] isn’t just one thing, you see.

You sit there and you talk with your friends. That’s part of the fun, and you certainly get to do that. You all share talk on an imagined world, and you get to throw your piece into the pot to make it what it is. That’s also part of the fun, and, it forms a meta-layer of play in its own right, which you’re also playing. (The goal of this meta-game is to get your particular vision of the world and what you want it to be like to have as most traction as possible. In a sense, the GM authority gives you an edge in this game – but in another sense not! Though, I digress.) But, I mean the lower layer of play. The mechanics. The part where you take a bunch of mechanical toys and you link them together to try to make a shape where you win. The puzzle of gameplay, the thing the players do all the time.

You might have some of the toys, but you’re not playing like they’re playing. You don’t get the game.

You don’t get the win condition, in particular.

RPGs get talked about, a lot, as though they’re games you don’t win. And that’s kind of true, but not actually. More specifically, it’s more like a segmented series of games with discrete win/loss conditions that you move through as you go. Here’s a fight you need to win or lose. Here’s a vault you need to break into and get this thing out. Etcetera. The shape is different from a board game where you have your final victory in mind from minute one, but, having a goal and working towards it is one of the most common activities in RPGs – especially ones that lean heavily on a GM role. The GM is very convenient for this, in fact! They get to play the opposition.

Which should be a counterargument. Right? The players play on one side, the GM plays on the other, whichever side plays better wins. It’s competitive solo-vs-team, the GM absolutely plays with the mechanics like the players do. What am I on about?

And if you’re like me, that actually is the position you take in response to this argument. That’s where I’m at, at the very least as an idealized mode of play. That’s how I’ve played and ran games (ones that can work with it as an approach, at least), and I’d argue I was playing, then.

But, statistically speaking, you probably aren’t me. And I would expect that that isn’t your response, either.

I wrote about this before. (Among other things.) The prevailing sentiment in RPG spaces would object to a GM playing like that at all, and directly declare, hey, you’re not supposed to be opposing the players. That’s adversarial, or something to that effect. When you do play against the players, you’re presenting obstacles to challenge them, but they should overcome them. If you actually defeat them, you’ve screwed up.

Now, sometimes, players catch flak for the same thing. They might get judged if they try to win too much, called a powergamer or selfish or what have you. But there’s, comparatively, a lot more wiggleroom. One of the privileges of being a player and not a GM is the right to try your hardest. And, really, that’s the [play] as I’m discussing it here – having a goal, doing everything in your power to make that goal happen, and, whether you win or lose, having the satisfaction of being pushed and really making an effort.

And, how much that matters is going to depend on the person in question. I know some people who don’t care at all about that experience, and it’d make no difference. I know some games that don’t put meaningful pressure on anyone to try to overcome them. But, for the games that do, and for the players that do… usually, the players get that experience, and the GM gets the job of ensuring the players have that experience, rather than actually getting it for their own.

The GM’s got a different goal, see. A meta-goal. Hiding right in the implications of the last one on the list. The GM is there to ensure every player has fun, and that’s what they’re optimizing for. And that means…

Well, it means a lot of stressful and confusing social maneuvering, and figuring out the particular tastes of every player involved. (That thing I said about some players not caring about being pushed and forced to try to win from their challenges, and other players really caring about that? You have to spot that, and cater to both.) But, more relevantly, that means it’s a goal! The GM does have something to aim for, and win or lose at, after all. They’re playing a social game, while everyone else is playing a puzzle game.

The trouble with that, of course, is that everyone is playing the social game.

How to win at RPGs

People don’t really come to RPGs to overcome obstacles.

That’s, a lot of the time, the gameplay. And it’s fun to do! [People] doesn’t cover everyone, in the above sentence. It only sort of covers me. If you drop a game in front of me, tell me about the fights I’ll have to win and the puzzles I’ll have to solve, I can be satisfied with that. If I read through a rulebook to see what it’s about, and the rules are all about exactly that, then, yeah, you should come to the table prepared to focus on beating some sort of challenge.

But if you’ve read, like, any RPG before, and how their introductions frame themselves, you’ll know that’s not it. Hell, if you’ve just asked someone to explain RPGs as a medium to you before, you’ll know that’s not it. It’s not what people sell you on.

RPGs are collaborative storytelling and improv exercises and opportunities to indulge in playing a character, and all that jazz.

These aren’t activities people conceptualize as having winners and losers. And, for good reason – improv where you’re trying to have your way the whole time and not let the other guy do anything to contradict that, sucks. It’s a bad experience for the both of you, and also just rude to do. On the same token, however, even if your partner isn’t trying to push things away from what you want, if things never go in a direction you enjoy, and you don’t try to pull them towards that, at the end of it you won’t have gotten any satisfaction out of the deal. And, probably, you’ll make them feel bad, if they wanted you to have fun as well. (Which, ideally, they do.)

So, the ideal state is somewhere between those two. Everyone is getting some amount of what they find fun to play as an improv story, and nobody’s getting boxed out because of it. That’s what idealized freeform roleplay feels like, among other things.

That’s what winning at RPGs is. (Kind of. It’s actually slightly better practice to err in favor of satisfying yourself a bit more than other people, because you know your interests better – and really, all of this is going to be messy unless you’re openly discussing things as a group and willing to shift gears if someone is uncomfortable, but that’s its own beast.) That’s a goal that everyone is working with when playing, including the GM.

This goal is, at times, directly contradictory with the goals of the game itself. What if you, personally, would find the flow of the story more interesting if you lost this fight? Should you try to lose on purpose? Is that against the spirit of the game? Is that rude to your fellow players? The answers to those questions are contextual on what the game wants you to prioritize, and the answer can be pretty hard to tease out if it isn’t open about it. Even worse, if your fellow players have different answers than you do about what’s kosher and what isn’t. (And if they do, then, they’re unsatisfied and upset with you – which means you haven’t won the social game after all!) It’s kind of a mess, and navigating that mess is one of the reasons RPGs have a diversity of different priorities to their design. You can make games that privilege the mechanical play over player satisfaction, games that contextualize their mechanics as props to be arranged to produce that player satisfaction, games that focus on different sorts of satisfying shapes of narrative as what they produce, etcetera. It’s neat! And a massive headache, unless everyone’s on the same page.

Which brings us back to GMing, ’cause the GM is assigned with ensuring that everyone wins this social game. That means unless everyone is satisfied, and on the same page of how this game will satisfy them, the GM is on the hook to fix it. The trick for getting what you want in an RPG, for “winning” as I’m talking about it here, is left as, don’t worry, the GM will ensure it. Or else.

One of the inspirations for this post was reading through the quickstart of Realis. Realis is an interesting game. Essentially, the gameplay is informed by having a list of declaratively true statements about yourself and the world, and, if your actions are in line with one of those true statements, you just succeed. For instance, “I am a master of swordfighting” as a sentence lets you narrate yourself beating someone in a swordfight, and unless they have a statement of their own, you win.

It’s neat! I enjoy tech like this, I’ve appreciated it in Nobilis and I appreciate it here. And, here, too, there are adjudication rules for when statements clash. Each statement has a numerical power level, which increases over time as specificity also increases. It makes for an interesting balancing act – “I am a master of swordfighting when wielding my trusted blade” is a more powerful statement than “I am a master of swordfighting”, but it takes more work to engage. Similarly, it’s the duty of the GM to define the statements the enemies are working with, including their power rating. Without that, any statement would be just as true, so, incentivizing specificity means the GM needs to be putting solid numerical pressure on them. In fact, they only grow in power and specificity by experiencing failure at the hands of a stronger enemy statement. Mechanically, posing statements that are stronger than the players’ is the main point of interaction the GM has.

Anyway, here’s a snippet from Realis.

This is, I think, a fair thing for the game to say. I’m always in favor of games communicating how they want you to play them. But, on the same token, it’s asking GMs to not use the one mechanical form of engagement they primarily have. Or, at least – to only use it for the purpose of the social game. Posing sentences for the players to overpower or have to outmaneuver isn’t gameplay, in the sense of a challenge two sides have to overcome one another through, it’s props for the production of satisfaction.

That’s the trick to it. That’s what’s going on when people expect victory while playing D&D. That’s what is expected of how a GM engages with their mechanics, a lot of the time. Not playing a game, but arranging props for other people to feel good as they play around.

And… I have complicated feelings about that.

Murkiness

There’s a lot of GMing advice out there, much of it contradictory with itself. You can find advice on how to run a meaningfully difficult combat encounter and advice on how to make enemies make realistic mistakes and flee when they hit half health so the players don’t face difficulty that feels too much like a game. You can find advice on how to prep scenarios in advance so the players have things to engage with; advice on how to never prep anything because that’s violating the spirit of player freedom and truth to an emergent world; and advice on how to prep things in such a way that the players don’t notice, so they still believe in their freedom and the emergent world even if that isn’t really there. You can scroll to the first section of this post again. And there’ll always be more of it, and all of it is pulling in contradictory directions.

That’s because games want different things from their GMs. Realis up there wants you to be focused on the social layer of appealing to the players’ sensibilities, in particular appealing to their sensibilities of what an exciting narrative might be. Any number of OSR games out there will tell you to hold to realism of the world as a highest principle, and then never quite state that you should play the social game to be sure you appeal to whatever the players think realism is like, because their immersion is necessary for them to keep playing. And… heck. I’m a big proponent of games standing on their own merits, but, the quintessential bit of GM advice that always grinds my teeth in? [It’s fine to ignore the rules or get them wrong, as long as people have fun]? It’s true, right? Even just from a pragmatic angle, the people you interact with being socially unhappy just affects you more than not following the rules does. In the long run, adhering to a system gives more consistency to the experience of playing, but, when you’re a human being, a social primate, and not me on the internet yelling about elfgames, that’s a very distant priority. This is a social affair, and that means placating people and managing expectations to keep things smooth.

That’s the murkiness of GMing. It’s the murkiness of RPGs as a whole, but, a lot of the arrangement is set up to resolve that murkiness by pushing it onto the GM to navigate.

The role of the GM is, in many ways, a scapegoat. You can blame them if the system is bad, because they should’ve fixed it. You can blame them if the social dynamics don’t work out, because they should’ve moderated it. You can blame them even if you do everything you’re supposed to, and they do the same, and everything went exactly as it should; but you didn’t feel satisfied at the end.

And… I can’t in good conscience come on my blog and go “hey, being a scapegoat gets a bad rap, you should try it,” right?

So, then, approaching GMing starts from figuring out how to not be that.

Those priorities I listed at the beginning? Most games don’t ask all of them of you. Some of them do. You don’t have to listen.

Some of these priorities you’ll find easy. Some of them you’ll find hard. I’m good at charting out thematic events to hit the players with, and worse at holding a simulation of a world in my head. More importantly, some of them you won’t find fun. It’s worth experimenting running different kinds of games to see, but, start with what excites you. With only the responsibilities you would want to do.

Most importantly? Always ditch the last one. Having one person responsible for everyone being happy is a cursed position to be in, and, it’s got conflicts of interest with every other piece of authority you’ll have. That’s the one that makes you a scapegoat, more than anything.

Once you’ve got the set of what you’re up for trying, be open about that. Be clear about what the play will and won’t engage with. You’re not here to make meaningfully difficult fights, but you can make them feel in line with the world as it’s been presented? Cool! There’s players who will jive with that, and players who won’t. If they know that in advance, everyone will be a lot happier from the get-go.

And… yeah. Yeah, that does mean that some of your friends won’t want to play what you’re running. A friend group tends to cover a broad spectrum of RPG play interests, usually incompatible. And that’s okay. It’ll be a better social experience this way. (And, even if you value the game as a consistent friend group activity more than any of this – being open that they won’t get what they want if they play is still the best answer.) People deep in the RPG trenches tend to cluster around enjoying certain sorts of play as a direct result of this. It’s easier to play with people who like the same kind of play you do.

All that’s pretty hard for a newcomer to GMing, or a newcomer to GMing in an ideally-less-exploitative manner, and you’re gonna slip up here and there. Some of these concepts may sound fun, but not be something you can pull off in execution. I’ve been there myself. The important thing is to learn your limits, be able to adapt, and explain as such to your players.

To a lot of players, the GM is a black box do-anything machine.

Probably the biggest asset a GM can have is players that understand just how much that’s not the case.

After all that, I still recommend giving GMing a whirl. It’s a lot of fun, when it’s going well. Not fun in the way being a player is fun, they’re not even close to the same experience… but even so, it’s fun in its own way. There’s value in that experience, too.

(And, to the designers in the audience – all this is a short introduction to my real agenda, why I think the GM role is deprecated and contradictory tech and a more distributed or systematized structure for running games is well worth investing time in. Check out Emberwind‘s enemy AI, or, any one of several Belonging Outside Belonging games. There’s so much cool GMless tech out there.)

(But, I digress.)

So… yeah!