Five hard questions about OSR play

I have read a bunch of OSR theory and played a bunch of OSR games, and I have the following questions (some of which I have tentative answers to, some of which I do not).

(1) When people say that in an OSR game, we make decisions from the point of view of our characters, do they mean what our characters would do, or what would be in our characters’ best interests?

(2) To what degree is an OSR GM responsible for pacing, and what tools are they allowed to use in service of that end?

(3) To what degree can OSR games handle player-authored material without losing integrity of setting? Does player-authored material threaten any other aspects of OSR play?

(4) To what degree can OSR games handle situations/characters/etc created specifically to target individual PCs and their interests without losing integrity of setting?

(5) If OSR games offer total freedom, why are they almost always about murderhobos wandering around and smashing pots in search of rupees?

I don’t know if this is the optimal place to have this discussion, since I think most people here will bring the same assumptions that I do to these topics, but we’ll see.

  1. I think the difference between adventure focused games and story focused games is that with the former you want to make optimal decisions for your character and with the latter you want to make interesting decisions (that may or may not be in the character’s best interest)
  2. You can skip obviously boring moments and fast forward in time if none of the players want to do anything. Not sure if you need a tool to do that.
  3. Do you mean players establishing facts about the world? It would be the same for e.g PbtA games with an established setting (like The Green Law of Varkith for Dungeon World). As a GM you can ask prompting questions that limit the scope of the answer / shape what a possible answer could be or veto answers that would contradict already established facts
  4. Good question!
  1. I basically agree, although I think that the character in an OSR game functions as (a) a handicap - i.e. you might make a person who gets into trouble or doesn’t do the smart thing as part of the challenge, and (b) a face to wear as you interact with the setting.

  2. Can you elide complex actions into a single roll with minimal fictional positioning to save time if people aren’t that interested?

  3. agreed.


1 The decision to go on adventure is already a suboptimal choice and pressing the red button for shits and giggles is part of the fun especially in one-shots, funnels and when you have torch-bearers to spare.
2 I think that is happening all the time e.g. hiring retainers, buying equipment, travelling (if it’s not part of the adventure itself) and I don’t see why that couldn’t be expanded to other activities as well.
4 In an open world / sandbox style of play over many sessions the PCs will necessarily entangle themselves in the world and in a few-shot you can use the Beyond the Wall questionnaire to tie PCs to a specific problem or element of the world without losing integrity to the setting. @RichardRuane does that all the time in his games.

  1. I struggle with this question in some OSR games, personally. (I think I mentioned something about it between sessions 1 and 2 of a Troika game @shanel ran, even.) I feel like gonzo exploration modules call for reckless abandon (so you’re doing, in theory, “what my character, an adventurer, would do”), but high-stakes problem-solving scenarios call for making cautious, informed decisions. There’s kind of a tradition of giving as little guidance as possible in certain veins of OSR gaming, but I appreciate articles like the Bastionland Players Handbook for more explicitly spelling out the intended style of play. I would say that more often than not, if you’re playing an OSR game with extremely fast character generation rules, no mechanical support for character personality or background, and especially if looting their corpse effectively allows you to pick up their advancement where they left off, it’s probably safe to assume the character is more of a player avatar than someone with their own desires and inclinations separable from the player’s.

But I’d love to find out I’m wrong, as the alternative is often more fun.

  1. Are most OSR games assuming “murderhobo” play? Robbery, sure, especially ones that base advancement on gold acquired. I’m not sure most expect you to kill much, though.

The easy answer for why most follow patterns of earlier games is “genre and demand”—people still want that stuff, and that’s where the known market is for it. Kind of like how a guitar can “play anything,” but there are cultural and market reasons why you hear it used for some kinds of songs more than others.

That said, I am not sure I buy the “it can do anything” argument, except insofar as you could say that about any RPG that’s mostly freeform. I’ve read it argued that these games have combat rules just because that’s what makes the stakes and danger feel real, even if they’re not “about” combat, but that feels a lot less plausible an answer to me than “these come from war games, and that’s just the way we’ve always done it.” If RPGs had developed from worker placement games instead of war games, maybe the new retro games would all have detailed rules for city building or downtime work or whatever instead.


I agree with Jason’s answer to question five but will add that the classic D&D mechanics re-enforced this style of play. In general, it seems to me that the classic D&D mechanics are based around one central idea: exploring dangerous physical locations. Most or all of the mechanics support or encourage this: saving throws, combat, healing, reaction rolls, spells or abilities to read unfamiliar languages or detect secret doors. Early editions awarded XP for gold. This provided a mechanical incentive for PCs to thoroughly explore locations instead of just grabbing a MacGuffin and running. But it also created a mechanical incentive for characters to be treasure hunters/grave robbers as opposed to exploring a place for other reasons. In theory, PCs would break away from this in higher level, domain management play as PCs worked their way up from common tomb raiders to nobles to perhaps one day kings by their own hand. But in practice, it seems relatively few people ever actually make it to that level.

Many newer OSR systems, for example Black Hack and Into the Odd, provide alternatives to XP for gold, but adventure design still seems to be based mostly around the old assumptions, probably largely because of the reasons Jason gave but also because those are the types of adventures that the classic D&D mechanics encourage.


I’d peg it as being a mix of the two, depending on table culture and framework.

Again, this to me is less an “OSR” thing and more a “individual table” thing. I’d say that there’s no question of a GM being ‘allowed’ to use tools, but more what they and their players feel comfortable with.

Personally, I prefer to avoid GMs ‘tipping the scales’ regarding pacing, save for out-of-game concerns (“let’s call it a night here because it’s getting late and the next few rooms will likely take y’all a while”).

My colleague (and friend) Luka had a good post about this very thing. I think that player-authored material re: setting detail can be great, but there definitely does need to be a soft (or hard) veto option available to the others around the table to maintain setting integrity. I’ve been very frustrated with games that have had a broad inclusion framework with no veto, where things tended towards maximally gonzo with minimal consistency.

In terms of other player-authored material? It depends. If Player A brings me a new, well-written class that fits well into the world, cool, we can start using that. (Again, GM veto power is crucial here.) If Player A tries to assert fictional details about a location, rather than asking whether they’re present or not, then that would likely be an issue.

To the degree that those situations can reasonably arise organically from elements put forward in the game. If Evil von Moustachetwirl has fought the PCs several times before and knows their capabilities, it makes sense for Evil to construct a plan that is designed to specifically counter or neutralize the PCs. If Evil just knows that “some opposition is coming in” then having a plan specifically designed to neutralize the PCs would seem dubious to me.

All I can say is that this is frankly not my experience, so… :man_shrugging:

Like don’t get me wrong, there’s lots of dungeon delving and the like, but there’s also engagement with the social environment as well. Despite the original West Marches framework, IME folks generally do wind up going into town, interacting with the people there, and forming meaningful connections (positive or negative) which wind up playing roles in how gameplay proceeds.



OSR = D&D. People make decisions the same way they make decisions in D&D. All sorts of ways. Hopefully whatever is the most fun.

I might be missing key jargon definition regarding “pacing” but I like to stock random encounter tables with things I’d like to happen as the characters explore and then I’ve got:

  • Hand-wavey time vs. exploration time vs. combat time to move between
  • Cost and opportunity cost. Most obvious suggested by the rules is spending time to be thorough or quiet. Speed can be bought by being less thorough or more noisy. Failure to be quiet or thorough means you can take a second-chance shot the noisy/messy way. Excess time (a “turn”) or noise or breadcrumbs means a random encounter table check - the things I want to happen that I put in the random encounter table now get a chance to be trotted out.

Never comes up. If it’s some intricate backstory I’m afraid I’m disinclined to work it in but I’ll usually make a song and dance about not putting work in in front. I want the players to build lairs and research spells in game, not before.

Seems off-topic IMG. Party-centric.

Freedom is in great supply but little demand.

1 Like
  1. Both, depending on group. Classically I think the latter would be the normal approach.

  2. I don’t think pacing is an OSR thing. Pacing is a general RPG thing and applies to OSR the same as any other game. That said many OSR games have strict rules about timekeeping and hence “fast forwarding” in those is not usual as it would mess with ressource tracking like torches burn time and rations.

    1. Is a matter of group agreement. Some games don’t want this at all and are very strict on who has world authorship and who doesn’t. Others less so (see Beyond the Wall, my favourite OSR game that is full of shared setting creation and works just fabulous)
  3. Because most people want that from their games I guess. There are alternatives like the aforementioned Beyond the Wall though. It’s certainly not the way I like to play.


I’m not sure the point about timekeeping is true. Obviously you can adjust pacing while still enforcing timekeeping rules. I do this all the time in my (only kinda OSR) game. “OK, you travel for three days until at last you reach the foothills. Cross off three days’ rations. Albrecht the torchbearer complains about his feet.” You can keep track of in-game time even as the ratio between it and table time varies.

Some OSR DMs would say that this is cheating a bit – I’m not rolling for random encounters and I’m basically making the decision for the players about which parts are the interesting parts. But I think it’s pretty common nonetheless.


I run GURPS DF, which has both built in avenues for player generated fiction, like the serendipity advantage, which allows you to find just the thing you needed as you describe it, the Ally advantage, which allows you to design a partner or henchperson ir the Enemy disadvantage, which lets you do the same with an opponent, and also runs with a number of OSR sensibilities.
I had to hard veto some player created fiction-" I have a general knowledge skill about this town, and made the roll, so I find a manhole that leads directly into the secret ninja lair"
I also rewrote an adventure based on lies told by the party’s bard to reduce interest in having some rando npc’s (actually bandits) try to join their quest for buried treasure. He said that they were seeking the tomb of a lich king who had once marched with an army of liches, and wore a three foot blazing crown set with tourmalines the size of plums. The warlord lich was actually part of the setting, they had already met (and liberated) a collossal earth elemental enslaved by his artifact (the setting was partly based on The Giant Under The Snow) and I incorporated the story by having the treasure they sought be buried in his tomb, where he had a small company of wights, and wore a one foot silver crown set with tourmalines the size of plum stones.

Correct, but as you stated yourself it’s not by the book as rolls for random encounters are required by RAW of many old school games. It’s not my preference either but wanted to point this out as it is one aspect that’s very different from many modern games. As Gyax said " You can’t have a meaningful campaign if strict time records are not kept". I don’t agree with this but it’s the old school way. That said it does not mean you have to play out every minute of a game, certain shortcuts are valid. Usually there are rules telling you how to handle time, even if you gloss over it in a “You travel three days” way though.

1 Like

(1) I don’t really do either of those things. I mostly treat my character as an avatar that I’m inhabiting, a bit like the character you play in Skyrim. Most of the time you’re just doing the thing that seems the most interesting, but occasionally you slip into “character” because the situation would be more interesting if you did that and you felt like it. Most of the time I’m doing the thing that gets me out of trouble, grants the most rewards, or occasionally pokes an interesting-looking hornets nest because I want to see what will happen.

(2) The main “tool” is probably the random encounter table. Things don’t stay safe for long. Generally, the players can control pacing by doing what they want. Want to drop a mission and start a new one? Fine with me as long as you can get out of your current situation.

(3) I wouldn’t say that player-created content threatens OSR play per se, but it can make exploration less fun if you get to make up the things you discover.

(4) I’m not sure what you mean here. OSR games threaten individual PCs all the time. How would you say this threatens the integrity of the setting?

(5) Players like being rewarded for figuring out how to survive dangerous situations. They like it so much they will ask for more dangerous situations to throw themselves at. It’s one of those primal human drives, I guess.


I think 4) means “to what extent in an OSR game can you add plot or setting elements that relate directly to one of the player characters without messing up the neutral sandbox?” So the players have decided to go and investigate the bandit camp, and the GM, looking to tie things into one of the PCs’ backstory, decides that the bandit leader is actually their former mentor. Cue emotional conflict, but some people would argue that it messes with the perceived integrity of the setting because, I mean, what are the odds? That’s my understanding of the question, anyway; I speak subject to correction.

Again, I think this is maybe not an OSR/not-OSR issue per se.

I actually wonder: what if the player makes that suggestion about knowing the bandit? Does that break everything?


I’m usually OK with the player’s helping create a narrative, but it really depends on the player(s). However, I will always stress the Czege Principle, which states, “When one person is the author of both the character’s adversity and its resolution, play isn’t fun.”

When I run PbtA games on the other hand, I do create a culture of asking questions as the GM, and using the answers.


Again, I think the answer varies. I’m no expert, but my sense of the community is something like this:

Example 1

Player: “Hey Alice, I was thinking, you know how my character’s backstory has that thing with a disgraced mentor who fell to the dark side? I think it would be super fun to see him again!”
GM: “Hmmm … I’ll bear that in mind!”

Some people might think this spoils the surprise or is pandering or whatever, but I think it can fit within the boundaries of OSR play.

Example 2

GM: “As you round the corner, you see five bandits seated at a filthy table, apparently dividing the spoils of the heist. In the flickering firelight, you see the leader’s face, and it’s --”
Player: “Oh! Oh! It’s my old mentor!”
GM: (who had not previously decided this) “Yes, it’s Kroth himself, the livid scar your axe left on him twisting his smile into a sinister leer.”

I think this is something most people would consider not OSR, no way, nohow.

Given the OSR’s focus on problem-solving, investigation and exploration, I think that any system whereby people can just change the givens of a scene on the fly is probably outside the boundaries.

Again, no expert, just my sense of what the community thinks.

As to whether you can do this with a system that most people think of as OSR: absolutely, you bet, but most such games don’t provide any support for that idea.

For a great dissertation on this very subject, check out John Harper’s piece on “Crossing the line”.


Blockquote Cue emotional conflict, but some people would argue that it messes with the perceived integrity of the setting because, I mean, what are the odds?

As I read OSR style play, a big principle is that “setting logic trumps story logic.” Which is what this seems to bump against. But that doesn’t mean things that satisfy story logic can’t happen, they just have to feel justified through setting logic. As the GM, you have to do the work of making it feel plausible, which probably means finding a way to drop hints of it earlier on in the game. (Like knowing that your former mentor is out there somewhere. Maybe hearing that they’d fallen in with a bad crowd. Knowing that they might be in this part of the world. Hearing somebody describe an encounter with the bandits and something about their leader sounds so familiar that you can’t put your finger on it.)

I dunno. Personally, I find that principle to be hard to enact in practice. It feels like I need to employ a lot of smoke and mirrors to create the illusion of this self-consistent world, without actually planning the whole thing out in detail. On the other hand, I also think I’m sometimes doing myself a disservice by trying to shove the kinds of story and emotional conflict I can get from a narrative game into an OSR game, the same way I would be doing myself a disservice by trying to bring all of the puzzle, exploration and “skill-based” play of an OSR dungeon crawl into a PbtA game.


On the flip side: it’s your table. As long as you and your players are cool with it that is ultimately all that matters. If you and your table agree that this stuff is what you want then you are playing right.

In my opinion OSR systems are rules lite because they are meant to only adjudicate the most fraught of decisions (life and death of PCs). They are meant to be hacked, pulled apart, etc.

I remember being on a distillery tour in Scotland and someone asked the guide something like “is it okay to drink your scotch on the rocks?” Their answer was: there is no wrong way to drink our scotch. If you wanted to buy 20 bottles and bathe in it we could care less.

Do what you (and the other players) want at your table. You might be doing things that a majority of tables playing with the same rules aren’t doing, but that’s not the point. Especially of OSR games.