[Blog Post] RPGs and Creative Constraints: What’s behind the OSR and FKR?

New blog post from me that starts with an analogy to poetry and then tries to explain what’s behind the OSR and FKR on that basis. I hope it will provoke some discussion here, but I never know whether I’m going to catch peoples’ attention or not with these, haha.

There’s a fun Joesky Tax if you can make it to the end!


I really liked this blog post. I particularly liked the “authentic challenge” categorization of the OSR.

Authentic Challenge

  1. The game should consist of challenges. The player’s skills should be tested (usually, problem-solving skills). There should be an approximation of “winning and losing,” or at least the possibilities of success and failure.
  2. The challenges should be authentic. Victory should not be given to the players; the GM should not “go easy” or pull punches.

I also particularly like the focus on “player skill” opposed to “character skill”.

Two minor quibbles from me.

  1. If you’re going to re-contextualize RPGs as “play pretend with friends” I think there’s a lot of stuff that pre-dates D&D in 1974, that add structure to “play pretend with friends” , such as the Brontës’ activity to “pass the time”. Totally agree that generally the more “modern” conception of RPGs has a lot to do with D&D.
  2. I don’t like “Safety Tools” being labeled as creative constraints. I think that does them a disservice. I believe I understand the sentiment. To me, safety tools, are often just formalized explicit consent procedures. They are their to make sure everyone agrees and wants what happens to happen. Yes, I can understand how some see it as “limiting creativity” but, to me, that seems like a poor way to label them. Because of their nature as explicit formalized consent tools, I also understand how some folks don’t understand why they are needed cause they only play with trusted folks and already have an easy time getting enthusiastic consent from all the folks they play with. So for me, these tools are way more useful in any situation where you might not know everyone from kindergarten or don’t necessarily have that deep trust and it’s harder to have those conversations about consent for any number of reasons.

Glad you liked it! I realize it takes some time to build to where it’s going, but I didn’t want to split “it’s all creative constraints on play-pretend” and “here’s the foundations of the OSR” into two separate posts. The arguments feel too connected.

I definitely want to look more into the Brontës stuff! When I say “creative constraints,” I mean “constraints that aid you in being creative,” so maybe I ought to have clarified that — but otherwise, point taken that my argument is a little shaky in the “safety tools” part.


I like this a lot. I had some thoughts, but as I was typing it out I found out that I was misinterpreting you. Still, I wanna thank you for this post since I’ve been thinking about some of these things on and off for years now, and more recently as I’ve been thinking about the difference between story-games and more traditional games.

On another note completely — and unrelated to any of the content of your blog — I really like your formatting with the little numbered headings. I love it. I rarely have time to read blogs, so having a way to know where I left off when changing trains or whatnot is really helpful.

1 Like

Numbered headings is another habit from poetry, haha. It’s a nice way of dividing things up without having to come up with header titles, which can be a little distracting from the body text, I think.

1 Like

I like the analogy of poetry. I think that probably is a bit more fitting than we realize

Nice post - while I’m not an FKR fan (I like exploration and think it needs mechanical limitations to create tension), I think your distinctions are well founded. Having been through and deeply involved with the OSR scene from about 2011 - 2017 or so it’s close to my interests and profoundly informs my own design.

So the legacy of the OSR is something I’ve been thinking about a fair bit lately – it seems there’s so much knowledge that’s currently being lost or ground under currently from it’s era. Whole chunks of carefully hashed out design are dismissed either in the name of nostalgic reactionary orthodoxy or a desire for lighter and lighter play (often with shared, improvisational narrative or procedural adventure generation). I personally find myself in an odd procedural corner of the post-osr, wanting both the rigor of fairly robust mechanics and procedures to save referee time/attention and the freedom of aesthetic/setting design and openness to rules change. I don’t really feel at home in nostalgia distorted, gray stone and orc haunted corridors of the old school reaction, /TG/, or long-standing early OSR forum types or the nebulous weird fantasy hex and point crawls of .itch pamphlet adventures. I appreciate the emphasis on procedure and exploration (though not the orthodoxy of unchanging 1970’s war game mechanics) of the first and the openness and creativity of the second (though not its dismissal of anything with greater rules density then a single universal mechanic or its insistence that referee virtuosity is shown by improvising everything).

Anyway, that were I’m at - not loving the fractal profusion of one-shot focused ultralights, and not enjoying the revisionism of the OSR into AD&D RAW.


I’ve read at least one of your posts, “Classic vs. the Past” — it seems like you’re working in an interesting space. By “mechanical limitations to create tension,” I assume you primarily mean inventory management, light tracking, that sort of thing?

I suppose you might put me in the latter camp. I do still prefer a referee, a prepared (rather than improvised) adventure, and some simple combat rules, but inventory management and time tracking have always felt to me like a distraction from what I find fun in games like these.

Can I ask for more detail on what you thought of my characterization of the OSR approach, particularly the “don’t go easy, don’t pull a punch, don’t contradict what’s been established” part?

Can I ask for more detail on what you thought of my characterization of the OSR approach, particularly the “don’t go easy, don’t pull a punch, don’t contradict what’s been established” part?

I mean that seems like a reasonable way of describing some aspects of some kind of “OSR” play. I think it’s always hard to do so though, and hard to define “OSR” because there are many versions of it. I suspect that late/P OSR play is less concerned with these sorts of play style choices, which to me seem largely to exist to create predictable/consistent fictional settings and establish referee neutrality and trust. The OSR spanned 15 plus years and was frequently very much about idiosyncratic and individual play style choices. It also had a variety of movements within it. Part of the reason definitions of it fail is that they tend to aim to define clear shared principles, when I suspect a lot of these things evolved and that the OSR is best examined as a historical scene. Certainly post 2016 or so when LotFP was really starting to professionalize and other projects like Hotspring Islands or MotBM began to focus more and more on OSR as a brand identity core gameplay and design principles started to diverge rapidly as different studios and auteurs wanted to solidify their system and their unique aesthetic while retaining the OSR as a sort of larger parent brand (or trying to claim it as there’s). This still seems to be the major way “OSR” is used today. Even before that though the distinction between the “revival” play style of the early forum driven OSR and the “renaissance” version of the G+ OSR created plenty of friction.

By mechanical limitations I generally mean the whole pile compared to something like FKR - for exploration based play this means encumbrance, supply limits, turnkeeping and such. I think without them you quickly lose a lot of what makes exploration (in the sense of navigation of a fantastic space as a puzzle) fun. Likewise when the only threat to risk to the PCs is combat and/or HP loss it’s much easier to get stuck in a combat game or to feel compelled to design adventures with more symmetrical encounters.

My suspicion is that in designing the earliest D&D editions, there was an expectation that the players would be familiar with war game style movement, that they’d be interested in logistics and such (also that parties would be more 10 - 20 PCs/henchmen for skirmish style combat, sessions would be 8 hours long etc). That many people outside the war-gaming community picked up the game lacking these expectations and interests almost immediately created tension. That the original rules (and AD&D, and B/X) all have either confusing, complex or even nonsensical mechanics (I really hate AD&D’s encumbrance system which meanders about multiple texts and is somehow in lbs and coins with very low weight limits determined in multiple ways) meant these subsystems were quickly abandoned by most groups and sort of hand-waved even by the war game ones (what is a bad of holding except a way not to worry about encumbrance). To use them, they need better systematization/proceduralization and I think more intentional mechanics (as well as more streamlined ones).


Like Gus, I was “around” during the nascent times of what would become the OSR (back when it was a handful of blogs/forums…eventually accreting somewhat in the G+ days) and found it useful to locate people who were still having fun and talking about those older games I still preferred. I wasn’t as much into the retro-clone phenomena (I had my Rules Cyclopedia, and didn’t really need new rules) but it was very interesting to see the types of hurdles and perceived faults other Referees and Players were encountering and trying to fix. Some good stuff came out of those conversations, mostly in terms of the re-emphasis on procedures and development of new procedures (yet another House Rule for Thieves is always fun and interesting to discuss, but that’s less about the game and more about “your table” to me, so I tend to prefer those kind of things happening “in play” for this reason).

But it’s the procedures that seem more universally applicable to play for me, and I’d hazard that sticking more rigorously to these procedures is one of the things that contributes to the “pull no punches” mindset, and by contrast, hand-waving or ignoring some can have a very powerful impact on play. If you choose not to use Encounter Distance/Reaction Rolls/Morale, or replace those with just “a ruling” too often, I think it’s likely to generate some subconscious bias creeping in eventually, to say nothing of losing some of the wonder of being surprised which is what I’ve always adored about them. At best it might just cause your game to live up to that “perceived lethality” thing whilst tacitly encouraging a potential combat-as-solution mindset, which are generally things I’m not trying to foster.

What struck me most about FKR when I first started seeing it bubble up was how much it reminded me of things we’d done ages ago. We had a thing called “walking D&D” we would play when it wasn’t convenient to roll dice or track things like HPs/equipment. There weren’t any rulebooks to consult for to resolve/clarify situations to adjudicate something like Spell Effects, etc. Very little Rules but very heavy on the Rulings. Coinflips/Paper-Rock-Scissor/Origami Salt Cellar-Fortune Tellers all occasionally served as randomizers, but even these were sometimes just eschewed in the interest of moving play forward. It was still that central “conversation” procedure but just without some of the other inputs generated by procedure/resource management/rules interactions that could do some of the “heavy lifting” to generate those surprising/interesting situations for us to work with :slight_smile:

But I still think fondly about that type of play, and some of it has probably colored how I still run games to a degree. I’m probably more of a proponent of things like team brewing/shared world-building and assuming character competency than some other Referees working with the older rulesets for this reason. FKR can definitely still lean heavily into procedures, and resource management can be centralized as part of play in those games. I think FKR will always have a more Invisible rulebook and that’s not always an inherently bad thing. I think it’s just a matter of shifting weight from one foot to the other in most cases.


When I wrote that the OSR approach takes pressure off the GM, I wasn’t sure whether that was just a feeling I had or if I was really onto something. But now I’ve run into a couple things:

First, a post on @GusL’s blog that reads:

The lack of overarching story also liberates the GM who is no longer burdened with the task of directing the characters and players through a series of narrative peaks and valleys, balancing risk and reward to make sure that the adventure is properly plotted; instead the GM becomes only set designer and judicious arbitrator charged with providing interesting information about a world full of puzzles (combats, mazes, traps, secrets, negotiations and strange effects or artifacts) while responding fairly to the players efforts to unravel them all.

Second, in his interview with Sam Doebler on the Bastionland podcast, Chris McDowall spoke about why he didn’t include difficulty modifiers for Saves in Into the Odd:

I really hate having to set difficulty classes and target numbers, and I hate asking GMs to do that as well. Somehow, that feels very complicated to me, very high-pressure, like I’m having to pick the right number. Whereas if you’re asking me to just describe the consequences or describe what’s going to happen if you succeed, it feels a bit more natural, and it feels more like just a natural extension of the fiction.


I enjoyed this article


I have a more recent blog post that expands on and sums up the material from the post in this thread and others I’ve done: The Four Channels of Creative Constraints on RPGs

1 Like

I also enjoyed that one, though I disagree with you here and there I am here for acknowledging the plurality of play, rather than lumping everything into ‘system’ as is common.

1 Like

I just now read this blog post, and it’s really inspiring. I take a slightly different direction, which will inspire my future designs. So, here’s what your poetry analogy got me thinking about:

  1. The GM of the a game (and to a lesser extent the players) are writing poetry. It can be free-verse, but it does not need to be.
    Ideally, I want a GM who uses my rules to use more and more free-verse as they become more experienced. I expect the players to depend less on standard-verse rulings as they become more experienced, but to not be disappointed when they get standard-verse instead of free-verse.

  2. The game has a core mechanic, or several core mechanics / procedures, which should be considered the language the poetry is written in. So, for D&D variants the d20 being the core resolution mechanic is a part of the game language. 5E adds advantage/disadvantage to that language, 3rd used typed bonuses, other setups exist. The core resolution mechanic is not the only thing that’s part of the game language, but most rules aren’t core to the game language.

  3. The game then has “rules” which are designed to work with this language. These are that New Book of Forms. If you always follow the rules for your language, you get decent results pretty consistently. Once you know them well, you know how to use parts of those rules without always adhering tightly. This is like how slant-rhymes are as good as full rhymes when used well, but people who don’t know how to make real rhymes probably can’t make slant rhymes either.
    It’s important to note that these rules are NOT entirely tied to the language. Consider poetry again. Sonnets originate in Italian but are a form in English too. The use of internal line-rhythm and alliteration in Beowulf and other old nordic poetry is a rule of that language but one useful in English. Haiku is a Japanese form but it can be used in English.
    A ruleset that explains how to run stealth can be useful in any game-language, but it will need translation. When people write Haikus in English, they generally up the syllable count, and will often even break from it slightly if the words in English feel more natural that way. This because English doesn’t match Japanese syllabalization. Similarly, if I decided to use clocks from Blades in the Dark in a D&D-style game with its d20 language, I would probably adjust them, but I wouldn’t need to throw them out entirely.

So, what’s the point of that rambling?

When I’m writing rules, I need to keep in mind that in their most valuable form, they are to be aids for a GM making free-verse poetry. To do this, they must work reliably when used cleanly, and they need to be fairly explicit. However, they need to be accompanied by indications about how to violate them.

To go back to poetry, imagine I’m making rules to teach slant-rhyme. My rules MUST explain rhyme correctly. If the reader doesn’t know what a rhyme is they won’t have anywhere to start from. However, if my rules ONLY explain what a perfect rhyme is, they are unlikely to invent slant-rhyme on their own, and even worse might think they screwed up when they occasionally do deviate from my strict rules.

This is all very esoteric, but I at least found it useful to write this, so hopefully other people find it useful to read it.

So, in general, thanks for a great blog post.