Here are my thoughts on game systems, and what I’ve learned about my taste in RPGs.

Apocalypse World (2nd ed)

Apocalypse World, and other Powered by the Apocalypse games, describe themselves as fiction-first games.

The way they mechanically achieve this is having “moves” which trigger in response to your character doing something. The book says that you shouldn’t look at the list of moves and decide what your character does, rather you should do what feels most natural for your character and then, if it meets the conditions of a move, dice get rolled. If it doesn’t meet the conditions of a move, you resolve it narratively.

I have three main problems with how the game works out in practice:

  1. The moves are super-broad, and I’ve had multiple situations in my game where all the players have different ideas about which moves are being triggered, which makes us stop the narrative and discuss it. That’s definitely not how the book describes them working. So in practice the system rewards you doing exact what it says you shouldn’t do (look at the list of moves to decide what your character does next) because if you don’t, everyone can agree that a move has been triggered, but not agree which has been.

  2. When a move is triggered, it usually calls for rolling dice. All dice rolls in the game are 2d6 + stat, with no difficulty modifiers. A 10+ is a full success, a 7+ is a partial success. So characters succeed at almost everything they try, regardless of the situation. Furthermore, there are only 5 stats, so characters feel really… I’m not sure how to describe it… lacking in depth? Successes don’t feel like they’ve been earned by the actions of the player, they just feel expected.

  3. It’s not just the players that have moves, the GM has moves too. They’re not suggestions, the book says that Apocalypse World calls for a particular style of GMing, and these moves are it. Codifying good GM behaviour sounds good in principle but, in practice, there have been times where I’ve wanted to resolve something non-narratively (eg, “there are three men approaching the junkyard, roll+sharp to see if you find the doo-dah you’re searching for: 10+ you find it and get away before they arrive; 7+ you find it and are loading it into the car when they arrive; 6- they arrive as you’re knee deep in the scrap heap, digging it out”) but if there’s no move for that situation, I can’t do that. I have to just pick an outcome myself, which feels way more arbitrary than a skill check.

The sessions are fun, but feel like they’re fun despite the system, not because of it.

Ultimately I ended up ending the campaign I was running early. I had started running a Call of Cthulhu game for a different group, and realised how much less fun I was finding Apocalypse World. On reflection, the main difference between the two campaigns was the story: I didn’t feel like my Apocalypse World game was building up towards a conclusion. The Apocalypse World book suggests the system works best for shortish (12 sessions or so) campaigns, naturally building to a head, and ending after some big moment. But after 7 sessions, I didn’t feel like that was happening at all. There were long-running plot arcs which could be tied up, but I saw no reason for one of them to be the end of the campaign. And not having any idea how a campaign will end 5 sessions before it’s supposed to is not a good feeling.

It also doesn’t help that, in certain online communities, whenever someone says they don’t particularly like PbtA games, people come out of the woodwork to explain how the OP must be playing it “wrong”, because if they were playing it “right” they would like it. Some of the fans can be almost cult-like in their devotion.

Call of Cthulhu (7th ed)

Call of Cthulhu is an investigative horror game. It’s also the first system I ran which, in hindsight, may have been a mistake because I couldn’t do a lot of it justice.

The system reinforces the horror theme by having sanity mechanics, where your character goes slowly insane as they’re exposed to the awfulness. This mechanic may not be for everyone, as it deprives the player of agency on occasion.

My main problems with the system are:

  1. There’s a very wide range of skills. On the one hand, this is good, because it means it’s hard for one character to be good in every situation. It encourages working together to solve problems. On the other hand, it means you can easily spend points in character creation on a skill which just never comes up, which is frustrating. Furthermore, it can be difficult to justify why such diverse characters know each other.

  2. The mechanics for chases are really clunky. I can see why they wanted to introduce them—in a horror game, running away is often a sensible thing to do—but I don’t think it was done particularly well. I think in my campaign we gave up on the chase rules at some point. If I were aware of D&D 4e’s “skill challenge” mechanic (the party has to accumulate x successes before y failures, using any skills as long as you can justify them), I’d have used that for chases instead.

  3. A lot of mechanics require weeks or months of in-game time: for example, recovering from insanity in a sanitorium, being healed from the brink of death, or studying a mythos tome and learning the spells within. This works if you’re running fairly disconnected scenarios where months of in-game downtime can be explained away with something like “and you return to your normal life, until you get a creepy letter from your uncle 3 years later…” but it doesn’t work so well in a campaign where there’s always the next lead to follow. I ended up house-ruling that you can just do those things on a week-long ocean voyage.

There was also a problem that, being an inexperienced GM, I was pretty bad at ensuring players got clues without railroading them, and pretty bad at knowing (a) when a skill check was appropriate (as opposed to an automatic success or failure) and (b) how to handle failure (beyond just saying “oh, it didn’t work…”). But those are problems with me, not with the system.

I particularly liked the skill check mechanics: a d100 roll-under system, with three task difficulties, and an advantage/disadvantage system. It means you have a lot to think about when deciding what should be rolled, but it’s easy to come up with some rules-of-thumb. I graphed all the types of roll in a memo.

City of Mist

A PbtA game, but with elements of Fate too. It’s a noir detective game, set entirely within “the city”, where all the legends are true. Most people are fully human but some, like the player characters, have felt a myth awaken inside them, granting them power… at the price of needing to keep up a constant balancing act with their humanity at stake.

Characters and situations have “tags” (like Fate aspects) and the basic die roll is 2d6+power, where power is derived from the tags which help or hinder you. There are mechanics to introduce or change tags, and tags can be temporarily used up to give increased bonuses. This feels much more satisfying to me than the 2d6+stat resolution mechanic in Apocalypse World, as it means difficulty is context-dependent. We didn’t seem to have many discussions about which move applied to a situation, it was usually pretty obvious.

In general, City of Mist feels like a PbtA game that does PbtA better than Apocalypse World does.

I’ve only played this system, not run it, so I don’t have any opinions about that side of it.

Fate Accelerated

A generic system, which takes a fairly abstract approach. The game mechanics are based around creating, manipulating, and using “aspects”: aspects of PCs, aspects of NPCs, aspects of the location, the setting, the game…

You deal with these aspects using a metacurrency called fate points. For example, the GM can encourage you to do something troublesome, like “your character has the ‘no respect for authority’ aspect, so I’ll give you a fate point if you insult the king”, if you do that, you get the fate point, and can later spend that fate point to gain a bonus in a roll, introduce a new aspect, and so on. The fate point economy is designed to lead to good things and bad things naturally happening in the narrative. Sure, you insulted the king and got locked up, but you then did an incredible job escaping from jail, and in one of the other cells you found someone we’re looking for!

It’s a bit tricky getting our heads around this style of play, but it’s working fairly well. It’s definitely not a system I’d recommend to someone new to RPGs, though.

I’ve only played this system, not run it, so I don’t have any opinions about that side of it.

Golden Sky Stories

Golden Sky Stories is a pretty unique system in that it’s got a really small scope. It also doesn’t use dice or randomisation at all.

In Golden Sky Stories you play helpful animal spirits, helping people in a small town with their problems. These problems can be as small as “my friends moved away and I’m lonely”. There are skill checks, but you resolve those by spending points of “feelings” to boost your attribute high enough to pass, it’s totally deterministic.

The game is largely about roleplaying. The mechanics are pretty simple, and the system relies on a few different metacurrencies. The main way to gain more metacurrency is to:

  1. Do cute things. Any player can award any character with a point of “dreams” at any time. The player doesn’t pay a dream to do that, the new dream springs into existence. The balancing factor is that you can only award one dream per thing.

  2. Form bonds with more characters (or use dreams to strengthen existing bonds), as bonds give you “wonder” and “feelings” (the other two metacurrencies) at the start of each scene.

It’s a nice system for one-shots, though it does absolutely rely on everyone getting into it and roleplaying. If you approach it from the angle of trying to solve the problem efficiently, you’ll get through a scenario in no time at all, and probably won’t have any fun.

I don’t think it really works for a campaign. I was a player in one campaign and we stopped after three or four sessions. The game doesn’t have much depth or character progression to it, and I feel those are essential to give a game longevity.


This is someone’s unofficial Pokemon RPG.

Honestly, I don’t really have anything positive to say about it. I was in a campaign, but we decided to switch to Fate Accelerated after a while because almost every time the mechanics came up, they caused more problems than they solved, and we realised that some of our best sessions hadn’t involved the mechanics at all.

There wasn’t a core mechanic. Every aspect of the game had a totally different system. Shortly after we switched to Fate, a new version came out with some new mechanics for competitions: which was literally just a boardgame, and one which looked a lot like Monopoly. Glad we got out of that.

I’ve only played this system, not run it, so I don’t have any opinions about that side of it. Though the GM has often complained that the difficulty of making pokemon means he can’t really do random encounters.

Systems I’ve used in one-shots

Because of my limited experience, it’s hard to give any particularly well-thought-out opinions for these.

Ars Magica (5th ed)

My group started a campaign, but most of the players didn’t really like the system and we called it quits after one scenario. I do like feel of the system, so I want to give it another chance, but I’ll need to find another group first.

Blades in the Dark

I know this is really popular (both it and its derivatives), but I wasn’t too impressed when I did play it. But that could be because the GM was trying to make all the game mechanics show up in one four-hour session so we’d get to see everything.

Dungeon World

I think this one-shot was the first RPG I actually played. I didn’t get much out of it because I was so unsure of what to do, so I didn’t do much. I’d need to try it again to actually form an opinion on it but, given how I feel about Apocalypse World, I’m not sure I’d like it.

Monster of the Week

Another one I played when I was still pretty new and passive. But it’s a PbtA game, so I’m not too sure…

Systems I want to try

I collect free RPG PDFs whenever I see them, so I’ve got a lot. Here are the few which I particularly want to try.


Horror game which uses a jenga tower to build tension. That’s pretty much all I know about it, but it seems like such a neat mechanic that I want to give it a go.

Dungeons and Dragons

The prototypical RPG. A lot of people don’t like it, but it feels like I’d be missing out to not try it at least once.


A family of systems, rather than a system as such, with lots of setting-specific variants. It’s an investigation game but with less dice rolling than Call of Cthulhu. A lot of people seem to like the GUMSHOE game Trail of Cthulhu.


A generic system, like Fate, but much crunchier. The main unique thing about the system is that it uses custom dice to give a wide range of possible outcomes for any action. There’s success/fail, but also good/neutral/bad side-effects (both minor and major), which seems like a great way to make sure every roll pushes the story forwards.


An OSR-style game about being a newly-awakened demigod doing whatever they want. I like the sound of this because I’ve not played an OSR-style game before, so that would be a good learning experience, and also because it operates on a different scale to most games: dungeon crawling isn’t a challenging task for demigods, so the challenges are more things like “I want to found a cult”, “I want to turn this corrupt monarchy into a theocracy”, or “I want to build a new heaven for the souls of my worshippers”.


A Basic Role Playing-derived kinda-generic system usually applied to fantasy settings. It’s got a bunch of features I like: hit locations, lethal combat, and a gentle power curve which means things which are threatening in the early-game can easily still be threatening in the late-game.

Thoughts on mechanics

Before getting into specific mechanics, it’s worthwhile looking at how, broadly, I think a game system should work. I’ll characterise this in terms of GNS and stance.

I generally feel that the rules of a game should provide the “physics” of the game world, so I prefer simulationist systems over narritivist ones. If the games rules commonly boil down to “you decide based on whatever is best for your story”, events in the game just feel arbitrary to me. I like to feel that there is some in-setting reason for the way things happen.

As a consequence of preferring simulationist systems, I think ideally players will spend most of their time in actor stance00Note: that doesn’t mean acting, they don’t have to “do the voice”.

, using only in-character knowledge and motivations to make their decisions. In practice that’s hard, so I’m also fine with players spending most of their time in author stance.

I generally feel that director stance ruins immersion, and I don’t like that. A player might briefly adopt director stance on occasion, for example suggesting a detail of the scene which the GM hasn’t described, but that should definitely be the exception rather than the rule.

I strongly think that story should arise out of the actions of the characters, not because the players are actively reshaping the narrative through rules which encourage them to take director stance.

Now here are some thoughts on particular areas of systems:

  • Core Mechanic: I think skill checks are a good core mechanic. I like the possibility of characters having substantially different competencies, so I like systems with a range of attributes/skills broad enough so that no character can be good at everything. I also like the ability to account for situational factors increasing or decreasing the difficulty of a task. It feels much more real.

  • Failure: I don’t like dead-end failure being common, it can be ok sometimes, but normally failing a check should nevertheless advance the game somehow.

  • Structured Gameplay: Normally gameplay should be unstructured, with the exact order of events fuzzy. But sometimes order does become important, like in combat. I like the ability to drop into a structured mode where events can be ordered (eg, turns and rounds). I suspect I would like popcorn initiative as a way to handle this, but I’d need to try it out.

    A related mechanic is that of positioning. I’ve not played a grid-based tactical RPG, but I suspect I would find that fairly unwieldy. Genesys uses a few different ranges (“engaged”, “short range”, “medium range”, “long range”, “extreme range”), with rules for moving up and down levels, and that seems like a nice solution to the problem.

  • Improvisation: I like the ability for the GM to easily improvise in situations there isn’t a rule for. For example, asking for a check to decide how the game advances, even when there isn’t a rule saying “now you ask for a check”, should be an acceptable thing to do.

  • Metacurrency: I’m not sure how I feel about metacurrency. I suspect it works in more narrative-style games, and doesn’t work in more simulation-style games.