Game Master Responsibilities

I saw a recent discussion about why Game Mastering is considered harder than playing. One reason is that there are many, many responsibilities that tend to default to the Game Master. Different games and groups will have distribute responsibility differently, but it’s common that if nothing says otherwise, these fall solely on the Game Master:

Master the rules

  • Learn the rules well enough to do all of the following:
  • Teach the rules
  • Decide which rules to engage and how
  • When someone takes an action in the fiction that engages the rules, understand and flesh out the situation well enough to set consequences, target numbers, or whatever else the system requires
  • Arbitrate rules disputes and have final say on rules questions
  • Fully stat out and balance challenges
  • Understand the subtleties of the game system well enough to create a fun experience

Be a host

  • Handle the logistics of getting everyone together on a specific date and time to play
  • Provide a physical space to play
  • Provide or coordinate food
  • Provide a digital space to play:
    • Pick a text, voice, or video chat solution
    • Figure out what to use for dice, character sheets, and any other play materials
    • Get everyone set up and comfortable with the tools picked
    • Act as tech support
  • Check-in on other players’ comfort, energy, and engagement levels
  • Make sure everyone else is having fun
  • Indirectly figure out what kinds of fun other people are looking for so you can give it to them
  • Decide when to take breaks

Be a moderator

  • Set expectations about the tone, subject matter, and style of the game
  • Distribute spotlight time
  • Moderate the discussion – make sure people stay on topic and get a chance to talk
  • Facilitate decision making – check if the discussion is going in circles and figure out how to unstick it
  • Arbitrate non-rules disputes
  • Set or frame scenes: decide the where, when, who, and what
  • Decide when to end scenes
  • Introduce and model safety tools and check-ins
  • Initiate the group’s transition from socializing to game time
  • Take notes
  • Keep an eye on the time, especially if people need to leave at a specific time
  • Lead a debrief or retrospective at the end of the session

Convey the fiction

  • Learn the setting well enough to do all of the below:
  • Teach or explain the setting
  • Describe locations and environments
  • Play NPCs
  • Play out their agendas and/or tactics
  • Do voices
  • Play multiple NPCs in the same scene or conflict and make them distinguishable
  • Be entertaining: clever, dramatic, funny, etc.

Put things in motion

  • Provide the main source of forward motion for the narrative
  • Provide resistance or antagonism to other player’s actions
  • Decide results and consequences of actions
  • Decide long-term or off-screen consequences of actions
  • When people look at you expectantly, say something that keeps the game going

Set the table

  • Own all the materials for play: books, special dice, miniatures, etc.
  • Prepare and maintain all the aids and materials for play: reference sheets, character sheets, maps, miniatures, props
  • Prepare a scenario, mission, adventure, challenges, etc, ahead of time
  • Remember what happened last session, and all the previous sessions
  • Maintain logs and reference material for everything that has happened
  • Bring the energy that we are going to play this game

(2021-04-05: Updated to add more duties from Gerrit Reininghaus’s excellent post, Sharing the Cognitive Load )

2018 Retrospective

On the first day of 2019, I’m looking back at my year in gaming and blogging.


My regular gaming group had ten sessions of Blades in the Dark this year. I GMed two and played a crew member in eight. It’s been interesting seeing the variety of GM and player styles at the table, and the different approaches to session prep. I’m back to GMing next session, and figuring out a way to pick up all the dangling plot threads from my last time at the wheel.

I made it to two cons this year, and played a mix of board games and RPGs. I got to playtest a few RPGs: Stephanie Bryant’s Last Monster on Earth, MonkeyFun Games’ A Town Called Malice, and Daedalum AP’s Roar of Alliance. I always learn something from playtesting other people’s games, whether it’s about the game itself or the process surrounding it.


2018 is the year I decided to get serious about posting to this blog. I made 19 posts this year. My three most popular posts:

Blades in the Dark is my trendiest topic, followed by Ryuutama. Still, with a total of 29 posts and an average of 7 visitors a day (many of which were me), there’s not a lot of data to go on. If I were optimizing for visits I would go in for more Blades content. Since I’m currently one of three rotating GMs in a Blades campaign, that seems likely to happen anyway.

This year I also started paying my RPG Tax, an idea from Ray Otus. I am woefully behind, but I have drastically reduced the amount I spend at Bundle of Holding. I frequently fell into the trap where I wasn’t up to  reading all of my purchase, so I read nothing. For next year, I’m going to reduce my threshold for what counts as “read” to include quickly leafing-through and commenting on what jumps out at me.

At one point in November I was able to finish three posts in quick succession. Instead of posting them immediately, I scheduled them for advance posting. That gave me more time to work on my next post, which was much appreciated. I’m trying for a weekly posting schedule this year, but with a buffer of scheduled posts. This means that some of my RPG Tax reviews may go up well after I first read them, but hopefully this will be more maintainable.

Here’s to a great gaming 2019!

Blades in the Dark Resistance Roll Stress Values

Someone on G+ asked about the expected Stress values for Resistance rolls depending on the size of your dice pool, so I calculated them.

At 3 dice, on average you’ll spend just under 1 Stress

At 4 dice, you have a ~52% chance of not spending Stress, with a ~13% chance that you’ll actually gain back a Stress.

I first tried to tackle this in Anydice, but my skills are rudimentary. I wound up writing a Python script to count the possibilities for each Stress outcome.

import itertools

def stress_outcomes(dice):
    counts = {1: 0, 2: 0, 3: 0, 4: 0, 5: 0, 6: 0, 66: 0}
    all_rolls = list(itertools.product([1,2,3,4,5,6], repeat=6))
    for roll in all_rolls:
        if max(roll) == 1:
            counts[1] += 1
        elif max(roll) == 2:
            counts[2] += 1
        elif max(roll) == 3:
            counts[3] += 1
        elif max(roll) == 4:
            counts[4] += 1
        elif max(roll) == 5:
            counts[5] += 1
        elif max(roll) == 6:
            if roll.count(6) > 1:
                counts[66] += 1
                counts[6] += 1
    print("\n%d dice" % dice)
    total_count = len(all_rolls)
    print("\n".join([("= %d/%d\t= %2.1f" % (count, total_count, 100.0*float(count)/total_count)) for count in counts.values()]))

for i in range(3, 7):


3 dice
{1: 1, 2: 63, 3: 665, 4: 3367, 5: 11529, 6: 18750, 66: 12281}
= 1/46656       = 0.0
= 63/46656      = 0.1
= 665/46656     = 1.4
= 3367/46656    = 7.2
= 11529/46656   = 24.7
= 18750/46656   = 40.2
= 12281/46656   = 26.3

4 dice
{1: 1, 2: 63, 3: 665, 4: 3367, 5: 11529, 6: 18750, 66: 12281}
= 1/46656       = 0.0
= 63/46656      = 0.1
= 665/46656     = 1.4
= 3367/46656    = 7.2
= 11529/46656   = 24.7
= 18750/46656   = 40.2
= 12281/46656   = 26.3

5 dice
{1: 1, 2: 63, 3: 665, 4: 3367, 5: 11529, 6: 18750, 66: 12281}
= 1/46656       = 0.0
= 63/46656      = 0.1
= 665/46656     = 1.4
= 3367/46656    = 7.2
= 11529/46656   = 24.7
= 18750/46656   = 40.2
= 12281/46656   = 26.3

6 dice
{1: 1, 2: 63, 3: 665, 4: 3367, 5: 11529, 6: 18750, 66: 12281}
= 1/46656       = 0.0
= 63/46656      = 0.1
= 665/46656     = 1.4
= 3367/46656    = 7.2
= 11529/46656   = 24.7
= 18750/46656   = 40.2
= 12281/46656   = 26.3

Types of Fun and Blades in the Dark GM/Play styles

I’m one of the rotating GMs in an ongoing Blades in the Dark campaign. We started with two GMs, me and E., and last session a third GM, W., started. It’s been interesting to see the differences in GMing styles, how those map to Types of Fun, and how they’re supported by the system.

When I GM Blades, I focus on providing my players with the Fantasy of being competent thieves, and to a lesser extent, to give them a compelling Narrative that wraps up nicely within the session. The focus on Fantasy is a tricky balancing act: to showcase their competence, and to make the victory feel earned, there must be some amount of challenge. But if there’s too much challenge, the crew might fail the Score, which would violate the fantasy of being competent. I’ve realized that my Scores are often easy because I’m leaning too hard on the “they must succeed” side. My score prep method also grows out of this focus: the list of potential challenges allows me to throw them in as a pacing mechanic, and dial them back down when it feels like the Score should wind down and the players have faced enough adversity.

Blades as a system has mixed support for this mode of play. On the plus side, the flashback and resist consequence mechanics are all about showing the characters’ competence. The playbook special abilities all give unique opportunities for characters to shine. The Score as a main unit of play gives clear goals with a reset in between, so messing up one Score doesn’t necessarily mess up the overarching quest. It’s also handy as a pacing mechanism for making sure each session has a satisfying conclusion. The XP triggers encourage players to bring up backgrounds and relationships in a way that makes an interesting narrative.

On the minus side, every Blades action roll comes with the possibility of things going horribly wrong. Some of the players in our group come at the game with a focus on Fantasy or Challenge, and they try their best to avoid triggering action rolls for that reason. This can bog the game down with long discussions about which course of action has the lowest risk, and lead to some weaseling.

My impression of E.’s GMing style is that he focuses on providing Discovery and Challenge types of fun. He recently finished GMing an arc that hammered home the differences to me. The crew was supposed to steal some blackmail evidence from a relative of Lord Scurlock. We successfully broke into her house and didn’t find it, but found the portal to her magical bunker. We got a key to the portal and broke into the bunker, but still didn’t find it, but found other valuable possessions. So we bargained those back to her, and she revealed that her blackmail material was hidden in a house taken over by an eldritch horror. We’d picked up some random clues about it earlier but had been so focused on robbing her directly that we ignored them. For a game focused on Fantasy and Narrative, this would be a failure on the GM’s part. For a game focused on Discovery and Challenge, this was a failure on the players’ part. We didn’t investigate enough to discover the secret of the horror house, and instead jumped to conclusions: we failed the Challenge. Meanwhile, the GM stayed true to the facts established in his prep and the Challenge he laid out for us.

Blades has pretty good support for this style of play. For Challenge, the core Action roll mechanic ensures that there’s a constant incoming stream of obstacles, while the action ratings and playbook abilities give the players many tools to address them. All the mechanics that involve Stress make an interesting resource management challenge. On the downside, Challenge-focused players may feel Action rolls are too risky and avoid engaging the system. For Discovery, there are two pages on Gathering Information, a specific roll type for it, and various perks and abilities that give bonuses to it. On the downside, it leans on the players to remember to Gather Information in situations that may not immediately scream “you need more info.” And by necessity, it also relies on the GM to give good answers.

I think W.’s GMing style is closer to mine than E.’s, but it’s too early to tell. There’s enough overlap in the way these types of fun show up in a game that it can take several sessions to realize there’s a difference. If the obstacles are well-tuned, you can’t tell whether the session was focused on Fantasy of competence or Challenge. A Narrative-oriented game with secrets will give players a strong sense of Discovery as they find them.

I think the designer of Blades focused on Narrative and Expression. Scores, complications, stress, and vices all provide avenues to explore and reveal character. Most of the system is about getting characters into trouble so they have to answer the question of what they will give up to succeed. I think our Slide player’s preference falls closest to this, but they’re outnumbered!

RPG Tax: HexKit v2.0 and The Black Spot tileset

What is it?

HexKit is a tile-based hex mapping application designed to work with the Hex Kit Fantasyland tileset. It comes with one set of black-and-white tiles, with more (including Fantasyland) available for purchase. The Black Spot (, DTRPG) is a treasure-map-themed tileset for HexKit.

How did it get my attention?

I saw the original Hex Kit Fantasyland tileset posted on G+ and reddit, and thought the tiles were beautiful. Later, the artist launched a Kickstarter to create an app for working with the tiles, and develop more tilesets.

Why did I actually buy it?

I was GMing Ryuutama and potentially going to GM a hexcrawl of some kind, so the tiles and hex software looked useful. The Kickstarter also coincided with a sale on the Fantasyland tileset, which bumped me over the fence.

What are my first impressions?

The tiles are gorgeous. One of the selling features is how easy it is to lay down tiles with subtle variations. The software groups tiles into categories such as “forest tiles.” Each category holds variants with the same background but slightly different features, e.g. the forest tiles have a mottled green background and different numbers and placements of trees. You can select a category to paint with, which fills hexes with random tiles from that category. It’s effortless to make forests or grasslands that have more visual interest than the same tile repeated.

It makes me wish I did get to run that hexcrawl campaign.

What are my second and/or post-play impressions?

The Black Spot tileset is fiddlier to use because it has more tiles where the orientation matters. This was true of the river and coast tiles from the Fantasyland kit, but the Black Spot has both coast and beach tiles, two path types, rivers, and map travel lines. For the water bordering tiles, it also matters how many of the borders are land and how many water, but they’re all in the same category. Even a simple map looks pretty cool though:

Prepping Scores for Blades in the Dark

I’ve co-GMing a campaign of Blades in the Dark, and one thing I struggle with is prep. I know it gives you tools to improvise scores, but my improvised scores tend to be cakewalks as I struggle to come up with suitable obstacles on the fly. I’ve also had several sessions where events meant most of my prep went unused. But for the last session, I tried something new with prep, and it went swimmingly.

First off, it helped that the crew decided before the session what score to pursue. That meant I could focus on a single score instead of preparing two or three scores in the hopes the crew picked one of them. We ran the score selection between sessions as part of an email chain with a short interlude by voice chat, but it could also be done Ryuutama-style as a session closer. The score was to steal an artifact called the Tangle of Bones from the Church of Ecstasy, with a warning that a rival crew was after the same artifact.

The prep method I settled on is inspired by John Rogers’ Crime World supplement for Fate, and The Covetous Poet’s Location Crafter. The high-level summary is to divide the score into zones, then prepare possible obstacles for each zone.

For heists, Crime World describes three key concepts. The Score is your target, whatever valuable you’re trying to steal. The Box is what directly protects the target, and the House is the building or area surrounding the Box.

Start by detailing the target. Why does the crew want it? Who is the current owner and why do they want it? If anyone else wants it, why? Decide on the target, at least at a high level, before going to the next step. For the Tangle of Bones, I generated a random prompt for its power, which was “Create Illness.” Since the crew’s client was a demon, the rival crew was a cult worshiping a demon, and the Church of Ecstasy researches demons and immortality, I decided that a demon could use the artifact to spread a plague that caused infected people to fall under its influence.

Next, think about the Box. What protects the target? What prevents people from accessing the target? How are intruders detected? What prevents people from just walking off with the target? What are some troubles or weaknesses of the Box? Decide the high level concept for the Box, and write down several ideas for the other questions. My high concept for the Box was a secret lab under a Church of Ecstasy.

Then detail the House. Divide it into three zones:

  1. The public areas. For my Score, the public parts of the Church where regular services are held. I decided on Whitecrown so they weren’t breaking into the Church’s main stronghold, and because it’s next to Doskvol Academy, where some of the crew has history.
  2. The secured areas between the public areas and the Box. The private areas of the church where only priests are allowed.
  3. The threshold to the Box. The secret labs under the church.

Make a table with four columns, one for each of the zones and one for the Box. In each column, write out possible obstacles and complications for the zone. Since I was prepping a score with rivals, I merged zones 2 and 3 into one column and added a fourth one for the rivals. Optional: sort the obstacles from easiest or most likely to least likely.

If your group likes playing with maps–mine does–sketch some rough maps for each zone. I picked an unlabeled building from the Whitecrown map. Then I looked up a few real church plans for ideas, and roughed out the zones.*

In play, whenever it seems like the crew should face an obstacle or someone rolls a complication, look up the list for the zone they’re in, pick one, and cross it off. If you sorted the lists, you can roll a d6 and use that obstacle, skipping any that are crossed off. If time is running short or the session is winding down, stop pulling obstacles from the list and deal with what’s already established.

For this score, part of the background situation prep was finding a reason for the rivals and crew to break in at the same time. Since services at the main cathedral involve dissolving spirits, I decided there would be an exclusive spirit destruction service/party at the Whitecrown location, held in the private areas of the Church. That would shake up regular security patrols and mean strangers wandering in the private areas, making it a golden opportunity for both crews.

This is the final table I came up with, marked with the obstacles I wound up using. The crew came in by a different route than the spirit destruction party, so most of those went unused. They were also most interested in the labs area, so I used more obstacles from that area than others. Overall, the score flowed smoothly without the awkward pauses where I rack my brain for a suitable obstacle. The players were engaged, and felt they could visualize what was going on better than any previous score I’d run.

Whitecrown Church Public Areas Church Private and Secured Areas Secret Lab / Vault Rivals
  • Need invitations for exclusive spirit destruction event
  • Guard checking guest list
  • (Spider) personal rival is there, will recognize him
  • (Whisper) Rival Whisper is there, recognizes him
  • Spirit destruction event requires  interaction from group and professions of faith
  • Rival Spider notices the crew and starts calling attention to them at the spirit destruction party, e.g. volunteer them for participation
  • *Door to labs area is hidden in secret passage, need to know sequence of brick/book presses to get in
  • Need to find which lab has artifact
  • Need to find location of vault with more immediately sellable  stuff
  • Researcher going to nearby lab who’s familiar with everyone that’s supposed to be working here
  • Artifact container is bulky and obvious, if any guards see it they’ll want to know what’s up
  • *Lab is guarded by a hull that wants proof that you’re an authorized user of the lab to open the door (Pick 2+: smell, sound, sight, knowledge)
  • Hull remembers everyone who comes through (+Heat unless addressed)
  • *Two keys needed to open valuables safe ether clean chamber
  • *Keys are on opposite sides and need to turn at same time
  • Research notes are ciphered
  • *Artifact is contagious in the ghost field and needs special container for carrying
  • *Artifact is in an ether clean chamber, need to remove it into container without letting ghost field in (Like those biohazard / chemical chambers with the glove sleeves)
  • Entire lab is an ether clean room, airlock style so you need to suit up (and suits are bulky / hard to move in)
  • Vault has combination lock
  • *Vault is set up like safety deposit boxes, need to crack individually to get contents
  • Rival Whisper alerts the rest of cult that the crew is here
  • Rivals find crew Whisper due to pendant
  • *A guard knocked out by rival Cutter or rival Lurk wakes up as PC passes
  • Rivals found the artifact first and are making their escape
  • Rivals corner whoever has the artifact and try to get it:
  • Cutter – by force
  • Spider – by fast-talking / distracting
  • Lurk – by distracting  / stealth
  • Whisper – summon ghost?



I’ve decided to copy Ray Otus‘s resolution to pay an “RPG tax” on any new games bought: immediately read them and post commentary. The posts have this structure:

  1. What is it?
  2. How did it get my attention?
  3. Why did I actually buy it?
  4. What are my first impressions?
  5. What are my second and/or post-play impressions? (Optional)

When I have time, I’ll also go through my backlog of games to read and post about those.

Edit: More clarifications on the resolution:

  1. If it was a Kickstarter, I must post when I receive the final version from Kickstarter. Writing posts for drafts is optional.
  2. If it was a bundle, I need to write posts for at least half the products in the bundle. If there are multiple versions of a product, like an adventure module with versions for 13th Age and Pathfinder, all versions count as one product.

Glass Bead Game Randomizer

A while back, Rob Donoghue posted about a game you should play to become better at GMing. Created by HipBone games, it’s inspired by the Glass Bead Game that Hermann Hesse wrote about in his book, Magister Ludi.

In the basic game, two players take turn placing ideas on the board. Whenever they place an idea, they need to state connections between the idea they just placed, and any other ideas connected to that space. Whoever comes up with the most total connections wins.

Played solo, it’s a neat exercise in divergent thinking. But I always hit a block when deciding what to fill each space with. To turn it into a pure divergent thinking exercise, I created a randomizer that fills each space with an icon from Game Icons. I also added the ability to set a seed in the URL, so if you find an interesting layout, you can share it around. Every time you randomize, it updates the URL.

Try the Glass Bead Randomizer here

Haven Swirl, a One-Page RPG

This was from a randomly-generated prompt:

Atmosphere RPG: Rock paper scissors resolution for a game about Farm focused on Discovery

Haven Swirl

The world was broken long ago, and life has been hard since. Wandering, you came upon Haven, a handful of people eking out a living on what was once a farm. Everyone has a story similar to yours, but nobody remembers who founded it.

In this game, you alternate between playing a member of Haven, and the world.

It uses two kinds of Rock-Paper-Scissors resolution:

Group Resolution:

Choose a reference player (the rules will tell you how). Everybody makes a fist, then simultaneously throws Rock, Paper, or Scissors. Compare your results against the object thrown by the reference player. Depending on if your result is Win, Lose, or Tie, the rules will tell you what to do. The reference player always ties with themselves.

Single Resolution:

You play one round of Rock Paper Scissors against another player.

Win – you get what you want
Tie – you partly get what you want, or it costs you
Lose – you don’t get what you want, and it costs you

Create your character:

Pick a specialty (duplicates between characters are okay):
– Crops
– Animals
– Technology
– Buildings
– Fiber arts
– Food

Pick what you’re looking for now:
– Protection
– Redemption
– Respect
– Belonging
– Knowledge
– Love

Decide why your character was wandering.

Decide what order your characters arrived at Haven. The character who arrived the earliest is called the First.

Choose one other character. You had an encounter with them that made you decide to stay. It could be someone who arrived after you.

Create the Map

Haven has one main house where everyone lives, and a few other buildings for the various specialties.

In the center of some large paper, draw the common room of the house.

Each player, draw the room of the house you live in. Draw a new building, plot of land, or room where you practice your specialty. If one exists, you can extend it.

Draw a rough multi-armed spiral extending from Haven, with one arm for each player. Each player claims a sector of the spiral; they have first say over what happens and what can be found in that sector.

Now decide on starting resources. Use Group Resolution with the First as the reference player:

  • Win – there’s a surplus of a resource relating to your specialty. Draw it in or close to Haven, in your sector.
  • Tie – you have just enough to practice your specialty most of the time. Draw the resource most limiting you in or close to Haven, in your sector.
  • Lose – there’s a lack of a resource relating to your specialty. Draw something in your sector or in Haven representing or explaining the lack.


You will play out 12 months at Haven, each focusing on a different character in turn starting with the First.

Each month, do these three things in any order:

  • Have a scene with another character
  • Explore
  • Practice your specialty

Have a Scene:

Decide whether this is a flavor scene (just roleplay to develop character) or a goal scene (you want something from them).

If it’s a goal scene, roleplay until the outcome is in doubt, then use Single Resolution.


Decide what you’re looking for, which direction you’re going from the 8 compass points, and how far. The player whose sector that falls in will frame a scene where you encounter something related to what you’re looking for. Roleplay until you feel the outcome is in doubt, then use Single Resolution.

Practice your specialty:

Describe your goal for the month and how you practice your specialty to achieve it. E.g. Have enough food for the winter by canning all the fruit that was just harvested.

Use Group Resolution with you as the reference player

  • Win – the player’s character or sector benefited from the activity
  • Tie – the player’s character or sector benefited, but the other (sector or character) lost something or was damaged
  • Lose – the player’s character or sector lost something or was damaged as a result

Special: the active player’s result counts as a Win

Draw something on the map representing the overall result.


At the end of the 12th month, reflect on how the year has gone. Take turns reminiscing about something that happened.

Then Use Group Resolution with the First as the reference player to determine your outlook, either for this year or the next:

  • Win: Rosy
  • Tie: Cautious
  • Lose: Pessimistic

Each narrate an epilogue about the coming year based on the result.


  • Avery Alder’s The Quiet Year
  • Ben Robbins’ Follow
  • Becky Chambers’ A Long Way to a Small Angry Planet
  •  Hitoshi Ashinano’s Yokohama Shopping Trip

January Game-A-Month: Cat Cafe

It’s done! (For certain values of done.)

I could keep polishing it (and it could use more), but it’s a reasonably complete game.


Stroke the cats by clicking and dragging the mouse. When you’ve stroked them enough, they’ll be happy and give you points. But if you stroke too much, they’ll hiss and bite you! If you neglect them, they’ll leave unhappy and you’ll lose points. The game ends after one minute or three bites. Try to get the highest score!