2021-04-05 The things I learned

“Every Wednesday morning I wake up with a sense of dread, remembering that on top of everything else I have to do that day I somehow have to run a game in the evening. Every Wednesday afternoon I seriously consider calling the session off. But every Wednesday night I sit down and log in and everything actually goes fine.” – Failing better: a GMing retrospective, by Joseph Manola

Last month, Joseph wrote about the campaigns he ran and what he learned from them, and at the time I thought to myself that someday I’d do the same. Well, I guess today is that day.

The format used in Joseph’s blog post is that he has four short paragraphs for each campaign: what it was, what worked, what didn’t, and lessons learned. We’ll see how well this goes!

I got Das Schwarze Auge from my mom and ran some modules for her and her friends when I must have been around twelve (1985). I ran three or four official adventures. The first edition rules were simple and characters were hard to kill: they started out with 20–30 hit points; there was an active defence roll (instead of using armour class); armour was damage reduction; magic used “astral” points; the game had no clerics. What worked: I remember we had plenty of players. I was a kid running the game for three adults and older teenagers and as far as I remember, it worked. What didn’t: It didn’t last long. One of the adults wanted to stop playing, and then the campaign stopped. Lessons learned: simple rules enable kids to run games. I also remember the last adventure I ran, where the party discovers a gate to another world. When one of the players said that we should develop trade between the two worlds, something opened up in my brain. An open world! My first whiff of the sandbox.

I ran some adventures of my own devising in high school when I was fifteen and sixteen. We used Das Schwarze Auge at first, switched to AD&D 1st ed. and then to AD&D 2nd ed.; we still had no concept of a “campaign” – we stopped buying Das Schwarze Auge modules and we didn’t buy TSR modules, we just wrote our own and played them one after another (1988-1989). What worked: we played D&D, we played in English, and we played our own adventures. What didn’t: we didn’t know how to handle party conflict and when the thief tried to steal another player character’s gems, we didn’t know how to deal with that and that player left the group. Shame on us! We also discovered that one of us was a rules lawyer and a power gamer. Lessons learned: The upgrade treadmill is relentless. There’s always somebody who wants to upgrade to the next edition. In hind-sight we managed to deal with the hot mess of AD&D rules pretty well.

I ran a few sessions of a light variant of Mutant Chronicles game for fellow students at university (1995). What worked: we played in some maintenance tunnels, which was creepy and cool. What didn’t: no rules, no structure, no adventure; I don’t quite remember why it fell apart but it did. Lessons learned: a cool place to game does not make a campaign.

Kurobano and the Dragons. I ran a D&D 3.5 campaign after a long hiatus (2006–2008). It started out with me using M20, but we soon switched to D&D 3.5 even though I dreaded the size of the rule books. I successfully managed to integrate my own starting setup with Red Hand of Doom; later, I added more Paizo adventures. What worked: I learned the D&D 3.5 rules. What didn’t: I ended up disliking the battle map; I felt I was being forced to play a wargame week after week, against five other humans, and I wasn’t even good at it. So I had to add tougher opponents to counteract the lack of tactics on my end. Lessons learned: high level D&D is weird; some players loved it even though they were bad at keeping track of it: the multiple attacks, the buffs, it was a chore. All they wanted was to be super-heroes! The rules were failing us.

Golden Lanterns. I played in the Shackled City adventure path using D&D 3.5 with DM James (2007–2010). I started running this game for a few sessions and then I handed it over to James. It was my first Adventure Path and I loved the idea. What worked: James managed to make the fights work; and given the rare opportunities to play, he skipped a lot of filler material. What didn’t: Advancement was fixed. I felt that sometimes I got new abilities even though I hadn’t even used all the abilities I had previously gained. My paladin found a Holy Avenger in an armoury somewhere and that was weird. I guess James had decided that we needed it for balance. Lessons learned: Adventure Paths are great and terrible. They are great because they promise a story arc that takes you to fantastic heights. They are terrible because advancement, opposition, power levels, it all has to harmonise and the more freedom the game offers, the harder this is.

Hagfish Tavern. I ran the Rise of the Runelords adventure path using D&D 3.5 (2008–2011). Another Adventure Path. What worked: Again, the beginning was awesome. What didn’t: Again, the it started to fall apart towards the end. High level D&D is complicated because what works at lower levels doesn’t get abandoned at higher levels. Bless is still there. If you have three or four attacks like +16/+11/+6/+1 you still roll that last one because who knows, you might get lucky. Lessons learned: the next campaign I would run was going to be a sandbox for sure!

Krythos. I ran a small Burning Wheel campaign inspired by ancient Greece (2008). It was my first attempt at Burning Wheel. It turned out to be a short campaign of six or seven sessions, by my standards. What worked: It was my first online game using TeamSpeak. I was able to use Burning Wheel’s Bloody Versus (simple opposed rolls) and the Duel of Wits (social combat). What didn’t: The game didn’t “sing” for me. I think I’d need a lot more practice. Lessons learned: online gaming can work. All you need is voice chat. At the time we had no video, no dice roller, no virtual tabletop.

Sohn des Schwarzen Marlin. I played in this D&D 3.5 campaign set along the Dragon Coast in the Forgotten Realms with DM Peter (2008). It was a sort of nautical sandbox. I don’t remember us ever leaving that initial island, though. What worked: We didn’t have a cleric in the game and it worked. What didn’t: I don’t remember. The campaign ended after a handful of sessions; I don’t remember why, though. Lessons learned: An island hopping game with pirates might be interesting.

City of the Spider Queen. I played in this D&D 3.5 adventure set in the Forgotten Realms with DM James (2008–2009). It’s a long adventure; dare I say a short Adventure Path? We started at a higher level, if I remember correctly. Somewhere around level ten? What worked: I liked playing a cleric of Shaundakul; I liked creating a character with a strong bond to another character, a kind of tag team. I liked leafing through that Forgotten Realms setting book for D&D 3.5 in order to write up a backstory. What didn’t: I didn’t enjoy the high level fights. Lessons learned: Tag team characters from the get go are fun. Avoid high level D&D 3.5.

Grenzmarken. I played in a D&D 3.5 campaign with a homebrew setting by DM Peter (2008–2010). I played a dwarf wizard licking his toad familiar and doing other small disgusting things. This was a kind of West Marches game: almost every session was an excursion that ended back in town. By this time we had established a way to split the game between the session itself and writing on the wiki. Reports got longer. Things happened between sessions. What worked: sandbox exploration was great, with many monsters to defeat and small dungeons to clear. The wiki saw a lot more use. What didn’t: After a while it was a bit formulaic, travel to the destination, fight, fight, end boss, skip over the trip home because we were running out of time. As fights took longer, the non-fighting was relegated to the wiki. Lessons learned: I need my sandbox to be more interesting than monster fighting. The excursion structure needs some tweaking but it’s very promising.

The Alder King. I ran a D&D 3.5 campaign in Lenap of the Wilderlands of High Fantasy and later switched the game to the Solar System RPG (2008–2012). I used very slow advancement compared to the games I had seen with DM James since I didn’t actually want to reach those higher levels. I had decided to start the game with practically no high-level non-player characters, either. My thinking was: we’ll play the next campaign in the same setting and the surviving player characters of this campaign would be the high-level non-player characters of the next campaign. What worked: There were factions all over the place. Finding allies against Yarshag the lizardman and his giant wasp riding dragon-blood infused super-soldiers provided for a nice campaign arc. What didn’t: When the low-level henchmen formed a new party to play through the Caverns of Thracia, I didn’t enjoy myself as much. The dungeon and I did not quite agree. Lessons learned: my default implied D&D setting doesn’t work well with Greco-Roman sensibilities; I’m starting to suspect that by now I might not like playing through bought Adventure Paths, high level adventures, and big dungeons. Time to buy a lot less!

Die Reise nach Rhûn. I played in a heavily house-ruled Rolemaster campaign set in Middle Earth with GM Berni; we later switched to Legends of Middle Earth (2009–2010). What worked: Rolemaster was weird in a good way. What didn’t: the campaign was short lived. We tried switching to a rules light system but that didn’t work, either. Lessons learned: both rules heavy and rules light cannot save a campaign; I think we should have talked less about the system and more about what we actually wanted from the game.

Kaylash. I ran a Mongoose Traveller game which was later switched to Diaspora using a randomly generated subsector (2009–2010). What worked: The random subsector creation inspired a campaign. What didn’t: The Traveller rules seemed to imply that trading was the way to play the game, a bit like the old Elite computer game, except that nobody actually cared about trading; when using Fate, the game shifted away from trade to fighting zombies in space. Lessons learned: again, switching rules does not save the game if you don’t talk about what you actually want from the game.

Lied vom Eis. I played in a few sessions of an Song of Ice and Fire RPG with GM Berni (2010-2011). We rotated GM duties a few times. We had one player via video call. What worked: the new game rules worked for us. What didn’t: we had created all sorts of characters that then had no in-game reason to actually adventure together; at the same time we did not think of running our house like a stable of characters with people picking the appropriate ones, depending on the adventure at hand. Lessons learned: do not create parties that are too far apart, conceptually. You can make it work if you run your game like a fast-paced movie but it takes a lot of energy. I also remember once telling those players that I had come to play, not for the small talk. In retrospective, perhaps I should have invested into that small talk. We stopped playing together.

Desert Raiders. I played in the Legacy of Fire Adventure Path using Pathfinder RPG with GM James (2010–2011). What worked: Pathfinder worked a bit better than D&D 3.5 for us. The campaign also didn’t go all the way to level 20. Once again, the lower levels were super cool. What didn’t: I had created a mounted archer and got to use a horse maybe once. Lessons learned: talk to players if you don’t know how to incorporate their character concepts into the campaign. Also, never let anybody borrow your RPG books because now I’m missing one of the books from this Adventure Path.

Burning Six. I played in a six session Burning Wheel campaign (2011). It was a short game set in an Italian Renaissance town, or something like it. What worked: It was Burning Wheel. What didn’t: again, I feel that the lure of the Tolkien-style troupe of everything and everybody foiled our plans. Instead of being laser focused on a thing, we all created characters and dropped them into a situation. I think I was playing an abbess, there was an elf, and others… Lessons learned: at the time, I felt that Burning Wheel simply wasn’t for me; these days, after hearing the Shoeless Peasant podcast by Sean Nittner and Judd Karlmann, I feel that perhaps we just thought we could play Burning Wheel like we play D&D: create a bunch of weird characters based on the rules and it’ll all work out. Not so. First, laser focus on the kind of game we want to play!

Durgan’s Flying Circus. I played in a HARP game with a homebrew setting by GM Stefan (2011–2012). It was short lived. What worked: It was HARP. As I think back, I now get the feeling that often we played campaigns because we wanted to try new rules and had no other plans beyond that.

Mondschein Saga. I played a handful of sessions in an OSRIC campaign set in the Forgotten Realms with DM Peter (2012). It was a sandbox and Peter was searching for rules that were simpler than D&D 3.5. Unfortunately, he picked OSRIC, i.e. AD&D 1st ed., and those rules aren’t actually very simple. The printed book is substantial. What worked: the island hopping sandbox was once again improved. What didn’t: OSRIC was not smooth sailing. The players didn’t like it. Lessons learned: OSRIC and AD&D are not rules light. The island hopping sandbox can still be made to work.

Ymir’s Call. I played in a Barbarians of LemuriaCrypts & ThingsAdventure Conqueror King campaign with DM Florian (2011–2013). It was a frost sandbox. We were people in some sort of polar region and there was trouble all around us. What worked: I loved it. There were many places to visit, people to talk to. What didn’t: The rules for Barbarians of Lemuria did not convince me. On the one hand, they were too simple, and on the other hand, spending that point to mow down dozens was too weird. Crypts & Things was better, but still very much on the simple side. Lessons learned: I really started to love the wilderness sandbox. Just make sure there’s plenty of stuff to do wherever players go. As for the rules, I started to realise that I liked something somewhere between Crypts and Things and Adventure Conqueror King.

Ritter von Salisbury. I ran the Great Pendragon Campaign for the years 485 to 510, when Arthur pulls the sword from the stone (2012–2014). What worked: we had a rough campaign outline and we loved it; the traits that sometimes make characters do things the players did not expect. What didn’t: character creating took a long time; the deadly battles were frustrating if your character bit the grass; the inability to actually influence the outcome of these battles took some getting used to; you could build infrastructure but they had no effect on the game; the winter phase felt like an integral and badly designed part of the game. We always wanted something important to happen in winter but the rules did not deliver. We should have just narrated it. Lessons learned: There’s a cool, simpler game with those traits hidden somewhere in those rules.

Karameikos. I played in a Labyrinth Lord game set in Mystara every second Monday evening (2012–2015), by @oliof. We played B10 Night’s Dark Terror and it was very cool. What worked: the siege of Sukiskyn was fantastic; fighting vampires felt good. What didn’t: I’m not sure how the campaign ended. We arrived at the hidden valley and then… I don’t know. Lessons learned: B/X D&D is the level of D&D I like. After some initial enthusiasm regarding the Mystara setting, I’m much less enthusiastic about the incredible emptiness. Large maps with many empty hexes simply don’t inspire me as much.

Berem and Beyond. I played in an Adventure Conqueror King campaign with DM Florian (2013–2015). It was another sandbox, except no longer up in the icy north. What worked: I liked the sandbox; there was also a second group playing in the same area and I remember at least once we came upon a little mausoleum that had already been plundered. What didn’t: I saw Florian improvising a lot, rolling up encounters, and I didn’t like it. It broke my immersion because I no longer felt we were exploring a “real” place. It was being made up as we moved through it. The ACKS upkeep and other economy rules did not interest me. It felt like homework. Lessons learned: ACKS is too involved for my taste. When I run a game, I make sure to use a screen in order to better maintain immersion. If the players stray of the prepared material, I want to maintain the illusion for as long as possible.

Montag in Zürich. Various people interested in one-shots got together and experimented with new systems (2009–2015). What worked: I never again played so many different games as back then. It was amazing. What didn’t: eventually things broke down as people no longer volunteered to run new games. Lessons learned: there are a gazillion games out there. Also, not many of them seem to be made for the kind of long campaigns I like (50+ sessions). What I totally learned was to present a game, hand out characters, and run a game in 2½h.

Fünf Winde. I ran a Labyrinth Lord game set in the Wilderlands of High Fantasy with a big dash of Planescape and Spelljammer on every second Tuesdays (2010–2017). What worked: plane hopping, B/X + house rules, building projects to spend gold and enrich the setting; references to the old player characters from the Alder King game. What didn’t: I had placed a few mega dungeons in the campaign setting but the players didn’t really want to go there; Planescape looks cool but if you’re actually looking for adventure material in the setting boxes, there is very little; the same is true for Spelljammer. Lessons learned: playing multiple campaigns in the same setting is a good idea; visible changes from one campaign to the next is great.

Wilderlande. I ran a Labyrinth Lord campaign set in a Points of Light campaign setting for my best friend and his three kids for two hours on a Friday evening every four weeks (2010–2018). What worked: one of the kids is my godchild and I managed to stay in touch for all these years. What didn’t work: I was unable to have the spark jump. As the older kids turned seventeen and eighteen, they showed up less and less. None of them wanted to pick up the referee mantle. Lessons learned: kids can play Labyrinth Lord; when they’re five and six, they might just sit there and draw pictures, making the occasional decision for their character, but as they grow older, they simply absorb the game. There’s no need for special games for children. Children don’t want toys made for children. Children want toys made for adults. They might ignore a lot of the rules, just like we did when we were kids, but that doesn’t prevent them from wanting the real deal.

Greyheim. I ran a Labyrinth Lord game around The Castle of the Mad Archmage (a megadungeon) on every second Wednesdays (2015–2018). It was a big dungeon, with a computer-generated wilderness map surrounding them. What worked: we played until we got down to level seven of the dungeon. We managed to invest the riches retrieved into infrastructure: a castle was being built soon after the campaign got going; we had a construction site, masons, carpenters, guards, a palisade, a quarry, and on and on. What didn’t: the wilderness map wasn’t populated with enough monsters, tribes, factions, and all that. We did have orcs and ogres and treants, and evil elves and necromancers, and whenever they were encountered, the game was great. I should have added a lot more at the beginning of the game.

Mondschein Saga. I played in a D&D 5 game with DM Peter, with no regular dates (2018-2020). This was a reprise of the earlier campaigns. Peter had refined his concept: islands, factions, sandbox, links to the rest of the Forgotten Realms (e.g. a visit to Baldur’s Gate), no single safe haven but a variety of safe harbours, more dungeons, and a group that took to D&D 5. What worked: Peter’s sandbox gained even more depth; a less formulaic approach. Player Rafael taught me the value of well written session reports to show character development and I started taking to it. Some of my best writing for role-playing games happened (in German, translate using DeepL if you’re curious): Der Wert eines Zwerges (talking at the campfire), Caer Corwell auf dem Opferaltar (burning down a city), Der Feuerzirkel von Rottesheim (epilogue after the party lost the last fight at the end of the campaign). What didn’t: I still have no great love for D&D 5, but playing with just the Basic Rules for D&D 5 actually works. I like that! Lessons learned: as a player, writing vignettes on a campaign wiki is great (or posting them on whatever medium you use to share writings for your campaign).

Rasiermesserküste. I ran a Halberds and Helmets game using Razor Coast on every second Wednesday (2017–2020). I was looking for a nautical sand box, islands and pirates and all that. What worked: a nice campaign with cool ideas using were shark and cultists. What didn’t: the book tries to square the role-playing book circle, it has everything – events, maps, dungeons, characters, prep sheets, and on and on. It’s huge. It’s dense. It takes a lot of work, I think. Lessons learned: try to run the next game without using any books and without big plans that need to be prepped.

Die dampfenden Dschungel von Chult: I played in a D&D 5 game with DM Peter, with no regular dates (2020-2021), via Skype. Same deal as before: I used the Basic Rules for D&D 5. Instead of a fire wizard I played a rogue and it was great. What worked: the mix of sandbox, interesting locations with smaller dungeons, the ruins of Omu, the big final dungeon. What didn’t: if at all, the big dungeon at the end was a bit too long, but just barely so. It was still a great game. Lessons learned: the D&D 5 sandbox still works! Yay!

Die drei Wälder. I ran a few sessions of Just Halberds, with no regular dates (2020), for a bunch of kids, via Zoom. What worked: it was a play test for the Just Halberds rules. What didn’t: I wasn’t really able to get into a good 2d6 groove and eventually the game just petered out. Perhaps the kids lost interest; perhaps gaming remotely using Jitsi, Zoom, and friends, isn’t made for larger groups of players. Lessons learned: for future games, I’m going to try and limit the number of players to three.

Die Zeit der Waldbrände. I’m playing in a Humblewood campaign using D&D 5 with DM Lars (2020–now), via Jitsi. What worked: D&D with anthropomorphic animals works better than I expected! The game is still ongoing, so it’s too early to talk about the things that didn’t work or the lessons learned.

Der Fluch des Stradh: I am playing in a D&D 5 game with DM Mircea, with no regular dates (2021–now), via Zoom. I’m playing a fighter that picked the Skilled feat to pick rogue skills. So, a rogue without sneak attack, haha. I’ve started using the D&D 5 Player’s Handbook, at long last. I guess it works for me because I’m playing a fighter. What works: the mix of interesting locations with smaller dungeons everywhere. The game is still ongoing, so it’s too early to talk about the things that didn’t work or the lessons learned.

Tau Subsector. I’m running a Classic Traveller game, with no regular dates (2021–now), more or less weekly, via Jitsi. I used a ton of random tables to provide even more details for the generated subsector and now I’m hoping that one thing leads to another, with not too much prep required. The game is still ongoing, so it’s too early to talk about the things that didn’t work or the lessons learned.

Lessons learned, overall: Keeping the notes of the games you played in, is future pleasure – in a jar. Do it! Open it a decade later and enjoy the memories, like I did.


Very interesting and a good read!

My lessons from the last decade:

  • Campaign wikis are great for lore, player background stories and interactions with the world. I also think they work better for character development and many aspects of role playing, as most people are more comfortable and creative at writing between sessions than at speaking during a session.
  • D&D 5e rules work. They are the best version of the rules in my opinion. Easy to get into, quite well balanced and without any really serious flaws. I would prefer them to be a bit more gritty and less Marvel Super Heroes style. The exploration/survival aspect is also too trivial.
  • Sandboxes are great. I love creating and running them. But pure wilderness hexploration as in the Grenzmarken campaign is a bit bland and can feel meaningless soon. Who cares about how many arch demons from Hell your hero killed in a forsaken dungeon in the wilderness? Killing an arch duke in a city or just having the power to possibly do so, is much more interesting and also character defining. So social sandboxes with fame and factions are more interesting for everyone, I think.
  • Dungeons are still the heart of the game and what the D&D 5e rules are made for. The players are on the edge of their toes and fully engaged most of the time: Thinking about tactics, spell selections, resting points etc. Everyone can contribute. The problem with social encounters is that usually it is just 1 or 2 players that are really engaged or relevant and the rest is fiddling their thumbs. I am just experiencing this again as player in DM Mirceas campaign.
  • Now here’s a big conflict with the point about sandboxes: For the campaign to be interesting and meaningful, you need a social setting rife with conflict and interactions. But in actual play it seems to me that running dungeons still works best. It is also less demanding of the DM than running and improvising interesting social encounters. So this led to my current kind of preferred gaming style of running mostly dungeons and combat in actual play and running a social sandbox for character development and world exploration on the campaign wiki.
  • There’s a also a social aspect of relevance at the meta-level: A smaller party size (3 or 4 max) would certainly make it easier to have more social interactions and character development during a session like we had in the past. But I also came to the conclusion that having a bigger party size makes the game and campaign much more relevant for everyone at the meta level. Many of my sessions with just 3 players felt a bit meaningless or bland in retrospective. If you kill the BBEG with a critical hit with just 2 companions at the table it doesn’t feel the same as with 5, where there’s always a lot of cheering, high-fiving etc. going on. It’s a bit like watching sports now in empty stadiums compared to fully packed ones in the past. It’s kind of stupid, but it really seems that among humans, everything becomes much more meaningful and long lasting in memory if there are a lot of spectators/companions.

– Peter 2021-04-05 20:59 UTC

This is a great post! Thank you for it! Is it just where you were in life or something about B/X based games that led to them lasting so long?

– Oliver 2021-04-05 22:39 UTC

It’s not just the B/X campaigns that went on for a long time; the D&D 3.5 and now the D&D 5 campaigns all went on for at least 30+ sessions, if not twice as much – even if I did not enjoy high level 3.5. I suspect that it has to do with how spell levels structure D&D gameplay: on the one hand, every new spell level attained changes the gameplay itself (suddenly you can fly, or fireball large groups of kobolds), and it also advertizes that change ahead of time in the rules: if you play until you get to level so and so, you’ll be able to do this and that. And immediately, people start dreaming.

See 2012-01-24 Changing Gameplay Over Time.

– Alex 2021-04-06 06:58 UTC

Please make sure you contribute only your own work, or work licensed under the GNU Free Documentation License. Note: in order to facilitate peer review and fight vandalism, we will store your IP number for a number of days. See Privacy Policy for more information. See Info for text formatting rules. You can edit the comment page if you need to fix typos. You can subscribe to new comments by email without leaving a comment.

To save this page you must answer this question:

Just say HELLO