Today was the last session of my Dungeon World game. We ended things very dramatically, with the destruction of the universe. The characters from this game are probably going to be the pantheon for the D&D game I’ll be running next, and I’m always a fan of that sort of continuity. Good endings all around, including at least one heroic death. I was most happy to reveal that Iggy the mule had been a dragon all along, largely because that explained why he had consistently been the most effective member of the group. 
I learned a lot over the course of running this, and there are definitely some things I’d do differently. I have a much more comfortable relationship with the system now, which includes a much stronger sense of what works for me and what I’d change. That said, there are definitely a few lessons I’ve taken away for any future games of Dungeon World I run or play in.
#1. Be Stingy With XP
In retrospect, advancement may have been my least favorite part of the game. The difference of the best bonus going from a +2 to a +3 is really huge, and the general improvement in stats meant that the awesome game-driving goodness I got out of bad rolls in early sessions kind of ran out of steam. The addition of moves was fun at times, but the most interesting ones often felt like things that should have been part of the class to begin with, so that took away some of their luster.
Barring a complete revamp of advancement (which I wouldn’t rule out) I would be more conservative with XP sometime. My default was that if the advancement questions could be answered vaguely yes, I would give the XP. If I’d been stricter, it would have at least slowed advancement down a little.
#2. Bonds Break Down
As written on the character sheets, the bonds will totally hold up for a quick game, but over the course of the campaign, they need to evolve. This is exacerbated if, as in my game, the cast of your game changes regularly.
This has play problems, but there’s a weird practical problem too – it violates the cleanliness of the character sheet. One of the great things about DW is the self-contained nature of the character sheet. Things are largely static and binary – checked or unchecked – with only a handful of values that can be changed, and those only occasionally. The character sheet is designed to hold the starting bonds, but once you’ve moved beyond those, the game has no physical place for them.
There are a couple possible solutions. I might use a bond sheet – a dedicated extra sheet that explicitly replaces the bond section in the sheet. It’d be lines and checkboxes, and when a bond pays out, you check it off, take your XP, and write a new bond. Alternately, keep the bonds on index cards so they can be shuffled, dealt around and physically manipulated in play.
All that said, there is very clearly a trick to writing good bonds, as evinced by the uneven nature of bonds out there. Some are clear an obvious play drivers. Others may not push play, but can establish an interesting truth or dynamic. And others just lie there like a turd on the lawn. Unfortunately, I don’t think there’s any simple rule of thumb that makes for grippy bonds. That might merit more thought later.
(By extension, bond ratings get really weird over time, but that’s a simpler problem).
#3. Hit Points are currency, not description
There is a whole discussion to be had about the nature of damage in Dungeon World. it has some definite issues, especially when the dice get swingy. This is hardly unique to DW, but it’s a bit of a hurdle. I get why they moved away from the simpler wounds system – damage dice give a sense of greater granularity, and they give a chance to use other die sizes. But I feel like there needs to be a middle ground where we don’t have enemies still int he fight because the PC rolled 2 points of damage too few.
But the issue of PC damage and enemy HP are easy to fix, it’s the reverse that’s complicated. Random monster damage is a weird absolution of GM authority over pacing to the whims of the dice, and some folks might find that more fun, but it feels awkward to me. Specifically, damage feels like a one-size-fits-all solution to a pacing problem, and it has the problems you could expect. I found damage got a lot more interesting when there was something at the bottom of the HP pile that the player could do (like sacrifice armor or take conditions rather than go straight to taken out).
In retrospect, I actually wish I’d cheated a bit more on monster damage. Not, like, totally arbitrary cheating, but coming up with categories of damage relative to hit points to get a similar effect as optimal damage play, but with less swinginess. Min-mid-max damage did me well enough, but I didn’t fully appreciate the problem I was solving when i started using it (and will never stop).
There are some corollaries to this, the two big ones being Armor matters a lot and the second that I am not 100% in sync with DW combat (still working on the latter).
In any case, I imagine I may have some more thoughts once I’ve had time to chew on things, but those three were very much top of mind today.
- We had a few players who couldn’t show up, so their characters were laid out on the table to be grabbed for flashbacks and hooks in. ↩
- The ability to write a new bond with the same character would be a requirement for getting the payout. If the bond has not lead to a next step, then it hasn’t really been tapped. ↩
- It also casts into sharp relief that combat is the one area where we discard the principle of letting a roll stand. ↩
- One curiosity. In our last session, every character’s armor value was either 1 or 4. That seems weird. On some level, I wonder if Armor should be handled like weapon damage, by class rather than by gear. ↩