Stunt Dice: A Story in Design Failure

I haven’t written anything in a while, have I? Sorry about that. Life problems, lack of inspiration, excuses, excuses, excuses. But hey! Here’s something new!

There’s been a bit of talk, lately, about failure. The big primary is the Angry GM, who wrote a Strong Article about failure in terms of the game itself, and how accepting it as a possibility can make you a better DM. He also touched on it later in a couple pieces, and I’ve seen the idea of allowing players to fail at tasks tossed around a bit on other blogs as well as on forums and facebook.

One thing that wasn’t really talked about is how to accept failure as a designer. Not necessarily as a DM, but stepping even further back into the role of game/campaign designer. I’ve always considered myself a bit of an amateur designer. My design work has never been published, but I’m still constantly designing new things, so I’m a designer. Ever since my very first game of D&D, I wanted to play with the system and make things my own. It, of course, manifested as a desire to carry around the same cool weapons as my favorite characters from fiction (in particular, I wanted to wield the Soul Edge from Soul Calibur when I was a kid). That desire shifted, however, as time passed and I learned more about the game in which I was participating. About the time I bought my first core rulebooks (edition 3.5), I had finally taken the step from player to DM. And as a DM, I wanted to make everything my own. I introduced new magic items, new classes, and new systems from my earliest days. I remember lancers, dragoons, hunters, mages, dark knights, and something I called the “Fearless” (a concept I’m still trying to make work). My first campaign boss had a Final Fantasy limit break (he was a dark knight riding on a nightmare, and he shot a giant purple laser beam). And this desire to create new and expanded options only grew as I got more and more involved in the games I played.

Pathfinder was a particular area of creation for me. Once the concept of Class Archetypes was introduced, I went wild. The archives of this blog are a testament to how much I wanted to create my own work. However, with that desire to create comes the inevitable result. Failure.

If you decide to create anything, you are going to fail. A lot. And that’s okay. In fact, it can be a good thing. It can be clarifying.

Take, for example, my attempt at Stunt Dice in D&D.

Ever since the latter days of 3.5, when the Tome of Battle: Book of 9 Swords was introduced, I’ve been fascinated with special attacks and techniques in D&D combat. I didn’t really find the Tome of Battle balanced at the time, but the concept of a warrior entering battle with a repertoire of special attacks and maneuvers was appealing. I wanted to spice up combat, rather than just having characters say “I full attack” over and over again. So, a friend and I made our own class: The Swordmaster. It was horribly unbalanced. I then made another: the Artes-Fighter, inspired by the characters of the “Tales” series of video games. This one was, as far as 3.5 is concerned, more balanced than my first attempt, but it still wasn’t quite right. It used a daily pool of points, and I wasn’t entirely satisfied with the result.

Then 4^th edition hit, and gave me everything I could have wanted. At-Will, Encounter, and Daily powers, all perfectly balanced against each other? An entire system built around using cool attacks in combat, so that “attack” was almost always the worst option? It was perfect!

And I rejected it out of hand. I saw how different it was from the D&D I had learned to love, so I immediately hated it. Because it was different, and I was a teenager. I didn’t like things that were different.

Lucky me. Pathfinder was right around the corner. I latched onto it as if I’d never look back. It was everything I loved about 3.5, but turned up to 11. More powerful classes, cooler artwork, and a level of adoration for 3.5 that I found entirely relatable. And I actually played it in its vanilla form for a while. Granted, the vanilla form was constantly being updated as more and more books were released, but I didn’t care. This was a system that understood my needs. It was around the time of Ultimate Combat and the Gunslinger, however, where the Maneuver bug bit me again. I did my own redesign of the gunslinger, which I viewed as superior to the core version. It had access to more varying special techniques. Looking back, I actually really like the design of the Pathfinder gunslinger, and I wonder if some of its methods made their way into D&D 5e.

I then took that idea and ran with it for the Pathfinder fighter. I got wrapped up in the Fighter-vs-Wizard debate that was going on at the time, and wanted to try my hand at making my perfect version of the fighter. I came up with one version that was actually Pretty Neat. It gained Martial Training and Combat Arts that gave it a more varying scale of improvement as it increased in level, as well as some access to martial maneuvers. Thinking back, this was actually one of my favorite things I ever designed in Pathfinder. I designed another that when whole-hog into the martial technique pool, using its own weird version of the gunslinger’s Grit mechanic. It wasn’t good. And even I recognized that at the time. I scrapped it.

It was also around this time that I discovered two RPGs that had been completely unknown to me. The first was the Iron Heroes RPG: a modification of the d20 3e ruleset that completely based itself around varying types of warriors, all using different fighting styles and a “token” system to execute special attacks. It’s actually a pretty cool system, though I don’t think I’ve ever tried to replicate the token gameplay in any real sense. I remember bouncing it around in my head, but nothing every got put on paper. It’s something that works really well for that particular game, but would probably be a bit too complex for other systems.

The other RPG I discovered was the Dragon Age RPG. And while some of this RPG didn’t really click with me (I wasn’t a huge fan of the 3-class system, nor its method of specializations), I absolutely fell in love with its stunt system. Essentially a variant critical hit system, the DARPG stunt system has you choose from a list of stunts to perform if you roll doubles on an attack roll (it uses 3d6 for attacks). Stunts range from the ability to dart around the battlefield to extra damage, and even provoking or threatening foes. It was elegant. It was genius. It was perfect.

Shortly thereafter, however, I grew disillusioned with Pathfinder, and 5e was on the horizon (then called D&D Next), so I instead toyed with different types of games and systems. Numenera, Call of Cthulhu, GURPS, etc. And then 5e hit, and I was there, man. I got all the core books. I got Hoard of the Dragon Queen. I got back into the blogging scene and I was IN IT TO WIN IT!

Then I heard that the Dragon Age RPG was getting a Core Rulebook. And I heard that its system was being published, divorced from its setting, as the Fantasy Age RPG, with years’ worth of improvements thanks to what had essentially become a very long, very expensive playtest.

Holy.

Shit.

So, of course I picked up a PDF of Fantasy Age. And of course the first thing I looked at was the Stunt system. Did it change? Is there more to it, now? I was so excited. And what I saw was exactly what I expected to see. Without going back to check, I’d say it’s exactly the same as the very first time I saw it. And that’s fine. If it ain’t broke, don’t tinker.

I decided that now was the time. I’ve wanted to integrate a stunt/maneuver/technique system for so long, and up until this point, I had only ever met failure. This time, I would succeed. This time I had something solid.

Stunt Dice, I decided, were my best option. It was partially inspired by the expertise dice from early iterations of the D&D Next playtest, and partially based upon the Battlemaster’s Superiority Dice. I couldn’t use the traditional Fantasy Age system, because it was effectively a critical system. And while I could figure out the appropriate percentage chance and apply it do a d20 roll (I think it ends up being around 14-20), that wouldn’t give the same feel as rolling doubles on 3d6. Therefore, I thought of giving characters a number of stunt dice which they could spend to accomplish similar feats: skirmishing around the battlefield, taunting enemies, changing one’s initiative, etc.

I spend a stunt die to skirmish my way onto the tree limbs. This guy's going down!

However, that on its own isn’t enough. The problem is one I’ve encountered before. Presenting the players with action dice and giving them a list of things to do with them is cool, but if it’s divorced from their character’s race, class, and background, then it’s just another tacked on thing to remember. It’s like a lego cannon on a castle that shoots a miniature cannon ball. It’s cool, yes, but the fact that it does a cool thing doesn’t really impact the set itself, and some people might forget that it even shoots a thing, instead focusing on the more central details of the castle.

In order to circumvent this, I decided that I should integrate stunts into the character’s classes. I’m already in the process of designing subclasses for my new campaign (you’ll see more on that as it progresses, I’m sure), but I thought that integrating stunts into them would really create a strong sense of cohesion. If a character needed to spend a stunt die in order to execute a core class feature, then they’d be damn sure to remember that they have the stunt dice.

So, I wrote up a playtest. Instead of using my new subclasses, I decided to use the subclasses included in the Basic Rules, since everyone has access to those, and most people are familiar with the Champion, Thief, Evocation School and Life Domain.

You can take a look at it yourself by following this link here: Stunt Dice!

As I wrote it, however, I noted something. Things weren’t exactly…working out. First, I had to add in a class stunt at first level, so everyone started out with a cool technique they could spend their stunt dice on. This upsets the balance of first-level play, which is very particularly designed to feel a very particular way. And as I designed them, I started asking myself questions like “is that bonus enough to incentivize the ability’s use? Maybe I should add more,” or “this is cool, but I think it’s too powerful. How can I nerf it?” Or the terrifying “this ability seems fine on its own without stunt dice. How do I add them in?”

The answer to that last question, in case you were wondering, is “Don’t.” And does a subsystem have purpose if it doesn’t aid its parent system? The moment you start asking yourself “would this existing feature be cooler with different numbers?” is the moment you should re-think what you’re doing. And that’s what I did. As I wrote, I realized that it wasn’t fun. I wasn’t thinking about the art of design. I was thinking about the math. And that way leads to Pathfinder. Which is fine, if you’re playing Pathfinder.

I’m not. I’m playing D&D 5e. And 5e is a game much more focused on incomparables than math. It’s a game where you don’t ask “which is better? +1 to hit and damage, or +1d6 damage?” Instead, you ask “which is better? Being able to summon my weapon on a whim, or being able to summon a fighting wolf?” The answer to the “which is better” arguments in 5e should always try to be “it depends.” Is Action Surge better than Sneak Attack? It depends. Is Burning Smite better than Hunter’s Mark? It depends. Is it better to spend a stunt die to get 1d6 bonus damage, or 2d6 bonus damage?

2d6 damage. Hands down. This is not how stunt dice should be used. And this is not how a stunt system in 5e should be designed.

I also noted as I went that stunt dice would need some pretty serious balancing in order to work. Adding 2d6 damage at 3^rd level is really good. Adding 2d6 at 15^th level? Less good. Especially since it’s just on a single attack, and the rogue’s core Sneak Attack feature can add 7d6 every single turn.

Which is a better option? Seriously.

Now, a simple way to fix this would be to replace stunt “dice” with stunt “points.” Instead of rolling a d6 with every stunt, stunts have varying effects. The inherent problem here is that it suddenly makes the system much more complex. And this is 5e, where complexity is a choice by design. By using stunt points, you lose a unifying simplicity that stunt dice would give you. When you use stunt dice, all you really have to remember is that you’re rolling a d6, and you then attach that d6 to whatever stunt you’re using. Skirmish? D6 movement. Taunt? D6+8 saving throw. Sieze initiative? D6 + initiative check.

When you then change that to stunt points, things suddenly change. The abilities don’t necessarily use a d6. They COULD, if you chose to make it that way. But if you want to balance things like damage at higher levels, then you have to suddenly start asking: how many d6’s do you use? Do players now have to keep power cards like 4e, so that they can remember what all of their disparate stunts do?

My point is that it’s a mess, no matter how you swing it. COULD you figure out a way to make stunts really work in 5e? Yeah, you probably could. Doing away with damage dealing stunts would probably be step 1. But…I don’t really know that it’s worth it. I’ve wanted to use stunts in my games for years. For YEARS! And yet, once I finally write up a playable, testable system that I can integrate into my home games without much issue…I don’t like it. The idea of the system seemed so genius to me before I wrote it. But now that it’s on paper (so to speak), I’ve become disenchanted with the idea.

And in that way, I’ve failed. My greatest successes in design were things that I actively WANTED to use in my games: The Engineer, the Trained Fighter, my Lingering Injury table (It's on its way). If I design something that I don’t personally want to use—something that doesn’t get me excited at the prospect of using it in my games—then that’s a failure.

And that’s okay. Because I’ve learned from this failure. It has clarified the issue on whether or not I actually want to use a universal stunt system in my games. And the answer is “no, I don’t.” And rather than wallow in this failure, I’m going to use it to create something that’s worth using. Something that you could call a success.

And now I really wonder what it would take to convert Iron Heroes to 5e. THAT seems like it’d be a fun game.