Fairly early on in the development process of a product, I try to decide what categories of rules-crunch is going to be in the book. Sometimes this is fairly straightforward (if I’m working on a Mythic Menagerie book, it’s going to have a lot of monster stat blocks, if I’m doing the Genius Guide to Cheese Magic, there are going to be new spells related to dairy solids). Other times I begin with a goal (such as “create new rules to add some wonder back into mythic music”), and then I have to decide what kinds of things will support that goal. And even if I know what the core of the crunch is going to be, the details often vary (for example some Mythic Menagerie books have feats and magic items, others don’t).

So as I’m thinking about all the fictional harpers and minstrels with mythic music in the fiction that’s inspired the project I’ve opted to “develop in public” on this blog I have to ask myself: what is the core crunch going to include? Since a lot of rules systems are fungible (feats, spells, magic items, skill uses) the first thing I want to decide is what kinds of non-fungible rules, if any, do I want to use. To put it simply, do I want to introduce a new base class or alternate class?

I have this thought fairly often. Alternate classes have snuck into a number of products – such as Advanced Options: Inquisitors’ Judgments
and Ultimate Options: Grit and Gunslingers – where I wouldn’t originally have expected to put them, but my goals (“make judgments a flexible and customizable part of character design,” and “expand the grit and deed system” respectively) lead me to include them.

So, how do I go about deciding to add new classes to a game that already has so many? Like so much of my development planning, it’s a process.

First, an aside and a confession. Mentally, I capitalize class names. To me it’s not a fighter, it’s a Fighter. Not only does that help my mind codify the difference between “a random character who fights things” and “a class that grants the ability to take Weapon Specialization,” it also makes it easier to find references to specific classes in my writing. Sadly for me, the Super Genius Games style guide matches the Pathfinder standard of not capitalizing class names. This, however, is a peek into my mental process. As a result, for this post I’m sticking with my incorrect usage of Monk over monk, and Wizard over wizard.

Over the two years or so, Super Genius Games has developed and produced a lot of new base classes, which I define as character classes designed for entry at 1st level as a starting character and that contain a full 20 levels. So far we’re published (including alternate classes) the Archon, Armiger, Death Mage, Dragonrider, Enforcer, Fusilier, Godling (in four flavors of sub-class), Justicar, Magus (likely soon to be renamed the Magister), Mosaic Mage, Shadow Assassin, Shaman, Templar, Time Thief, Time Warden, Vanguard, War Master, and Witch Hunter. We have several more in various stages of production (including the Chronomancer and Dracomancer). If the line continues to be popular, I expect we’ll do several more after those. I also note the rate at which we’re producing them has slowed down gradually, though hopefully it’ll never stop.

The reason we’ve slowed down is that I believe there really are a finite (though not necessarily small) number of good ideas for base classes. Even more importantly, I think there is a finite number of base classes that should be in the same campaign world. For example, while you may have room for an Archer base class even with the Fighter and Ranger, you won’t also need a Hunter, Sniper, Arcane Archer, Musketeer, Force Fusillader, Rock Thrower, Sling Chucker, Bowyer, Marksman, Sharpshooter, Tellite, and Nimrod. Even if those were all good ideas for a base class (and not all of them are), there’s absolutely no need to have them all in the same game. Base classes can certainly have similarities and fill the same niche on a team (Sorcerers and Wizards are the best example of this, though Barbarians, Cavaliers, and Fighters come as a close second), but they also need to be distinctly different.

Base classes also need to be fairly broad, to prevent them from becoming either boring or straightjackets that limit player creativity. It’s fine for prestige classes to be very specific (indeed, it’s often crucial), but base classes need greater flexibility. This is even more true for the Pathfinder RPG than the popular d20 game it grew out of, as all classes in Pathfinder have more customization built into them. And with the addition of archetypes (variant powers you can use to replace the normal class powers to focus on a different game element), that customization becomes even broader, further reducing the need for either prestige classes or additional base classes.

So, when looking at an idea for a base class, the first thing I have to do is compare it to our theoretical “campaign,” to see what base classes already exist. Obviously I include everything that’s official to Pathfinder, which was originally just the Core Rulebook, but now includes the new base classes from the Advanced Player’s Guide. That gives us a group of 19 classes (Alchemist, Barbarian, Bard, Cavalier, Cleric, Druid, Fighter, Gunslinger, Inquisitor, Magus, Monk, Oracle, Paladin, Ranger, Rogue, Sorcerer, Summoner, Witch, and Wizard) and three alternate classes (antipaladin, ninja, and samurai). Since I want fans who buy all of our products to be able to use them all in the same game, I also add all of our own above-mentioned base classes for a total of 38-41 base classes (depending on how you count the four Godlings) and 5 alternate classes. I don’t expect anyone to use all these classes in the same game, but I want them to be able to without difficulty.

When any new base class ideas get batted around in my planning, I look at how they compare to this massive set of exiting options. I frequently thrown some things out as too similar to existing or upcoming classes. Once I’m convinced a base class idea is different enough from the dozens of existing options, I have a few more criteria to test. Is the class something appropriate for a 1st level character? Does it match a common play style for fantasy RPG gaming? Is it going to be interesting enough to support a whole PDF product?

A lot of those questions have to do with what character ideas go well in the kinds of fantasy stories a Pathfinder campaign tends to tell, and what characters players are likely to be interested in. I could write a Packmaster, who excelled at cargo carrying and packing, and make it a viable base class. But even if such a thing existed, most GMs wouldn’t be too pleased to find one in their campaigns. Which wouldn’t be a big problem, because I doubt many players would be chomping at the bit to play the mystic porter. Similarly I could produce a Federation Explorer, a base class representing advanced aliens with teleportation technology and stun-ray weapons, to represent members of an away-team sent into a feudal fantasy world to study it. Some players might love it. A lot of GMs (and other players) would hate it. So already, I’m selling to a fractured market, There’s a place for kooky ideas like that, but it calls for more support material (like a mini-setting at least) that outs it outside of the scope of a single-class PDF.

(Yes, I have the Enforcer, from our currently 1-product long Anachronistic Adventurers line. Yes, it violates the rule above to some extent. I maintain it’s a much more common trope in fantasy stories than explorers from a starship. Even so, it’s outside my official core guidelines. I wrote it anyway. I love me some pulp, and what’s the point of being Lead Developer for a game company if you can’t break your own rules occasionally?)

Other questions are about in-game logic. Yes, I could build a 1st level character class called the Archmage, that would given vast flexibility over magic, and still make it balanced from 1st-20th level against all existing classes. But does it really make sense within the context of a game world for a 1st level character to be an Archmage? If you start life as an Archmage, what prestige classes still appeal to you? Why does anyone train to be a Wizard, if they can be an Archmage at the same point in their life. And, is what a first-level character reasonably can accomplish within the limits of game balance going to inspire anyone to call the an “arch” anything?

Sometimes I have to change gears with an idea mid-process. For example, I originally had plans to do the Genius Guide to the Justicar, to present a non-spellcasting enforcer character. When I saw how the APG’s inquisitor worked, I realized no one needed a whole new class to cover my justicar idea. Instead, the justicar became an alternate class of the inquisitor that focuses on judgments in Advanced Options: Inquisitors’ Judgments. I’ve similarly turned core class ideas into archetype packages (from the various ‘Genius Guide to X Archetypes’ products), new class options (the omens mystery presented in the Genius Guide to Divination Magic was originally a Soothsayer class I decided worked better as an option to allows Oracles to actually predict the future), and even just feats (The Genius Guide to Feats of Subterfuge took some things I had developed for a Charlatan class and turned them into feats anyone could take).

Much more often, asking these questions forces me to clarify my thinking on what a class is, and isn’t, supposed to be. When writing The Genius Guide to the Templar, I found myself struggling to decide how much the Templar should look like the Paladin. That lead me to ask how people within a campaign might view both Templars and Paladins, and that lead to a sidebar about the differences between the classes which went into the final PDF. Once it was clear in my head that my Templar served a very different function, and got its power from a different source, than my assumed “Paladin standard,” it was clear to me the two should have no powers in common even if they had similar concerns. This resulted in Templars having similar themes to Paladins, but expressing those themes through different powers and choices, and helped me build a class design that could work in conjunction with Paladins or replace them, based on the desires of the GM and players.

Once I am sure I have an idea I wish to progress with, I have to consider how it will work with other characters. Pathfinder is designed on the assumption that most groups will be 4-5 characters, and include a thief-type, a combatant-type, a healer/divine spellcaster, and a fireballer/arcane spellcaster. Lots of groups vary from that format, but they do so at some risk to themselves. And many groups are militant about making sure these roles get filled (often resulting in the last person to make a character being forced into a role they aren’t excited by). When making a new class, I want to make sure it can fill one of those four roles (so it’s a viable and exciting new option for players forced to fill them) without totally overshadowing any existing class. This is done both to avoid players being upset (as their original characters are upstaged when a SGG class enters into the game), and to avoid the slow increase of PC power that rules supplements often introduce. It’s a tricky process, that requires a lot of side-by-side comparisons, and as much playtesting as can be managed.

For Pathfinder in particular, I also try to make sure there’s a strong customizable element in any SGG base class. With the addition of things like bonds (nature’s bond, hunter’s bond, arcane bond, and so on) and class packages (sorcerer heritages and fighter group specializations), Pathfinder has made nearly every class as easily customized as the cleric was back in 3.5 rules. The addition of favored class benefits, archetype powers, and traits further expands how many ways you can write characters of the same class. Not only do I feel this is a powerful part of Pathfinder’s success (which I therefore wish to duplicate), I think it helps base classes’ find a role with a given group of gamers. The Death Mage, for example, has pale roads, which change the focus of the class and give it different kinds of creepy. I’ve had players tell me they love taking the anti-undead reaper mage road, while others have said that option is useless, but they love the creepy ghoul mage road, while some GMs just like using tomb mages and shadow mages as villains. Those are all the same class, but the custom options to change focus and style allow different groups to use it in different ways.

Two other things I always want to look for is making sure my idea for the class won’t be too appealing for a 1 or 2 level dip by other characters, and making sure it has something interesting to do even at 1st level. I mention those together, because they form a kind of opposing game-mechanical tensions. Obviously if you give a class a power to reroll attack rolls at 1st level (Genius Guide to the Witch Hunter), the power of that option is based on how many times the ability can be used. Since it’s one of the core powers of the Witch Hunter, I originally had it available several times even at low levels. Since low-level Witch Hunters aren’t going to deal that much damage even when they hit, I wasn’t worried about letting them use the ability several times. And my playtest run at 1st level was fine. When I bumped the playtest to 6th level, however, my players brought a Paladin 5/Witch Hunter 1, Fighter 5/Witch Hunter 1, and Barbarian 5/Witch Hunter 1. If every fighting class is better off with one level of Witch Hunter, the class is too attractive as a 1-level multiclass. (I limited the ability to once per day per class level maximum, moved the ability to know if an area is bewitched down to 1st level to give the Witch Hunter more spotlight time at 1st level, and the problem went away.)

So, looking at my long list of guides, do I want to add a full base class to my Mystic Music product?

Probably not, and overlap is a big chunk of why. If someone is already playing a bard, I doubt that player is going to be happy if another player comes along and runs a “harper” with even more control over music. One could reasonably argue that you can have two masters of Perform in the same way you can have sorcerers and wizards in the same team, but ultimately I think the roles would just be too closely intertwined.

I do see having an alternate class or two as an option. The official Pathfinder rules already have the rogue and ninja, and clearly those overlap quite a bit. So I think the player base would accept a “harper” alternate class of the bard that had some new subsystem (like ninja have ki and rogues don’t) without expecting it to be so different from the bard as to avoid all overlap.

In many ways, that’s good news. It means I can just worry about pinning down what mechanics and options I want to create my mystic music feel, and decide exactly how players get those options later. I’m looking at creating one or more fungible rules systems, and those are easier than entirely new classes. Once I have some systems I like, shoehorning them into templates, feats, skills and spells is easy.

Of course, coming up with those systems is pretty complicated all by itself…