Sunday, July 19, 2015
Why Modifying An Existing Game Is So Hard
I love game design, I've even designed a few of my own RPGs from scratch. It is a hard, mentally and spiritually challenging slog that your players may look at with disgust at the end. No one in their right mind should ever do it. For those of us who are cursed with that mindset however, there is a big dividing line: do you design your own system or modify an existing system?
I've done both, and designing your own system is easily a few thousand posts of concepts and philosophies - but I want to look at modifying an existing system. Specifically, since I am house ruling 13th Age, which is built off the D20 System, let's look at Pathfinder.
A few years ago I got the idea to start working on a tweak of Microlite20. I spent a lot of time on it, and managed to write 2 or 3 posts, but after many hours driving myself crazy the idea had as many holes in it as I'd started with; so the whole thing kind of fell by the wayside. Recently I stumbled upon an un-posted article I had written that I want to share with you.
-----here is a post I wrote about Pathfinder:
Deconstructing Pathfinder To Tweak Microlite20 - and creating my own Microlite system
Ever since I first read Microlite20 I was amazed by it. It took the Pathfinder/3.5 rules and boiled them down nicely to their bare essence, for a rules-lite, more old school feel to the system. It's one thing to tweak and twist a system, another to see it well enough to hit it's heart. However, being me I just had to look at how I could tweak microlite myself. One thing I thought of was with the BAB and saves. Microlite does not use the base attack bonus or saves. Everybody gets their level for the BAB and saves are mostly based on skills. Which is okay, but I was wondering about maybe adding a system to build classes with the BAB and saves to make them more in line with the core Pathfinder (and since Microlite classes don't do much). So I started looking at the core Pathfinder classes to deconstruct how they were built, and I was surprised.
I started with the 3 elements above, and wanted to see how many different types there were in the Pathfinder classes. So I just looked at the core classes: Barbarian, Bard, Cleric, Druid, Fighter, Monk, Paladin, Ranger, Rogue, Sorcerer, Wizard. Turns out there are only 3 possible BAB progressions, either maxing at 20 or 15 or 10. There are only 2 possible save progressions, maxing at 12 or 6. I decided to throw hit dice into the mix at that point, and there are 4 possible hit dice, d6, d8, d10 or d12. So we have 2, 3, or 4 possibilities for 5 measures (HD, BAB and 3 Saves). Not very neat. I wanted to see how they were balanced against each other, so I made a simple system where each possibility cost 1 point (so 1-4 pts for HD, 1-3 pts for BAB and 1-2 pts for each save) and totaled them up. The highest was 11 points for the Barbarian, Paladin and Ranger. 10 points for the Fighter and Monk. 9 points for the Bard, Cleric, and Druid. 8 points for the Rogue. And only 6 points for the Sorcerer and Wizard. Okay, not too bad, a spread from 11 to 6 points, but that's not counting skills or abilities.
Even though Microlite20 only uses 4 or 5 skills, at this point I was just curious about the underpinnings of the Pathfinder system, so I wanted to see how many class skills each class had, and how many skill points they gained per level. Skill points per level was pretty easy, there are only 4 possible progressions (2,4, 6 or 8 each level, plus Int of course). The fun was looking at how many class skills there were. Now, this is counting Craft and Profession as only 1 skill each, and each Knowledge as a separate skill - that's how the book has them listed in the class descriptions and the skill table. Turns out that for 11 core classes there are 8 different class skill totals- 9, 10, 13, 14,15,16, 21, 28. What a fiddly mess!
Having put my foot in it, I decided to go all the way. Time to total up the weapon/armor proficiencies and the number of class abilities. Weapons and armor are pretty easy. There are 2 weapon groups, simple and martial. Lots of classes have only partial access to a group, or access to a few exotic weapons, we'll call that worth 0.5 point. There are three armor groups, light, medium and heavy. There are two shield groups, shields and tower shields. With each group worth 1 point, and the partial weapon groups worth 0.5, we get a spread from 0.5 for the Wizard up to the full 7 for the Fighter.
Class abilities are the hardest part. It's easy enough to just total the number of class abilities, and I'm looking here from levels 1 to 20. But, some of those abilities increase over time and some are fixed benefits. Rage increases in duration, but uncanny dodge is a fixed ability (with improved uncanny dodge a fixed upgrade). So I really want to weigh this total. I'm going to count leveled abilities, including things like bonus feats for fighters and wizards, as worth 3 points, and all fixed abilities worth 1 point. I think it's more accurate than a single point for everything. This yielded the most incredible results. The Monk has a whopping 11 leveled abilities and 12 fixed abilities - 23 abilities total worth 45 points in my ad hoc scale. While the poor Cleric has the fewest, only 3 leveled abilities (spellcasting, channel, and domain powers) and 1 fixed ability (swap healing spells) - 4 total worth only 10 points. Now, here's where things get troublesome. Spells throw off the whole thing because they themselves are so random and out of balance. A level 9 spell is god-like Wish while a level 1 spell is only slightly better than a sword-swing. But casters can only memorize so many spells. So while in my scale spellcasting is a 3 point leveled ability it's a little harder to pinpoint than that. Still, going off what I did, I totaled up all the core classes. And they ranged from the 62 point Monk to the 24 point Sorcerer. Which actually came out surprisingly close to how I'd imagined. The monk does have a lot of spell-like abilities at higher level, half of his abilities level up, plus he gets a better attack bonus, better saves, more class skills and more skill points per level - all without having to juggle spell slots like the Sorc. Again the level 9 god-spell disparity messes the exact comparison, but really the Monk does kick the Sorcerer's wand overall as a class.
So what have I gathered from this little foray? Well, I knew that Pathfinder and its antecedent D&D 3.5 had no underlying balance or structure, and this just puts it right there on paper. Numbers and ranges and possibilities go all over the place. Class abilities were just chosen and thrown in at random, hoping it would all balance out in the end. Which it more or less does, because people are smart. Once they learn the system they adapt to it and make up for its weaknesses. But having someone Rule Zero your flaws is not a good design strategy.
----- and here is the table I made for myself showing how each class was built:
Class- Barbarian Bard Cleric Druid Fighter Monk Paladin Ranger
Hit Die: d12 d8 d8 d8 d10 d8 d10 d10
# Class Skills: 10 28 13 13 10 14 10 15
Skills/lvl: 4 6 2 4 2 4 2 6
Max BAB: 20 15 15 15 20 15 20 20
Max Fort: 12 6 12 12 12 12 12 12
Max Ref: 6 12 6 6 6 12 6 12
Max Will: 6 12 12 12 6 12 12 6
# Lvl Abilities: 4 6 3 3 4 11 6 6
# Set Abilities 7 1 1 9 2 12 9 11
Wpn/Arm Prof 5 4 4.5 2.5 7 1 6 5
Total # of pts- 39 43 27.5 34.5 34 62 47 52
Class- Rogue Sorcerer Wizard
Hit Die: d8 d6 d6
# Class Skills: 21 9 16
Skills/lvl: 8 2 2
Max BAB: 15 10 10
Max Fort: 6 6 6
Max Ref: 12 6 6
Max Will: 6 12 12
# Lvl Abilities: 5 4 4
# Set Abilities 9 3 1
Wpn/Arm Prof 2.5 1 0.5
Total # of pts- 45.5 24 26.5
how the point totals were calculated:
1-4 pts for hit die (d6, d8, d10, d12)
1-8 pts for class skills,counting Craft and Profession as only 1, but each Knowledge skill separately
(9, 10, 13, 14,15,16, 21, 28)
1-4 pts for skills/level (2, 4, 6, 8)
1-3 pts for BAB (max 10, 15, 20)
1-2 pts for Fort (max 6, 12)
1-2 pts for Ref (max 6, 12)
1-2 pts for Will (max 6, 12)
3 pts/ leveling abilities (ie, any class ability that gains power/level(s), includes bonus feats as one ability)
1 pt/ set abilities (ie, any ability that gives a fixed power/effect)
0.5 pt/partial weapon list or exotic weapons; 1 pt/weapon cat (simple, martial), armor cat (light, med, heavy), shield cat (normal, tower)
-----okay, so why do I drag out all these numbers?
The thing about modifying someone else's system is that you need to understand the framework, the concepts that make up that system. Like a skeleton in a body, each game has certain assumptions and relationships that lie under the numbers and hold them together. In a system like Pathfinder, just looking at the character creation alone (didn't even touch races or combat or spells) there is a complex system of interactions. And if you introduce a new ability or rule you need to think about how it might interact with that underlying structure and what unintended effects it may have on gameplay.
Now, Pathfinder is a good example to me, it's complicated but not as bad as it could be (imagine the Hero system analyzed this way - heck, the game really is all its framework out in the open). And it has the potential for all kinds of crazy as you add new abilities and feats and classes and spells and items into an already complex mix. Also, it has no rigid framework. It was a patchwork growth of the old D&D system that they pruned a little. Which is what's so funny about my modifying 13th Age, itself a modification of the whole d20 tree. And what makes it such a dangerous rabbit hole of first modifying stuff, they wanting to modify the foundational underpinnings of the game itself. Which leads to rewriting the whole game in the end - so the supposedly easier way, to just modify someone else's work, ends up with you writing your own game after all.
Anyways, just a peek behind the curtain of some of the things you have to consider when you want to modify a game, and just how deep the rabbit hole can lead :)