PDA

View Full Version : Balance of a System



Piccamo
2007-06-06, 11:32 AM
Warning: Wall of Text, if you don't feel like reading don't tell me about it.

The problem with DnD is class balance. All balance comes down to solely and squarely what a class gives. For some there are many class abilities and for others there are merely bonus feats. The visibility of a lack of balance can be seen more easily the higher level you scale, but is always present. Even having many class abilities cannot compensate if the class abilities are bad.

This thread's intent is not to ask what classes are imbalanced since all of them are, in one direction or the other. This thread's intent is to analyze what must be balanced and a good way to do this, from scratch if necessary.

The aspects of DnD or any roleplaying game are combat interaction, environment interaction, and social interaction. Combat interaction is how a character effects change in a violent setting. Environment interaction is how a character may involve oneself to effect change on the surrounding area. Finally, social interaction is how a character effects change in a social setting. All of these may be used in tandem or completely separate. The problem arises when a single class is too good at any one of these or can do all of them without aid.

The combat interaction is often considered the biggest the largest part of action RPGs. It is often the most complex and is how adventurers gain their money. In combat there are several roles and each class or type of class must be able to perform well at these roles. The roles are AoE Damage, Buffing, Control, Debuffing, Melee Damage, and Ranged Damage. All of these combat roles must be approximately as useful for the system to work.

AoE Damage: In order for AoE Damage to be effective in a d20 system there must be some extra effect attached to damage or an extra effect attached to the AoE Damage. To avoid stepping on the toes of other types of combat we could target different damages. AoE does not do HP damage, it does Initiative damage, skill damage, AC damage, etc. This greatly marginalizes the need for a Debuffer role so we can distribute that role to the others, leaving the AoE Damage role as the strongest as it affects multiple enemies.
Buffing: If Buffing takes actions in combat it won't be fun. If it is done out of combat there is nothing left for the buffer to do once the fight starts. To resolve this attach buffing to other actions: make it so the buffer may buff every time it attacks or changes form (wild shape). Its damage will be somewhat lower, but it will still be able to fill a role and feel like it is contributing to defeating the enemy at the same time.
Control:Absolute control over the situation is traditionally the easiest way to win, but also the least amount of fun when used on the players. Instead of stopping opponents from taking actions, this should modify the actions they may take. At low levels, the control character can reduce the opponent's speed; at high levels the control character can modify the opponent's attack so that it hits an adjacent target. It will still be useful, but less abusable.
Melee Damage: There are many ways to chop off someone's head. The main function of the melee damage is to incapacitate the opponent from a melee range. Melee characters are the most vulnerable to opponents as well so they should also have the strongest defenses.
Ranged Damage: The primary focus of these is to incapacitate opponents from far away. They should have mechanical benefits to being left vulnerable to opponents since they have the most safety most of the time.

Environment and social interaction is another large part of most RPGs. This involves disabling traps, breaking down doors, hunting animals, buying goods, intimidating foes, and more. Every character needs to be able to interact with the environment and other people. I propose a complete separation of Environment Interaction and Combat Interaction from a class standpoint. The easiest way to do this is just assign all characters the same number of skill points per level. Make the environment tasks more important and add new uses to the skills so that they are more desirable. Amber E. Scott's Knowledge (http://www.giantitp.com/articles/paBcfg1YaEccDMQACfu.html) rules are a great start to this.

After enabling all characters to have the same base number of skill points, certain changes to the skill system should be changed. First, remove Use Magic Device. Every character would and should take it if it is available. Second, use a variant Diplomacy system like The Giant's (http://www.giantitp.com/articles/jFppYwv7OUkegKhONNF.html) or Justin Alexander's (http://www.thealexandrian.net/) work in progress. For playing more skilled versus more combat oriented characters this is easily represented with greater ability scores.

Ability scores under this system will be the foundation of all characters. To reduce class imbalance, make every class have MAD. Attach Dexterity to all attack bonus and give a bonus to ranged damage based on Wisdom; knowing where to hit is as important as hitting it. If every class needs at least 3 abilities to be effective it cannot easily outshine other classes with MAD. Feats should also be a major factor in all characters, especially given their rarity.

Every feat should make a character do something new. Static bonuses to different abilities do not demonstrate this. An overhaul of the feat system that fundamentally changes what a character does would make them far more desirable. Power Attack follows this trend very well; it enables melee characters to do something they could not do before. To help differentiate obvious combat characters from obvious skill characters, feats that increase skill points or add class skills to lists (or both) are desirable. These could be prerequisites for other skill feats that allow new uses of the skills. The per day system is skewed as well.

The claim that DnD is based around a 4-encounter per day system fails. If you are trying to limit what a character can do in each encounter, then cut out the middle man and just function at the per encounter level. While an encounter is a variable time, it is easy to impose limitations to it. For example, a buff that lasts the rest of the encounter could be limited to a maximum of 5 minutes (300 rounds).

It is not the d20 system that is flawed, it is a very well crafted system. It is a problem with the class structure of DnD. Any other ideas about how to improve on this foundation are welcome.

Example: Buffer
Lets say a druid falls into the buffer class. We strip its spells and leave its wild shape. Every time the druid changes form it grants an ability bonus to all allies within 30-ft based on the ability the druid activates upon shifting. Further, grant the druid the ability to summon allies and every time it does so, all allies gain an ability they may use, such as a fire shield for fire elementals or scent for dog-like animals.
Example: Control
Wizards would likely fall into the Control role. It is easy to keep in the save system and make a wizard work. Make the DC of the save based on the Wizard's Charisma, to represent its force of personality, but only allow the wizard to effect change based on its Intelligence, to represent that it knows what to do because of how smart it is. A low level wizard might stop some of the enemies from reaching the party on the first round. By mid level it is causing some of the enemies to accidentally aim at the wrong target. By high level the wizard is causing some of the enemies to lose actions each round. Is it still powerful? Yes. Can it still end an encounter in a single turn? No.
Example: Skill Monkey
Under this system it is still possible to be a skill monkey. Start with a high intelligence and as a result gain more class skills and skill points, allowing it to be more useful outside of combat. Because of the high stat in intelligence it is less likely to have high stats for in-combat purposes.

Draz74
2007-06-06, 12:21 PM
Good ideas. I offer two cautions:

- If the ability to be a skill monkey is based only on having a high intelligence, then characters who need a high Int anyway (e.g. wizards) will be the dominant skill monkeys as well. That doesn't fit my idea of what party dynamics should be like. I don't see the party wizard usually being the one who sniffs out traps, or has a silver tongue at negotiations. Also, the "dumb rogue" who still has an OK amount of skills in spite of low Int is an archetype I like.

- Per-encounter abilities have one big problem: the way they can be used infinite times per day outside of combat. Some of the Shadow Hand teleporting-ish maneuvers from ToB, for example, can totally break the game, unless you implement a rule "maneuvers can only be used in combat," which I don't think makes any sense at all.

LotharBot
2007-06-06, 02:41 PM
The key point to creating balance is this:

Every class/archetype needs to grow in power at approximately the same rate. If class A doubles in power ever 4 levels, class B needs to approximately double in power every 4 levels as well. If C doubles in power every 8 levels, D triples in power every 4 levels, and E doubles in power from level 1 to 3 and then again from 3 to 12 and then again from 12 to 35, they won't be balanced with A, B, or each other over the course of a whole game.

One of the fundamental problems with the PHB classes in 3.5 is that the difference between a caster with level 8 spells and one with only level 6 spells is big, while the feats a fighters has picked up between level 11 and 15 don't grow him in power nearly as much. Tome of Battle is a really good attempt to fix this -- it gives the melee classes powers and abilities that scale with level in the same way that spells scale with level. The only thing really missing is the rogue -- the tricky, sneaky, scout/observer, master-of-machines type (Shadow Hand covers some of this, but lacks the class skills to reproduce the rogue.) Give them abilities that scale similarly to the other classes, and you have a basically balanced game.

Of course, you're right to note that "balance" depends on three game areas: combat, environment, and social interaction. One thing I think is important is giving all of the characters the ability to contribute -- not necessarily excel, just be involved -- in all three of these areas. At present, when you have a 5 charisma half-orc barbarian and a social situation comes up, the player is likely to tune out. When you have a rogue in combat with undead (and you haven't allowed any "sneak attack undead" abilities) that player is likely to tune out. And when the rogue is working on disabling a series of traps, all of the other players are likely to tune out, because their characters have nothing to do. It's important to give every class ways in which they excel, things that they do better than the others... but it's also important, along with that, to give every class a way to contribute in the majority of situations. And it's VERY important to make sure those contributions scale appropriately with level, so that everybody in your game gets to be involved equally over the course of the average gaming session. If a player's main ability is to disable traps by rolling a d20 and hoping for a high number, and he can also occasionally deal 8d6 melee damage, he's going to feel pretty left out in comparison to a player whose main ability is to cast reality-altering spells that have a variety of interesting effects.

Matthew
2007-06-09, 07:46 PM
I like the idea of Feats for Class Skills. I'm very much of the opinion that Class skills ought to be treated like Proficiencies. If you play using the 'maxed out' Variant Rule, it might even make sense to see them automatically scale once purchased.

Another thing I have been thinking about is allowing Characters to pick their 'Class Skills' at creation from discrete lists. Using the Weapon Group Rules, these two types of 'Proficiency' could be treated as interchangable.