D20 Spell System: Statement of Objections

[ruemere]

This is an extended version of my issues with d20 spell system. Posted here as per suggestion by Wulf Rathbane.

INTRODUCTION

This short document presents my view on the problems of d20 spell system. It is not intended to be exhaustive or definitive, however since it is a list of problems I have experienced during several years of playing and GMing the game, it may provide someone with a good way to start of bringing some improvements into the game.

The whole story could be traced to my first experiences with Final Fantasy system (sorry, Wulf, it was PC, not Amiga - my bad), when I thought about converting 9 spell levels of AD&D to 100 levels of the computer game. Yet, it was only during Pathfinder playtesting when we, my players and me, have begun to experience painful side effects of d20 spell system.

This critique is meant to assemble together issues. Solution proposals are beyond the scope of this short article.

Here we go.

1. The core of d20 system is a test of modified d20 roll vs static number (Difficulty Class or DC). And supplementary damage or secondary roll. The DC should be a static number unless you are fond of grandmas being able to arm wrestle dragons (i.e. the infamous natural 1 versus natural 20 during contested rolls).

2. D20 roll is modified by skills, Base Attack Bonus (BAB), saves and other statistics. By default, the modifiers scale with level.

2.1 Exceptions: neglected skills, neglected character abilities.

3. While numerous spells also scale with levels (directly, via caster level dependencies, indirectly, through feats and ability enhancments), quite a few of them fail to do so.

3.1 Failure #1: spell DC does not scale well beyond spell level (this is an obvious problem for epic campaigns) since spell levels run out around 9.

3.2 Failure #2: spell DC scales differently than saves (at different rate).

3.2.1 The proper rate of scaling for saves assumes a lot about defender's build and wealth.
3.2.2 The proper rate of scaling for saves assumes a lot about availability of effects which affect saves.
3.2.3 Spell DC is based on caster's forte while defender's save is usually defender's Achilles' heel.

3.3 Failure #3: there are spells which circumvent the default core test of save roll vs DC or default core test of skill vs DC.

3.3.1 Skipping the default test. Example: Knock.
3.3.2 Replacing with a nonstandard test. Examples: Touch spells often do not allow saves, some spells substitute the default test with caster level check (example: Dispel Magic).
3.3.3 Introducing non-relative, absolute effect. Example: Forcecage (as of Pathfinder BETA, Forcecage allows save), Antimagic Field, Wall of Force.
3.3.3.1 Absolute effects bring d20 spell system problems to a whole new level - that of underequipped classes becoming irrelevant. Or of characters who fail to bring appropriate tools for the adventure and thus become stage filler.

3.4 Failure #4: Binary/Lockdown spells are instants with no maintenance cost (read on below for explanation).

3.4.1 Binary spells (also known as Save'Or'Die or Save'Or'Suck) either work (and remove target instantly) or fail. Example: Finger of Death.
3.4.2 Lockdown spells instantly negate character to the point of that character losing ability to affect play. Example: Insanity, Feeblemind.
3.4.3 Instant spell with no maintenance cost should be contrasted with slow depletion of resources caused by other methods of combat. Basically, a spellcaster gambles (trying to use a spell with odds favoring spellcaster's side) while ranged/melee combat character laboriously continues war of attrition.
3.4.4 To bring two sides to the same level, either the spellcaster would have to settle for delayed onset of effects (with partial effects taking place at earlier stages of casting) or the ranged/melee character would have to be able to produce an attack, with usage limited daily, able to penetrate opponent defenses with a simple save or lose effect.

3.5 Failure #5: There are no passive magic defenses except Spell Resistance. An Archmage in pajamas and a king in a privy are open to nasty assassination attempts unless ludicrous amount of funds is expended on magic defenses.

----

That's it. There are already quite a few attempts to patch these issues (notably, quite a few of those were fixed in True20... at least according to quick start rules available for download), however there has been no real comprehensive alternative solution.

I cannot help noticing that most of these problems have been resolved in many online games, by either turning everything into hitpoint damage, damage over time effects or very short durations or easily broken up effects. However, such solution comes at the cost of losing the difference between spellcasters and non-spellcasters.

Two or three games introduced quite interesting ideas. Rolemaster (MERP incarnation) turned spells into standard tests with special bonuses. Storyteller games use dicepools with a chance of absolutely devastating success being relatively low. Savage Worlds introduce metagame mechanics (Wild Die and bennies) to allow protagonist cheat their odds. And so on.

Yet, these games are not D20. So, who knows, maybe someone can think of redesign d20 spell system?

Regards,
Ruemere

[GlassJaw]

Great post ruemere, and I think you've pretty much nailed the d20 casting system on the head.

To me, one of the best casting systems was Shadowrun (I'm slightly less familiar with SR4, although I think it's fairly similar).

In Shadowrun, no spell or "target number" was absolute, there was always a scaling degree of success, and the target always had a chance to resist.

For Trailblazer specifically, a complete rewrite of the spell system is not going to happen. Our goal is to offer universal fixes and rewriting as few specific spells as possible (Dispel Magic will get a rewrite though).

We can't do much with static DC's without a rewrite but we are certainly addressing binary spell effects, lockdown effects (great term by the way), and the lack of passive magic defenses, largely through an expansion of the Action Point system.

[Wulf Ratbane]

Originally Posted by ruemere

This is an extended version of my issues with d20 spell system. Posted here as per suggestion by Wulf Rathbane.

Thanks for posting. It's useful, and well thought out enough that I'm happy to revisit the issue with you.

3.1 Failure #1: spell DC does not scale well beyond spell level (this is an obvious problem for epic campaigns) since spell levels run out around 9.

The "core" progression for DCs is 10 + 1/2 HD + ability modifier.

As it turns out, spell progression actually advances one level faster. A 3HD caster can cast a 2nd level spell at DC 12+ability modifier; the same caster with a special ability would round his HD down, getting only DC11+ability modifier.

This hold true only up through 18th level or HD.

3.2 Failure #2: spell DC scales differently than saves (at different rate).

It scales at exactly the same rate as "Good" saves. The slope of the line is the same-- Good saves advance at 1/2 HD.

Back out the d20 roll (10.5) and the ability modifier to compare baselines. (We'll even back out opposed feats, like Spell Focus vs. Iron Will, etc.)

At 1st level/HD, the Good save baseline starts at +2.

At 1st level/HD, the spell DC starts at +1.

This puts these two opposed factors smack in the middle of the d20.

So far so good.

3.2.1 The proper rate of scaling for saves assumes a lot about defender's build and wealth.

Agreed-- but that is a problem best addressed in PC Wealth, Big Six, etc.

There is no easy analog on the caster side for increasing spell DCs that compares with Resistance items.

3.2.2 The proper rate of scaling for saves assumes a lot about availability of effects which affect saves.

Can you please clarify this point?

3.2.3 Spell DC is based on caster's forte while defender's save is usually defender's Achilles' heel.

If the caster is on his game! I consider this a feature, not a bug.

The "Poor" save lags the "Good" save by 2 points at 1st level and drops behind by another +1 every 3 levels.

However, the "every 3 levels" rate is the same rate at which Resistance bonuses advance. Assuming the DM allows the PCs to advance their Resistance item at a consistent rate, you'll end up with the following table:

Level = attacker or defender level/HD
DC = baseline DC of spell or special ability
Good & Poor = saves with no resistance item
G (res) & P (res) = same saves, with Resistance item +1 per 3 levels (+5 max)

Level	DC	Good	Poor	G (res)	P (res)
1	11	13	11	13	11
2	11	14	11	14	11
3	12	14	12	15	13
4	12	15	12	16	13
5	13	15	12	16	13
6	13	16	13	18	15
7	14	16	13	18	15
8	14	17	13	19	15
9	15	17	14	20	17
10	15	18	14	21	17
11	16	18	14	21	17
12	16	19	15	23	19
13	17	19	15	23	19
14	17	20	15	24	19
15	18	20	16	25	21
16	18	21	16	26	21
17	19	21	16	26	21
18	19	22	17	27	22
19	19	22	17	27	22
20	19	23	17	28	22

3.3 Failure #3: there are spells which circumvent the default core test of save roll vs DC or default core test of skill vs DC.

Agreed, it's a problem.

3.3.2 Replacing with a nonstandard test. Examples: Touch spells often do not allow saves, some spells substitute the default test with caster level check (example: Dispel Magic).

I'm not particularly concerned with spells that require some kind of roll, even if it's an "easy" roll like a touch attack. We've added more ways to defend actions such attacks (combat reactions and Action Points).

3.3.3 Introducing non-relative, absolute effect. Example: Forcecage (as of Pathfinder BETA, Forcecage allows save), Antimagic Field, Wall of Force.
3.3.3.1 Absolute effects bring d20 spell system problems to a whole new level - that of underequipped classes becoming irrelevant. Or of characters who fail to bring appropriate tools for the adventure and thus become stage filler.

3.4 Failure #4: Binary/Lockdown spells are instants with no maintenance cost (read on below for explanation).

3.4.1 Binary spells (also known as Save'Or'Die or Save'Or'Suck) either work (and remove target instantly) or fail. Example: Finger of Death.
3.4.2 Lockdown spells instantly negate character to the point of that character losing ability to affect play. Example: Insanity, Feeblemind.

Because of the many changes we've made to the core system in other places (again, action points being a very key component here), I limit my concern to "absolutes with no opposed d20 check of some kind."

.... AND:

"Lockdown spells that are just a no-fun drag for everybody."

(Put solid fog or black tentacles in this category.)

3.4.3 Instant spell with no maintenance cost should be contrasted with slow depletion of resources caused by other methods of combat. Basically, a spellcaster gambles (trying to use a spell with odds favoring spellcaster's side) while ranged/melee combat character laboriously continues war of attrition.

3.4.4 To bring two sides to the same level, either the spellcaster would have to settle for delayed onset of effects (with partial effects taking place at earlier stages of casting) or the ranged/melee character would have to be able to produce an attack, with usage limited daily, able to penetrate opponent defenses with a simple save or lose effect.

Hmm... As I mentioned to you when we were discussing this back and forth in private messages, what I don't want Trailblazer to do is to start modifying individual spells on a case by case basis.

So it's absolutely necessary to identify broad categories of problems and then find systemic solutions.

So-- and mind you, I'm speaking off the top of my head here-- I could live with a change like this:

Originally Posted by SRD

Burst, Emanation, or Spread: Most spells that affect an area function as a burst, an emanation, or a spread. In each case, you select the spell’s point of origin and measure its effect from that point.

/snip burst and emanation/

A spread spell spreads out like a burst but can turn corners. You select the point of origin, and the spell spreads out a given distance in all directions, expanding by 5' per round until it reaches its maximum spread. Figure the area the spell effect fills by taking into account any turns the spell effect takes.

Making systemic changes like that allows you to build a "trap" for common problems, without having to rewrite spells on a case by case basis-- which in any case could never possibly cover all of the third party spells out there.

Does this spur any further thoughts?