Monday, 8 September 2014

D&D 5e Monsters: Part 6: Construction: Traits & Spellcasting

D&D 5e Monsters: Part 6: Construction: Traits & Spellcasting

Surf digs into Traits and takes a closer look at Spellcasting...

 

Part 6: Construction: Traits & Spellcasting

Traits are a crucial part of many creatures, serving to provide both grit and flavour. Many creatures also rely on these exclusively for their combat relevance in the game. This can be the case with Spellcasting, which is an incredibly flexible Trait.

We'll start with a broader Traits Overview - the types of Traits, how those impact other areas and their distribution across levels. From there we'll consider how Spellcasters are typically implemented in D&D 5e.

Looking at Spellcasting leads naturally into an analysis of Damage and my original plan was to include that on this instalment. Unfortunately I ran over on both time and content before I could cover this. So I will do a post on Damage all on its own, just as soon as it's ready.

Finally we'll apply what we have determined to our Example Monster.

 

Traits Overview

Traits are facets that alter the way game rules apply to creatures with that Trait. The implementation of Traits is evidence that D&D 5e's modularity is underpinned by an exception-based ruleset. It is also one of the main areas where templating is evident, since so many Traits are trivially transportable and we see them used by many unrelated creatures.

Conditional vs Simple

All Traits can be considered enablers and the majority are conditional.

Conditional Traits enable a benefit (or occasionally a penalty) based on a specific circumstance being met. Let's consider the following...

  • While the gargoyle remains motionless, it is indistinguishable from an inanimate statue.
  • The badger has advantage on Wisdom (Perception) checks that rely on smell.
  • The crab can breathe air and water.

In the first case the condition is that the Gargoyle is motionless. In the second the Badger only gains Advantage on a Perception check if that check relies on smell. The third case is more subtle yet it is still conditional - the Crab must be breathing either air or water in order to not drown. It would still drown in oil or a vacuum, for example.

Traits that are not conditional can be considered simple enablers - they apply a flat modification to a creature without any prerequisite condition. Examples include the following...

  • The elemental can enter a hostile creature’s space and stop there.
  • The banshee can magically sense the presence of living creatures up to 5 miles away. She knows the general direction they’re in but not their exact locations.
  • The crocodile can hold its breath for 15 minutes.

These simple enabling Traits generally have minimal impact to the way the creature performs in combat and are relatively uncommon. That said some non-conditional Traits have direct combat impact. Some Traits can act as attacks (like the Medusa's Petrifying Gaze) or provide access to attacks and/or utility Actions and Reactions. The Spellcasting Trait being the most common of the latter.

Categories

A complementary way of viewing Traits is by their use or impact, as I see it these fall into three broad categories...

  • Combat Traits: Have a direct bearing on combat. Examples include Pack Tactics, Sneak Attack and in many cases Spellcasting.
  • Stat Boost Traits: Serve to provide some kind of boost to a creature's basic stats, including providing advantage on skill checks and possibly saving throws. Examples include Freeze, Keen Sight and Regeneration.
  • Utility or Flavour Traits: Provide additional modifications to creatures for purposes that don't directly relate to combat or other mechanical considerations. Examples include Shapechanger, Water Breathing and in some cases Spellcasting.

Because the two approaches are complementary they can be used togethor. What I see is that most combat and stat boost Traits are conditional, while most simple enabling Traits are of the utility/flavour type.

Note that the Spellcasting Trait is very flexible and variable, so we'll take a closer look at that shortly.

Impact On Calculations

It's important to note that Traits can impact basic stat calculations and that combat Traits and stat boost Traits sometimes need to be considered when assessing a monster's actual CR. For example the Regeneration Trait appears to factor into a monster's HP count and attack-type Traits (including Spellcasting in some cases) factor into a creature's Damage assessment.

Distribution

OK so that's all important and very good, but "How many Traits should I give my monster?" I hear you ask! Let's have a look at the numbers...

Monster Traits Scatter
Monster Traits Scatter

If we create a scatter graph showing the number of Traits each creature has we can see that the data is tight and at a pretty low level. If we then drop a couple of trendlines on this, one Linear and one either Power or Logarithmic, we can see that the variability of data is also fairly low. That is the data generally has a low Standard Deviation and Variability. While there are exceptions at a couple of CRs, Var and StdDev are mostly 1 or less, occasionally making it as far as 2.

That's a fancy way of enumerating what you can see plainly - on average monsters have 2 Traits and the normal variation is +/- 2.

Monster AC Averaged
Monster Traits Averaged

If we summarise the data, take the result and use that to produce another scatter graph with the same two trendlines... Does this change much? Well no, it really doesn't. This only serves to condense the data points, but it has no significant impact on the trendlines... on the shape of the data.

This essentially leaves us with two options. The first is a Logarithmic (or Power) formula that starts at a couple of Traits at CR1, plateauing by CR5 and only very slowly progressing from there.

CRNum
Traits✝
0 1
1
¼1
½1
1 2
2 2
3 2
4 2
5 2
6 2
7 2
8 2
9 2
10 2
✝ Varies by +/-2

The other option is a simple Linear formula. Based on the current crop of monsters I think we'll see that this is, in fact, a fairly flat Linear progression.

So exactly what kind of progression do I expect we'll see in the Monster Manual? Well after rounding we'd want CR1 to hit 2, but under CR1 we want 1. So we want to start CR1 at 1.5. At CR30 I believe we'll want to just hit 4, after rounding. So that at CR30 our target is 3.5.

A moment's work in Excel suggests something similar to this...
    Number Traits=0.07 x CR + 1.43

When using this formula it's important to keep in mind that that a variation of +/-2 is easily within the normal ranges. And realistically one can go to +/-4 by exercising a little care. What do I mean by that? Well adding six or eight combat-oriented Traits to a CR1 monster can produce creature much stronger than its CR indicates. Complementary and synergistic Traits can also be problematic here if not carefully considered.

Prediction: Analysis of the Monster Manual will reveal that number of Traits is a slow linear progression, as we have outlined above.

 

Spellcasters

The Spellcasting Trait effectively grafts the core PC system for spellcasters onto a monster. Spellcasting and its close relative Innate Spellcasting are arguably the most versatile and powerful Traits commonly found in the current crop of D&D 5e monsters. There are several caster archetypes obvious in the current crop of monsters and we'll have a look at these once we've considered the structure of Spellcasting.

Spellcasting Overview

Spellcasting and Innate Spellcasting are discussed on pages 5 & 6 of the DM Basic Rules v0.1 available in Basic D&D. This section sheds a little light on how to use monsters with this Trait and, if we read between the lines, even gives us a few clues to how they are built. Combined with the elements of Spellcasting that we see on monsters with that Trait we gain a pretty clear understanding of how that Trait is implemented. The components are...

  • Spellcaster Level is used to index several elements of the Trait from the appropriate class. It is usually several levels over the CR of the creature (typically 3 levels higher), but I don't believe there's a simple formula applied to this. Instead I think it serves as a type of "dial" to tune in damage, support or utility reuqirements for the monster. Spellcaster Level is normally followed by several dependant attributes...
    • Spellcaster Ability is usually the highest of Int, Wis or Cha.
    • Spell Save DC in most cases is calculated as...
          DC=8 + Proficiency_Bonus + Spellcaster_Ability_Modifier
      The occasional creature seems to have a bonus or penalty on top of this. I suspect this is either a racial bonus or a manua "tweak" to adjust particularly powerful or weak casters.
    • Spell Attack Bonus is simply calculated as...
          Bonus=Proficiency_Bonus + Spellcaster_Ability_Modifier
    • Spellcaster Class dictates the cantrips, spell slots, available spells, etc as detailed below
  • Cantrips are gained as normal for a PC caster of the appropriate Class and Level.
  • Spell Slots are applied as normal for a PC caster of the appropriate Class and Level.
  • Spell List details the spells the monster knowns or has prepared (as per its Spellcaster Class). Note that these are usually fewer than for a PC of the same caster class and level, typically one fewer at each spell level, however it may be up to the same number as the equivalent PC caster class and level. My suggestion is that only spells that fit the monster's underlying design concepts are added. Simply adding spells to match the number that PCs of that Class and Level have might seem intuitive, but ultimately it makes the monster more difficult to operate and therefore less effective.
  • Domain Features available to the class may also be added to the monster.
  • Note that Monsters may cast spells using a higher slot if one is available and if that spell allows casting at a higher spell level.
  • Note the section on modifying a monster's spell list and the warning that this may impact that monster's strength relative to its CR, making it stronger or weaker than its CR indicates. An obvious reason for this is that many monsters with the Spellcasting Trait achieve most of their damage via their spells.
Archetype: Attack Caster

Some monsters with Spellcasting achieve most or all of their combat relevance through this Trait, with the primary focus being on spells that inflict damage and/or conditions. For example, the Evil Mage from page 57 of the Starter Set only does 3 damage each round without its spells. But look at its Spell List...

  • Cantrips (at will): light, mage hand, shocking grasp
  • 1st Level (4 slots): charm person, magic missile
  • 2nd Level (3 slots): hold person, misty step

So this monster can cast Magic Missile using a level 2 slot three times, allowing it to do 14 Damage each round.

Archetype: Defense Caster

Some monsters don't use Spellcasting for direct aggression, instead their spells focus on assisting their allies. These creatures often play support roles in combat, healing and buffing, or rely on buffing their own attacks (ala "Gish" creatures). The Acolyte on p53 of the DM Basic Rules v0.1 is a good example of this archetype...

  • Cantrips (at will): light, sacred flame, thaumaturgy
  • 1st level (3 slots): bless, cure wounds, sanctuary

While the Acolyte does achieve its best damage using Sacred Flame that isn't its main role in combat. Instead this creature's best use is casting its 1st level spells.

Archetype: Utility Caster

Some creatures with a casting Trait don't use their magic skills for purposes that directly impact combat calculations, which isn't to say that that their spells have no direct impact on combat. All current examples of this are creatures with Innate Spellcasting, but there's no reason this can't be equally well implemented using Spellcasting. Let's consider the Rakshasa's spell list...

  • At will: detect thoughts, disguise self, mage hand, minor illusion
  • 3/day each: charm person, detect magic, invisibility, major image, suggestion
  • 1/day each: dominate person, fly, plane shift, true seeing

The Rakshasa's spell list clearly is not intended for head-on combat. Instead it is well suited to the creature's preference for working behind the scenes and for infiltration. This isn't to say its spells have no combat application, but in that context they are focussed on safely escaping confrontation rather than defeating enemies.

Archetype: Flavour Caster

We can expect to see the odd creatures with Spellcasting or Innate Spellcasting where the spells have no combat application whatsoever. The example that currently jumps to mind is the Yuan-ti Malison from the Horde Of The Dragon Queen...

  • At will: animal friendship (snakes only)
  • 3/day: suggestion

Very thematic, but with little or no combat application.

Archetypes vs Real Monsters

The thing is actual monsters closely matching the Archetypes listed above are pretty rare. There are some, as our examples show, but few of these are "pure" examples of their type. What's most common is for monsters to mix types. So a creature that is primarily an Attack Caster will frequently dabble in defensive casting and may also have a dash of utility casting.

This kind of selective, well-thought-out mixing is fine and can result in a more robust opponent. But I would again caution against going overboard and giving a Spellcaster too many spells. I feel a clear, succinct monster is easier to operate than an overly complex one. Of course, this is my own perspective and preference, not something based on any particular analysis or metrics.

Prediction: When we see casters of the Flavour archetype their Caster Level will be relatively lower (for their CR) than that of casters that focus on Attack or Defense casting.

 

Example: Human Pyromancer

OK let's apply this all to our pyromancer. Now what Traits do we want to give this monster? Spellcasting is obvious, as we've alluded to in both this and previous installments. But what about other Traits? I did toy with the idea of giving the Pyromancer something to mitigate damage to its allies, but I don't think that's really in keeping with the ideas behind this monster. Likewise some kind of spell penetration Trait could be appropriate, but I think that's simply going to complicate things without yielding any real benefit. So I've opted simply to go with Spellcasting.

Human Pyromancer

Medium humanoid (human), any alignment

Armor Class 12
Hit Points 77 (14d8+14)
Speed 30ft

STR
9 (-1)
DEX
14 (+2)
CON
12 (+1)
INT
19 (+4)
WIS
12 (+1)
CHA
12 (+1)

Saving Throws Dex +5, Int +7
Skills Arcana +7, Perception +4
Damage Resistances Fire
Lanuguages Common, Ignan
Challenge 5 (1,700 XP)

Spellcasting. The pyromancer is an 8th-level
   spellcaster. Its spellcasting ability is Intelligence
   (spell save DC15, +7 to hit with spell attacks). The
   pyromancer has the following wizards spells
   prepared.

   Cantrips (at will): dancing lights, fire bolt
   1st level (4 slots): burning hands, disguise self
   2nd level (3 slots): flaming sphere, invisibility
   3rd level (2 slots): fireball

The pyromancer is an intelligence-based caster loosely modelled on the wizard. We'll peg its Caster Level at 8 for now, 3 higher than its CR. The remaining stats are trivial to determine, as described above.

This brings us to its Spell List. We already know we are doing a fire-themed caster and as a glass cannon it makes sense for it to focus on the attack caster archetype. All this makes the core of spells we would choose pretty obvious: fire bolt, burning hands, flaming sphere and fireball. This works pretty nicely for the pyromancer being a ranged caster without a lot of close-quarter options - fireball will account for most of the pyromancer's damage and burning hands gives it some melee relevance.

Next we'll splash in some flavour and utility with dancing lights, disguise self and invisibility. I did think about adding mage armor or shield, but that makes our monster a lot more resiliant in melee combat than our original design. Instead I will be adding an interesting litle reaction in a later installment.

In the forthcoming Damage installment we'll reflect on how our example creature shapes up damage-wise and see if we need to make any adjustments.