5e Spell Progression is Weird (mechanically)

[Oramac]

So with all the talk of separate resource pools and in researching for my own homebrew, I had a thought: what if Divine Smite were a Channel Divinity option?

So I started looking into it. Ignoring the fact that 2014PHB paladins don't have CD, and OneDnD paladins get it after smite.

But the more I looked, the weirder it got. I made a google sheet to keep track of it all (https://docs.google.com/spreadsheets/d/1WaAlRAp-Vd87qv6DY42g6zERpF2HN6SLe69xYInqU7c/edit?usp=sharing). This includes full casters too, since I wanted to see how full and half compare. I didn't (yet) include quarter casters.

Some weird observations (in no particular order):


Proficiency Bonus scales poorly compared to total spell slots (especially for full casters)
PB as a percent of total spell slots for full casters follows no obvious pattern
Spell slots gained follows PB (which we already knew), but doesn't follow the Tiers of Play at all
This leads to some wacky scaling for Channel Divinity when compared to spell slots
Example: if you target CD = 50% of spell slots all the way down, you get a Every-Other-Level progression, except for levels 9 to 16, where half casters only get 1 increase, and full casters get single level and multi-level increases (also at 3rd & 4th level for full casters)
Targeting 60% gives a weird 2-4-2 progression for half casters (4 levels from 5-8 and 13-16, 2 levels everywhere else; 11-17 for full casters, plus a bunch of single-level increases)


I'm sure there's more wonkiness in there that I just haven't found yet. I think the real takeaway here is that Spellcasting is borked (which, again, we already knew). Not that proficiency is bad or wrong. I would also posit that the Tiers of Play are wrong as well, since tiers 2 & 3 take up 12 of the 20 total levels. Exacerbated by the fact that tier 3+ doesn't get played as much, so the whole scaling system is just kinda wasted in places.

Sheesh. I'm going to take a nap now. What else have I missed here?

[KorvinStarmast]

So I started looking into it. Ignoring the fact that 2014PHB paladins don't have CD Incorrect. They get it with their oath at 3rd level. Example:

Channel Divinity
When you take this oath at 3rd level, you gain the following two Channel Divinity options.
Sacred Weapon. As an action, you can imbue one weapon that you are holding with positive energy, using your Channel Divinity. For 1 minute, you add your Charisma modifier to attack rolls made with that weapon (with a minimum bonus of +1). The weapon also emits bright light in a 20-*‐‑foot radius and dim light 20 feet beyond that. If the weapon is not already magical, it becomes magical for the duration. You can end this effect on your turn as part of any other action. If you are no longer holding or carrying this weapon, or if you fall unconscious, this effect ends.

Turn the Unholy. As an action, you present your holy symbol and speak a prayer censuring fiends and undead, using your Channel Divinity. Each fiend or undead that can see or hear you within 30 feet of you must make a Wisdom saving throw. If the creature fails its saving throw, it is turned for 1 minute or until it takes damage. A turned creature must spend its turns trying to move as far away from you as it can, and it can’t willingly move to a space within 30 feet of you. It also can’t take reactions. For its action, it can use only the Dash action or try to escape from an effect that prevents it from moving. If there’s nowhere to move, the creature can use the Dodge action.

Proficiency Bonus scales poorly compared to total spell slots (especially for full casters)
It's irrelevant for "Known" casters. It defines attack bonus and save DC.

PB as a percent of total spell slots for full casters follows no obvious pattern
Not sure what pattern you are looking for. The two aren't even linked.

Spell slots gained follows PB (which we already knew), but doesn't follow the Tiers of Play at all Please explain what you mean by this in more detail.

This leads to some wacky scaling for Channel Divinity when compared to spell slots Those are separate resource pools. Tasha's option to burn a CD for a spell slot regeneration seems to me like a slipshod add on, but some people like it.

Example: if you target CD = 50% of spell slots all the way down, you get a Every-Other-Level progression, except for levels 9 to 16, where half casters only get 1 increase, and full casters get single level and multi-level increases (also at 3rd & 4th level for full casters)
They are separate resource pools by design. Tasha's makes an exception, and I am not convinced that it's a good idea.

Targeting 60% gives a weird 2-4-2 progression for half casters (4 levels from 5-8 and 13-16, 2 levels everywhere else; 11-17 for full casters, plus a bunch of single-level increases)
[/LIST] I am not convinced that there is an intended relationship.

I would also posit that the Tiers of Play are wrong as well, since tiers 2 & 3 take up 12 of the 20 total levels. Hardly. As described, Tier 1 is "beginner to competent phase."
Tier 4 is "End Game / Multiverse altering adventures" that not all tables want to tackle, and that no published adventures handle.

Tier 2 and 3 encompass a great deal of the genre, particularly the high fantasy aspect of it.

Sheesh. I'm going to take a nap now. What else have I missed here? I am not sure what assumptions you are operating under. A limit to spell casting for prepared spell casters is "you have to choose what you prepare" but known casters have no such constraint: their constraint is how many total spells they have to choose from.

[Oramac]

I am not sure what assumptions you are operating under.

That was right at the beginning of my post.


What if Divine Smite were a Channel Divinity option?

From there, comparing CD uses to spell uses to maintain a respectable number of Smites. Which led to the comparison of scaling uses and the unusual progression of PB and CD compared to number of spell slots available.

[KorvinStarmast]

That was right at the beginning of my post.
From there, comparing CD uses to spell uses to maintain a respectable number of Smites. Which led to the comparison of scaling uses and the unusual progression of PB and CD compared to number of spell slots available. Unlike the cleric, Paladin only gets on CD per short rest. Is it your intention to nerf/reduce the use of Divine Smite?

[GeneralVryth]

One thing you need to keep in mind (much so for full than half casters), is lower level spell slots tend to get relatively weaker over time as they don't scale the way a lot of non-spell options do. So what ends up happening is higher level lots take the place of lower level ones (which mostly get used for utility at that point).

[stoutstien]

One thing you need to keep in mind (much so for full than half casters), is lower level spell slots tend to get relatively weaker over time as they don't scale the way a lot of non-spell options do. So what ends up happening is higher level lots take the place of lower level ones (which mostly get used for utility at that point).

Eh. It hard to say that because so many low level spells have either stacking bonuses (shield/ AE) or effects that never become useless (fog cloud / FF).

Damage is about the only thing that this is true for and even then it never falls completely off.

[RogueJK]

One thing you need to keep in mind (much so for full than half casters), is lower level spell slots tend to get relatively weaker over time as they don't scale the way a lot of non-spell options do. So what ends up happening is higher level lots take the place of lower level ones (which mostly get used for utility at that point).

There are a number of exceptions to that generality.

A) Level 1 slots remain the most "efficient" use of a spell slot for Divine Smite purposes, getting you 2d8-3d8 extra damage for a 1st level slot, and only another +1d8 for each level above that.

B) There are a variety of non-utility 1st level spell options that remain very effective spells throughout all the Tiers:
-Bless
-Absorb Elements
-Shield
-Silvery Barbs
-Command
-Faerie Fire
-Healing Word
-Magic Missile (for forcing 3x Concentration checks, not necessarily the damage inflicted)
-Dissonant Whispers (for forcing OAs and the use of Reaction and subsequent movement on their turn, not necessarily the damage inflicted)

Similarly, there are quite a few non-utility 2nd level spells that remain very effective spells throughout all the Tiers:
-Aid (especially when used to yo-yo 2-3 allies back up from 0 HP; cheaper than Mass Healing Word)
-Blindness/Deafness
-Darkness (if combined with Blind Fighting or Devil's Sight)
-Mirror Image
-Misty Step
-Silence
-Spike Growth
-Spiritual Weapon
-Suggestion
-Tasha's Mind Whip (for the debuff, not necessarily the damage)
-Vortex Warp
-Web

[Oramac]

Unlike the cleric, Paladin only gets on CD per short rest. Is it your intention to nerf/reduce the use of Divine Smite?

It's a thought experiment. The intent is literally "What if Divine Smite were a Channel Divinity option", since it seems some people don't like it keying off spell slots.


Eh. It hard to say that because so many low level spells have either stacking bonuses (shield/ AE) or effects that never become useless (fog cloud / FF).


There are a variety of non-utility 1st level spell options that remain very effective spells throughout all the Tiers:

Similarly, there are quite a few non-utility 2nd level spells that remain very effective spells throughout all the Tiers:

snip

Yup. In the thought experiment, with Smite not using spell slots, there is no incentive to "save" spell slots for Smite.

The challenge is ensuring that [Number of CD Uses] is high enough to make Smite worth using when compared to [Number of Spell Uses].

[KorvinStarmast]

The challenge is ensuring that [Number of CD Uses] is high enough to make Smite worth using when compared to [Number of Spell Uses]. What number of CD per SR are you proposing?
As I noted, Paladins get one per short rest.
Is that enough Divine Smites for your purposes?
Clerics, at level 6, get 2 per SR or LR, and at level 18 3 per SR/LR.

Are you suggesting that paladin be modified to do that?

SR/LR have nothing to do with PB, which makes me wonder why the PB relationship was included in the post.
Seems to distract a little bit from the CD-to-Smite point.

[Oramac]

What number of CD per SR are you proposing?
snip

That's the whole point of the discussion. :smallbiggrin:

The OneDnD paladin gets the same 2-3-4 progression as the druid (and presumably the cleric once they revisit it). To me, this doesn't seem like enough. At least, not for Smite. Thus I thought about the PB progression, which has already been used quite a lot in other material. This feels better, but still scales poorly, leaving the paladin with only 40% of its "normal" smiting capability by 20th level.

That's why the Target Percent column of the spreadsheet is editable. So we can play with the numbers and see what (if anything) feels right. Or perhaps we decide that Smite keying off spell slots is in fact the better option.

[Kane0]

You could match it to the Warlock's Pact Magic slots (one/SR at level 1, two from 2-10, three from 11-16 and four from 17-20)

Or prof bonus times per short rest but the damage doesn't scale except for maybe level 11 or a subclass feature

Or flat twice per short rest at all levels with damage scaling with prof bonus

[PhoenixPhyre]

That's the whole point of the discussion. :smallbiggrin:

The OneDnD paladin gets the same 2-3-4 progression as the druid (and presumably the cleric once they revisit it). To me, this doesn't seem like enough. At least, not for Smite. Thus I thought about the PB progression, which has already been used quite a lot in other material. This feels better, but still scales poorly, leaving the paladin with only 40% of its "normal" smiting capability by 20th level.

That's why the Target Percent column of the spreadsheet is editable. So we can play with the numbers and see what (if anything) feels right. Or perhaps we decide that Smite keying off spell slots is in fact the better option.

Personally, I'd explore inverting the equation. Instead of paladins getting spell slots and then using them to smite, give them a pool of smite dice (d8s) and let them cast spells with them.

--------

That said, my loose investigation made me believe that paladins were generally designed to smite out of their bottom half of their spell slots. Effectively use 50% of their slots for smites, preferring lower to higher and saving their higher ones for real spells (hence the cap on damage meaning a 5th and a 4th are the same smite value). But that was very loose numerics done a while ago and may no longer be accurate.

[Amechra]

Honestly, spellcasters are wildly over-complicated for what they're trying to do in the first place. In practice, you essentially only have two kinds of spell slots:


Your top-level spell slots, which do big, powerful things.
Your low-level spell slots that you use for utility magic.



You could get a very similar playstyle by giving spellcasters a few major spell slots and then a scaling number of minor spell slots, with spells having different effects based off of the type of spell slot you cast them out of. A low-level Wizard might treat Invisibility as a major spell, while a higher level Wizard would treat it as a minor spell (with Greater Invisibility being the major version). Shield is always a minor spell, Fireball is always a major spell, etc.

[Kane0]

Honestly, spellcasters are wildly over-complicated for what they're trying to do in the first place. In practice, you essentially only have two kinds of spell slots:


Your top-level spell slots, which do big, powerful things.
Your low-level spell slots that you use for utility magic.


You could get a very similar playstyle by giving spellcasters a few major spell slots and then a scaling number of minor spell slots, with spells having different effects based off of the type of spell slot you cast them out of. A low-level Wizard might treat Invisibility as a major spell, while a higher level Wizard would treat it as a minor spell (with Greater Invisibility being the major version). Shield is always a minor spell, Fireball is always a major spell, etc.

Similarly, Pact magic slots and invocations/cantrips.

[Amechra]

Similarly, Pact magic slots and invocations/cantrips.

Pretty much.

[Yakk]

So 5e characters where balanced around a full day's effectiveness model.

You make a model adventure day, and you make a simplified model of the class, and you see how much it contributes.

Then for un-model'ed stuff, you eyeball it and make sure it is either not important (model-wise) or subsitutes for existing stuff in the model (ie, has a cost).

If I was to model a Paladin, I'd start with a model adventuring day of 2 short rests, 9 medium encounters, each lasting 3 rounds, and a 60% hit/5% crit rate, and using all slots to smite in encounters starting as early as possible (first round if possible).

That will give us a total damage budget baseline.

For a second order comparison, I'd consider what happens when I use slots for not-smiting. Is using Holy Weapon more efficient than a smite? Quite possibly. This will involve working out concentration break rates etc, but I can give it a range and look at when holy weapon is a better idea than a smite (or elemental weapon or whatever).

For something like Bless, I'd assume that if the paladin casts it it is because it is better than a smite slot. So such spellcasting can only improve the Paladin's performance. (I might test this, or not). Measuring its exact impact is hard, so I'd punt.

We can do the same for channel divinity.

...

Then if you want to change how smite works, what you'd have to do is make a model of how useful spell slots are for the paladin when not smiting. And then do a comparison of the two paladins.

You'd want to pay particular attention to the cases where the new model is better/worse, and try to figure out situations when those would occur, to see how big the change could be.

[Saelethil]

I think that Smite could work well as a Channel Divinity but it would need to be substantially more powerful than a basic 1st level 2d8 to make up for fewer uses.
Maybe 4d8 and enemies in your Aura’s radius make a saving throw taking half on a failure and none on a success? Then you just have to decide how the damage should scale.

[stoutstien]

So 5e characters where balanced around a full day's effectiveness model.

You make a model adventure day, and you make a simplified model of the class, and you see how much it contributes.

Then for un-model'ed stuff, you eyeball it and make sure it is either not important (model-wise) or subsitutes for existing stuff in the model (ie, has a cost).

If I was to model a Paladin, I'd start with a model adventuring day of 2 short rests, 9 medium encounters, each lasting 3 rounds, and a 60% hit/5% crit rate, and using all slots to smite in encounters starting as early as possible (first round if possible).

That will give us a total damage budget baseline.

For a second order comparison, I'd consider what happens when I use slots for not-smiting. Is using Holy Weapon more efficient than a smite? Quite possibly. This will involve working out concentration break rates etc, but I can give it a range and look at when holy weapon is a better idea than a smite (or elemental weapon or whatever).

For something like Bless, I'd assume that if the paladin casts it it is because it is better than a smite slot. So such spellcasting can only improve the Paladin's performance. (I might test this, or not). Measuring its exact impact is hard, so I'd punt.

We can do the same for channel divinity.

...

Then if you want to change how smite works, what you'd have to do is make a model of how useful spell slots are for the paladin when not smiting. And then do a comparison of the two paladins.

You'd want to pay particular attention to the cases where the new model is better/worse, and try to figure out situations when those would occur, to see how big the change could be.

Actually none of the PC options were ever balanced behind any attrition model. 5e whole thing was just making it feel like that well enough to allow each table to find the groove.