D&D General Path of Feats: a Superior Design than Subclasses

[SkidAce]

.... if your feat requires another to access you won't be getting it sooner than 8th level, which is often not too soon before most games end.

All I would say (agreed with your post), is that becuase of their value in character defintion, many DMs are being a little freeer
with feats.

Giving everyone a free feat at the beginning of near beginning of the campaign brings that "wont be getting it sooner" value down to about 4th level. which I feel is just right.

 

[saviirkad]

On metric value:
If the concern is that the metric itself isn’t valuable, then I need to know what you think a valuable combat metric would measure. Without either concrete features or an alternative metric, I can’t address the objection.

I don't think there ultimately is a combat metric to measure, because we are discussing a game with both hard and soft features that can be engaged with in various measures that incorporate both to different degrees. An analysis based on hard metrics will always exclude too much of the game to give valuable insight, even within the scope the metrics are pulled from.

Request for concrete features
If you think I’m wrong, then I need the specific support or control features you believe materially change that outcome so I can model them. Broad categories like “support” or “utility” aren’t actionable, I need the actual features.

I want to be clear that I don't think your math is wrong, I just think that its results have no impactful meaning beyond itself. If all one is doing is abiding by some sense of curiosity to see which of these separate dice rolls have a greater average total than the other, the math is right within its scope and curiosity is sated. But if the results are to be used in any way outside of that, up to and including validating the statement of "combat is abysmal," then its value in my opinion plummets to zero.

Can specific support or control features be added to the math and change the result of that math? Yes. I did name a few examples (commander's strike, heroism, bardic inspiration). Would it add value? I don't think so. If anything, what it suggests is that if you want to maximize just those metrics within  just this comparison, then the answer is both. You want to play either the fighter-bard or the barbarian opposite the other.

 

[Silam]

These below are all valid arguments, but they’re also not absolutely true. Feats can be designed either as cohesive or not.

i would assume the reasoning would be because subclasses can work on the guarantee that a character will have X, Y and Z features, both ones from baseclass and prior subclass levels, meaning they can make connections between those features in a way that feats can't really do, a monk subclass knows it gets to call upon ki/focus points (and by proxy, flurry of blows, patient defence and step of the wind), unarmoured defence, martial arts, unarmoured movement, a decent chance of high DEX or WIS and probably more i'm missing.

Let’s say we wanted to give Fighters more options, but by following a feat-based approach rather than a subclass approach. We could achieve this by setting a feat’s prereq as one of the Fighter class features. Here’s an example, below:

Come and Get Me
General Feat (Prerequisite: Level 4+, Second Wind Feature)
You have a way of getting under the skin of your enemies, making them overextend themselves and run out of steam before you.
Goad. Once per turn, you can talk or gesture to an enemy that can see you to entice them to attack you. Until the start of your next turn, that enemy has a -2 penalty on attack rolls against everyone else but you. If multiple people goad the same target, then the target suffers no penalty against any of them.
Payback. If you hit an opponent with a melee attack within one round of them hitting you with a melee attack, and then activate your Second Wind feature as normal, you deal extra damage equal to the amount healed for yourself.
Unstoppable. Once per day during a short rest, you can regain two uses of Second Wind rather than one.

Is there any reason why you wouldn’t let both a Champion and a Battlemaster have this feature? Then it’s better as a feat than as a subclass feature. Is it cohesive? I think so…

feats, by their nature of being individual packages, tend not to get those guarantees except for when they have prerequisites, which are generally not the most desired feats due to said gate to entry, they have to be accessible to a rogue, a barbarian, a warlock and all the other classes,

I think that’s an artificial constraint. If it were true, feats like War Caster would not exist, but clearly they do. Furthermore, there is nothing inherently wrong with making a feat have a prereq that’s even more restrictive than War Caster (as in the example above). Elven Accuracy is another example of a fairly restrictive prereq. Of course, it’s fine to make feats that are more accessible as well (e.g., having Weapon Mastery as a prereq which can be gotten from many classes and even from another feat). There is really no requirement that a feat needs to be accessible to all characters, neither in existing feats nor in future feats that have not been printed yet.

and even if you can access those features by another feat, feats in and of themselves are a very expensive currency that don't come frequently,

Agreed, but I don’t know why that’s a problem. Any class or subclass feature of level 11 or above is by definition a feature you can only ever get from a single class/subclass. So the whole bottom half of every class table is made up of very very expensive currency. But it’s fine right? I don’t think anyone is seriously complaining that they can’t be a Monk 10 / Fighter 11 (to get 3 attacks per action and 3 per bonus action). It’s just accepted as a fact of life. Getting somewhere between 4 and 7 feats is par for the course.

meaning for most people, if your feat requires another to access you won't be getting it sooner than 8th level, which is often not too soon before most games end.

Unless the other required feat is an origin feat, but yeah, some could take a while to get. Just like Epic Boons also take a while to get and most games never get them at all. But is it a design issue that Epic Boons exist at all? I don’t think so…

 

[FrogReaver]

I don't think there ultimately is a combat metric to measure, because we are discussing a game with both hard and soft features that can be engaged with in various measures that incorporate both to different degrees. An analysis based on hard metrics will always exclude too much of the game to give valuable insight, even within the scope the metrics are pulled from.

IMO, that describes almost every discipline where models are useful. No model captures everything, they only need to capture the parts that really matter for the question being asked.

I want to be clear that I don't think your math is wrong, I just think that its results have no impactful meaning beyond itself.
If all one is doing is abiding by some sense of curiosity to see which of these separate dice rolls have a greater average total than the other, the math is right within its scope and curiosity is sated. But if the results are to be used in any way outside of that, up to and including validating the statement of "combat is abysmal," then its value in my opinion plummets to zero.

Why do you think this? Do you apply the same standard to other models and metrics you rely on? If not, what makes combat uniquely unmodelable?

Can specific support or control features be added to the math and change the result of that math? Yes. I did name a few examples (commander's strike, heroism, bardic inspiration). Would it add value? I don't think so.

Why do you think adding relevant features wouldn’t add value? In every modeling discipline, adding relevant variables increases accuracy. What makes D&D combat the exception?

If anything, what it suggests is that if you want to maximize just those metrics within  just this comparison, then the answer is both. You want to play either the fighter-bard or the barbarian opposite the other.

How are you determining what “maximizes” anything when you aren’t comparing any metrics? What is that conclusion based on?

 

[I'm A Banana]

Let’s say we wanted to give Fighters more options, but by following a feat-based approach rather than a subclass approach. We could achieve this by setting a feat’s prereq as one of the Fighter class features. Here’s an example, below:

I think it's a bit of awkward design to patch an apparent class issues with a feat. If a class has an issue, fix the class.

There is really no requirement that a feat needs to be accessible to all characters, neither in existing feats nor in future feats that have not been printed yet.

True. There is, however, a practical limitation: the fewer characters a feat is relevant for, the less value that feat has to the system overall. If one has infinite space and design time (and play time!), then sure, no reason not to. But since those things are limited, a feat that delivers value to more types of characters is more useful to the game overall than a feat that only delivers value to 1/13th of the characters. Even War Caster is much more broadly useful than a single-class feat would be.

So the whole bottom half of every class table is made up of very very expensive currency. But it’s fine right?

I would personally argue that it's NOT fine, and it's baggage that D&D has been needlessly committed to since at least 2e!

 

[Silam]

I think it's a bit of awkward design to patch an apparent class issues with a feat. If a class has an issue, fix the class.

It’s not an issue with the class at all though!

First of all, it’s a melee-oriented feature. I thought of making it compatible with range but it diluted the flavor a lot and I didn’t like it. But I also don’t want to force Fighters into the melee niche. If the Arcane Archer comes out as a Fighter subclass (which I think is not as good as coming out as a prestige feat path, but WotC doesn’t care about my opinion so it will likely indeed come out as a subclass), then giving this feat’s ability to the Fighter would be deadweight to the AA.

Secondly, I considered making it a Fighting Style, which would have made it available to Paladins, etc. But I wanted to tap into the Second Wind mechanic because it felt right thematically for this to be related to endurance. Furthermore, Paladins already have a bunch of smites and I don’t think they need one more. If a Paladin wants to go out of their way to get hold of Second Wind then they could still take that Come and Get It homebrew feat, though the feat will be more rewarding to a pure Fighter because of how Second Wind is designed and how it scales based on Fighter level. Finally…

Third, I’m not opposed to having a feat which grants Second Wind and therefore fulfills the prereq of the above feat and technically makes it available to all builds, though it would probably not be as effective as the scalable Second Wind Fighters get… just like Metamagic Adept is nice but ultimately quite a lot weaker than a Sorc 2 dip.

There is, however, a practical limitation: the fewer characters a feat is relevant for, the less value that feat has to the system overall. If one has infinite space and design time (and play time!), then sure, no reason not to. But since those things are limited, a feat that delivers value to more types of characters is more useful to the game overall than a feat that only delivers value to 1/13th of the characters.

13 includes Artificer, I assume? If we assume 4 subclasses per class, then there are 52 of them. Therefore, designing one new subclass increases that to 53 and if we assume equal popularity then it means it’s applicable to a paltry 1/53rd of single-class builds. In comparison, a feat which is applicable to 1/13th of single-class builds is quite a bit more applicable than a subclass.

I would personally argue that it's NOT fine, and it's baggage that D&D has been needlessly committed to since at least 2e!

Now that’s a hot take

Would love to see you elaborate on that

 

[saviirkad]

IMO, that describes almost every discipline where models are useful. No model captures everything, they only need to capture the parts that really matter for the question being asked.

Why do you think this? Do you apply the same standard to other models and metrics you rely on? If not, what makes combat uniquely unmodelable?

Why do you think adding relevant features wouldn’t add value? In every modeling discipline, adding relevant variables increases accuracy. What makes D&D combat the exception?

This back and forth should only go on for so long, and that point probably already passed, so I'm going to make this my last response on the matter.

But there is something fundamentally missing here on what a model is or can be used for. No, D&D combat is not a unique exception, it's a consideration any model should be evaluated for.

I fully understand that when you have the mind for numbers analysis, it becomes a go-to tool for many a curiosity. I certainly do it. But there's scratching the itch and there's building or seeking usefulness. I'm always on the look out when building a work product model that I'm not including a metric that poisons the data set just because I was curious or interested about it. If there is some relevancy to it, then I need to make sure when provided up that it includes a detailed explanation of where it fails and why other factors have stronger relevance.

How are you determining what “maximizes” anything when you aren’t comparing any metrics? What is that conclusion based on?

Nothing new or unstated, and it's not a conclusion.

 

[FrogReaver]

This back and forth should only go on for so long, and that point probably already passed, so I'm going to make this my last response on the matter.

But there is something fundamentally missing here on what a model is or can be used for. No, D&D combat is not a unique exception, it's a consideration any model should be evaluated for.

I fully understand that when you have the mind for numbers analysis, it becomes a go-to tool for many a curiosity. I certainly do it. But there's scratching the itch and there's building or seeking usefulness. I'm always on the look out when building a work product model that I'm not including a metric that poisons the data set just because I was curious or interested about it. If there is some relevancy to it, then I need to make sure when provided up that it includes a detailed explanation of where it fails and why other factors have stronger relevance.

Given that, I’m not really expecting a reply - but I do want to point out that the objection is still very vague.

I’ve been trying to ask what specifically is missing from the model so it can be evaluated or incorporated. So far, the critique seems to rely on unspecified factors that aren’t being named or described in a way that can be worked with.

If there are concrete elements that materially affect outcomes, I’m open to including or modeling them. If they can’t be modeled, I’m also willing to acknowledge that.

If that clarification isn’t forthcoming, then I agree this exchange has probably run its course.

 

[I'm A Banana]

It’s not an issue with the class at all though!

I'd basically agree.

13 includes Artificer, I assume? If we assume 4 subclasses per class, then there are 52 of them. Therefore, designing one new subclass increases that to 53 and if we assume equal popularity then it means it’s applicable to a paltry 1/53rd of single-class builds. In comparison, a feat which is applicable to 1/13th of single-class builds is quite a bit more applicable than a subclass.

I do think this is a problem that D&D class design does keep running up against, but ultimately a subclass is a way to provide customization within a class. A feat has a different design goal -- a useful feature for many different kinds of characters. If you swap out subclasses for feats, and re-write feats to fit that design goal, you've ultimately just re-invented the wheel. The difference between an Arcane Shot feature that a subclass gets and an Arcane Shot feat that only Fighters can select that locks them into a feat chain is...mostly academic.

Now that’s a hot take

Would love to see you elaborate on that

I'm a big believer in designing games to be played. D&D -- and most TTRPGs -- are already such a huge time commitment. It takes about 2 years to go from 1-20, and a lot of tables last a lot less time than that (especially in the high school / college / early adult years where 2 years can mean different hometowns, different careers, marriages, children, etc.).

So I'm broadly in favor of a tighter game with fewer levels. This also helps a few big design goals that I think are pretty commonly advocated for in D&D (for instance, high-level spells being too disruptive to the game's expectations aren't a problem if we don't have archmage PC's that can cast them).

 

[Corinnguard]

A feat has a different design goal

I see feats as something the character learned while they were adventuring. Something that's kind of independent from whatever class they happen to be. I say kind of because sometimes a player picks a feat because of their class. But there probably are instances of a player picking a feat regardless of their class. Ex. A wizard picking the Nimble feat, a feat that's more suitable for DEX-based martials, in order to subvert a class stereotype.

Since 5.5e now have ASIs as a feat, it could be viewed by others as the character learning ways to be stronger, faster, hardier, smarter, wiser or charismatic. Ex. a +2 ASI to STR= physical strength training during the character's downtime.