Subclass bloat

[quinron]

With the release of Tasha's, I'm starting to worry that subclass bloat is going to be a real problem going forward (if it isn't already one now). I'm pretty sure this reaction was provoked as a total knee-jerk, but thinking through it, I've decided this bears some consideration.

My main concern is that subclasses carry a lot of content (always at least 3 levels' worth of features), but you only ever get one. Every subclass added in a new book adds one more opportunity cost that you have to balance your choice against, but there's not really an enhanced level of customization coming with it; you're typically just replacing a fixed set of features with a different fixed set of features.

I'm glad to see them adding things like new Fighting Styles, new Path of the Totem choices, new maneuvers, etc., that expand on the core subclasses, but I'm very conscious that these are always being added alongside new subclasses - new subclasses which, in almost every circumstance, totally invalidate these other new options. Which ties back into my main concern - new subclasses don't expand your options, they replace your options. Even the expanded subclass-specific options are totally irrelevant if, for example, you're playing a Champion instead of a Battle Master - both core options, but one of them is going to get a lot more support (and, due to the near-inevitable power creep, almost certainly become stronger) than the other.

Problem is, in the now-firmly-established idiom that 5e's created, I don't know how one might go about fixing or at least tempering this. The first idea that springs to mind is alternate class features, which I think Pathfinder employed quite well with their archetypes and which I think D&D could implement effectively. They toyed with it a bit in Tasha's, but like a lot of things in Tasha's, it feels like "toying with it" is as far as they got by the time the book went to publication.

I also wonder if this concern isn't just the system functioning as intended; 5e's initial design ethos was streamlining, which ipso facto means reducing options. I'm still worried about the content bloat, though; I'm foreseeing 3.5 PrC levels of redundant and/or useless subclasses by the time the edition ends.

[MrStabby]

I think that more content without customisation is kind of... ok.

The things you need to check for balance grows with the number of subclasses. The number of combinations explodes with the number of subclasses/options that can be combined.

By creating new content that doesnt interact strongly they avoid this.

On the other hand the leaning on proficiency bonus rather than class levels for more features now is a step in the opposite direction as your subsequent levels in class B impact your features from class A in multiclass.

Honestly, I think we are still a bit away from subclass bloat being an issue (I worry more about quality than quantity). I think that different races are more bloated than classes.

[MoiMagnus]

Subclass bloat is fine for me.

It's order of magnitude better than class bloat, and it's reasonable localised in term of features as long as they keep the new subclasses within the same design philosophy as the first ones ... which unfortunately doesn't seems to be the case (see new sorcerer subclasses), and is a much more worrying trend IMO than the bloat.

(Additionally, for most classes, subclasses don't start at level 1, so there is much less worrying about multiclassing by having a subclass bloat.)

Due to the much more "localised" nature of the subclass features, if a subclass is problematic at your table, you will have a much better time trying to houserule it and balance it than for a full class.

And in a no-multiclass game, the subclass choice is a one-time choice, possibly at session 0 like the race and class. It's not something like feats, magical objects or spells, that can jump out of nowhere in an existing campaign with unforeseen consequences with a lax enough DM.

DM-side, subclass bloat is probably the least threatening out of all the player-content-bloats.

[Theodoxus]

Subclasses are really the only place they can create new content. You mention Prestige Classes, and in a way, especially the level 3 subclasses, they ARE Prestige Classes. They have simpler pre-requisites (get to level 3 of your class), but they modify the base class in the same way.

WotC even notes your concern and provides a way to change your subclass if you feel a new one better captures your character (or you're (general you, not specific) a munchkin and just want the new shiny). But the game needs new content to remain viable. Players want new options. Not everyone likes the same classes. Even with Tasha's improvements, I'll never play a Sorcerer again, it's just not my bag. No matter what new Domains they come out with, I'll always play a Life Cleric - it fits my prototypical idea of a Cleric I've been using in my headcannon for 40 years and I'm not going to stop now. But other classes, each new thing might intrigue me enough to come out of my healer shell and try swinging an axe or shooting a bow for. The Hexblade finally convinced me to play a Warlock, and it was an enjoyable time for me. No longer was I relegated to EBing all the things like every other Warlock I've seen played. It was a true gish and I loved it. Had to retire him due to time pressures, but I'd probably bring it back out again.

The new summoning spells in Tasha's makes me want to try a micro-minionmaster. Probably a Necromancer. Likewise, the Beastmaster upgrade finally feels like a WoW Hunter (or really, the old Dark Age of Camelot Hunter, more specifically) and that makes me happy.

Yeah, it'd be nice if the same amount of QC were done for these books as the original PHB, DMG and MM were done - like, the same time were given to each, so if when Tasha's were released, the devs had been slowly cooking it for 3 years, not 6 months. The release rate would be the same, just each book would bake a LOT longer. But I get that no company works on that kind of time scale - faster product out means more money in - but it would be nice if the world didn't run that way sometimes.

[Amdy_vill]

so on the topic of Bloat, I don't think 5e will become bloated. I hear a lot of people talking about how 5e doesn't need more content and it's becoming too much to handle, as a dm of nearly 10 years I have found that unless you're playing hyper focuses system like bear heist or dread you will never know everything in the game and that's just something you will have to deal with, WotC makes money off of selling books. I have found the systems only really feel bloated when large amounts of content overlap, like having 6 subclasses for wizard that focus on summoning.

on the topic of expanding class options to make them feel more unique. this is something that they should do, WotC should also focus on making subclasses the can have expansions to their abilities, like how the totem warrior or battle master got new content.

Streamlining doesn't mean less content, it means easy to understand and use content. 3.5 problems wasn't that it wasn't streamlined it was that they published content too fast, which cause some streamlining problems but mostly just made dozens of overlapping ideas and some really bad unplay-tested things.

[OldTrees1]

Subclass design is less flexible than alternate class feature design.

Less flexible in this sense causes 2 symptoms:
1) To cover the same volume of character concepts, you need more subclasses than you would need ACFs.
2) Characters can use fewer subclasses than they can use ACFs.

This means there is more demand for more subclasses (to better represent characters) and less demand for more subclasses (because they are not relevant to many existing characters).

However subclasses are the way 5E chose, so you will see the number of subclasses grow proportional to the number of splat books. Which started rather slow in 5E, but might rapidly accelerate soon.

So, yes, expect some bloat by the end of the edition. This is not really a problem.

[MrStabby]

so on the topic of Bloat, I don't think 5e will become bloated. I hear a lot of people talking about how 5e doesn't need more content and it's becoming too much to handle, as a dm of nearly 10 years I have found that unless you're playing hyper focuses system like bear heist or dread you will never know everything in the game and that's just something you will have to deal with, WotC makes money off of selling books. I have found the systems only really feel bloated when large amounts of content overlap, like having 6 subclasses for wizard that focus on summoning.

on the topic of expanding class options to make them feel more unique. this is something that they should do, WotC should also focus on making subclasses the can have expansions to their abilities, like how the totem warrior or battle master got new content.

Streamlining doesn't mean less content, it means easy to understand and use content. 3.5 problems wasn't that it wasn't streamlined it was that they published content too fast, which cause some streamlining problems but mostly just made dozens of overlapping ideas and some really bad unplay-tested things.

I agree with the idea of both designing classes with more internal options to chose from (battlemaster manoeuvres, warlock invocations, fighting styles, spells etc.) and releasing more of them.

And yes, I like ACFs in general.

I am not sure that some of these options are untested - but I think that some of the testing has been done poorly. I.e. testing a new option in a party alongside an old option to see the extent to which it a) brings enough new to the table and b) doesnt overshadow or conflict with the old class.

[stoutstien]

I'm a fan of alternative class features as well but I can see the hesitation
some people have because it's one area where PC options has historically been that new is always better. I can count a least 4 subclasses that would work better as alternative features instead of trying to be complete option in thier own right. Even then I don't think we are at risk of subclass bloat yet out side of wizards and the druids which IMO struggle to be a class to begin with.

Race bloat is more annoying and talk about a prime spot for alternative features rather than just more for the sake of more.

[PhoenixPhyre]

Race bloat is more annoying and talk about a prime spot for alternative features rather than just more for the sake of more.

If I were to rank content types for power creep/bloatiness, I'd probably do it like this:

Low risk:
1. New monsters of non-beast type. DM only, generally. No risk to intra-party balance.
2. New subclasses. Because if you only get one (or maybe 2), the interactions are minimal. So you can do things with these that you couldn't do with a more mix-and-match system.

Medium risk:
3. New items. With a DM in control, these aren't so bad (but can get out of hand pretty fast with open access).
4. New feats. Much more of a risk. People start building their builds around these.
5. new races. Not so much for power creep, but for bloat. IMO, they should only publish setting specific races from here on. With the expectation that not all the PHB races are everywhere.
5.5 new beast-type monsters. Slight power creep for druids and polymorph.

High risk:
6. new classes. Here you get substantial interaction concerns.
7. new spells. This is a key power creep. Because every time you publish a new spell, all the spellcasters capabilities go up instantly and asynchronously. Especially "whole list" casters.
8. Anything that allows mix-and-match (ACFs, etc). Because then you have a combinatoric explosion of concerns. I'm a bit less concerned about variants that simply replace one set of features with a pre-defined alternate set, however.

As for pure bloat, the issue comes in when the field is full enough that any new option has a directly comparable old option. Because then you get
1. New strictly better than old --> old is now invalidated. I'd say some of the ranger options came darn close here, and some of the new sorcerer ones are pretty close.
2. New strictly worse than old --> why did you print it?

In either case, you've got a trap option. And I hate trap options.

[stoutstien]

I will say 5e has very minimum trap options which is nice. You have to actively try to deadend a PC and could still be highly effective. Hurray for high floor designs.

[MrStabby]

If I were to rank content types for power creep/bloatiness, I'd probably do it like this:

Low risk:
1. New monsters of non-beast type. DM only, generally. No risk to intra-party balance.
2. New subclasses. Because if you only get one (or maybe 2), the interactions are minimal. So you can do things with these that you couldn't do with a more mix-and-match system.

Medium risk:
3. New items. With a DM in control, these aren't so bad (but can get out of hand pretty fast with open access).
4. New feats. Much more of a risk. People start building their builds around these.
5. new races. Not so much for power creep, but for bloat. IMO, they should only publish setting specific races from here on. With the expectation that not all the PHB races are everywhere.
5.5 new beast-type monsters. Slight power creep for druids and polymorph.

High risk:
6. new classes. Here you get substantial interaction concerns.
7. new spells. This is a key power creep. Because every time you publish a new spell, all the spellcasters capabilities go up instantly and asynchronously. Especially "whole list" casters.
8. Anything that allows mix-and-match (ACFs, etc). Because then you have a combinatoric explosion of concerns. I'm a bit less concerned about variants that simply replace one set of features with a pre-defined alternate set, however.

As for pure bloat, the issue comes in when the field is full enough that any new option has a directly comparable old option. Because then you get
1. New strictly better than old --> old is now invalidated. I'd say some of the ranger options came darn close here, and some of the new sorcerer ones are pretty close.
2. New strictly worse than old --> why did you print it?

In either case, you've got a trap option. And I hate trap options.

I think that I would rate the risk of beast type monsters higher than you did for a few reasons:

1) I think there is some semblance of play-testing for most new player content. We can complain it isn't enough but there is soemthing that goes on. I am not sure monster manual content is checked so thoroughly. So the probabiliy of something bad slipping through is higher.
2) The classes that get polymorph on their spell list are already tending to be the most powerful in the game so there is less headroom before they become a problem... compared to say the risk of a new fighting style.
3) These give up nothing for the option. At least any new spell requires you to have one fewer other spell prepared for what it offers. This just straight adds another option
4) The flexability risk is huge - the potential stuff a beast could do but doesn't is big. There is a lot of space here.
5) There is a risk that someone might consider it a good plan to release higher CR beasts.
6) There is risk to wildshape, conjure animals AND polymorph - potentially three effects on a single character from a change that look innocuous
7) Since the MM got released there are more support options for beasts if the same MM mentality is used the impact of shepherd druids and twighlight clerics might be neglected.

There are similar arguments for most things that can be summoned, but beasts are the worst risk.

[ezekielraiden]

So. We can't add new classes, because that's bloat, and risks redundant options.

Now we can't add new subclasses, the things specifically made to be concept-specific focuses for classes so we wouldn't need to add more classes, because that's bloat, and risks redundant options.

Why not just come out and say, "I don't want supplemental material that adds character options"? Because that's literally what you're saying, just behind a smokescreen of "concerns" and other noncommittal phrasing. If you only want five subclasses per class, or some other arbitrary limit, just say it and be done with it.

[MrStabby]

So. We can't add new classes, because that's bloat, and risks redundant options.

Now we can't add new subclasses, the things specifically made to be concept-specific focuses for classes so we wouldn't need to add more classes, because that's bloat, and risks redundant options.

Why not just come out and say, "I don't want supplemental material that adds character options"? Because that's literally what you're saying, just behind a smokescreen of "concerns" and other noncommittal phrasing. If you only want five subclasses per class, or some other arbitrary limit, just say it and be done with it.

I think pretty much everyone here has mentioned something they would like to see more of? Be it classes , subclasses, feats, fighting styles, ACFs...

[stoutstien]

So. We can't add new classes, because that's bloat, and risks redundant options.

Now we can't add new subclasses, the things specifically made to be concept-specific focuses for classes so we wouldn't need to add more classes, because that's bloat, and risks redundant options.

Why not just come out and say, "I don't want supplemental material that adds character options"? Because that's literally what you're saying, just behind a smokescreen of "concerns" and other noncommittal phrasing. If you only want five subclasses per class, or some other arbitrary limit, just say it and be done with it.

Bloat is any added option that doesn't necessarily add any real value to the game. Sometimes it's not the new contents fault when some options are just so flexible so the new content need really strong theatrical elements to shine. Wizards are a good example here. I could see plenty of new subclass ideas but they need powerful setting ties not a bunch of new crunchy parts.

I'm on board for new classes as long as they spend as much time play testing as they did the arty. I'm not 100% satisfied with it but it is well balanced for as many new mechanics it has.

[TrueAlphaGamer]

I guess I'm echoing some ideas already touched on, but it does seem as though subclass 'bloat' or just the trend towards too many subclasses is necessitated by the design philosophy of minimizing the addition of new classes. In 5e, it's like the designers are construction workers who don't have permission from the zoning board to make any new buildings (new classes), so they just have to make the buildings they already have taller (new subclasses). It's mainly up to personal opinion whether this form is preferable, as some people may be quite content with the classes that are currently in place, or otherwise have their tastes satiated by the new additions.

Personally, I would want more classes, though that might be because I've gotten somewhat used to the general play style of the current ones and would want some variety that isn't just window dressing. Of course, that itself may just be due to discontent with the snails pace WotC prefer to go at when it comes to 5e books, but I do think there is room for yet unexplored class archetypes.

I suppose subclass bloat is better than class bloat, but it is perhaps disingenuous to assume that adding new classes would automatically (or inevitably) result in bloat. Though, again, would one rather prefer a set of content that is controlled to the point that it risks being stale, or a chaotic mess of options where character creation becomes like the wild west (with, perhaps, a higher likelihood of enjoyment)?

[MrStabby]

I used to be a bit against new classes. Then I saw the artificer and was dead against new classes... but since then I have had my view changed.

I hated the artificer, the theme of it, a lot of the mechanics, lack of it's own spells and a few other things.

The artificer still isnt for me and yes I am somewhat saddened that magitech can impose itself on my games, but I wasnt the target for the class. It does appeal to a LOT of people and (other than the alchemist) people seem pretty happy with the realisation (which is quite an achievement for something not overpowered).

The class doesnt really step on anyone elses toes, its mechanics are different and rewarding and, however much I dislike it, it does have it's own thematic identity.

New classes can work and I would be open to more of them especially if they are as well thought through and calibrated as the artificer.

New classes can actually be more robust - if you have an underperforming class you can boost it with new content whereas if you have a subclass that is weak you cant really boost it without boosting all the other subclasses that might be strong. For example if the artificer was weak as a class you could add more infusion options in splat, or add more spells. If something like a warlock subclass were to be underpowered (cough... undying... cough cough) then if you try making it better by adding class options like new invocations then you need something quite contrived to only boost that subclass.

[sophontteks]

Which ties back into my main concern - new subclasses don't expand your options, they replace your options.

Giving you more options is expanding your options. I think you'd need to expand upon this a little. Are you saying the new options are superior or something?

If you are worried about powercreep, I don't think the new options really made the old choices irrelevant. In many cases it made them better. The only definitive powercreep I see is the tashas sorcerers, and even then I feel like the new subclasses made the sorcerer more accessible to players. The old sorcerer sublcasses aren't very friendly to casual players.

Battlemaster has new options and remains a great fighter subclass.
Beastmaster was fixed.
Totem Barb remains one of the best barbarian subclasses.

Most of the duds were the same duds we had since the beginning. Like champion. It's just a bad subclass. Same with Berserker. It's be great if they fox these subclasses in new releases.

[PhoenixPhyre]

New classes can actually be more robust - if you have an underperforming class you can boost it with new content whereas if you have a subclass that is weak you cant really boost it without boosting all the other subclasses that might be strong. For example if the artificer was weak as a class you could add more infusion options in splat, or add more spells. If something like a warlock subclass were to be underpowered (cough... undying... cough cough) then if you try making it better by adding class options like new invocations then you need something quite contrived to only boost that subclass.

The issue with that is that WotC has made somewhat of a policy that new material doesn't assume that other, non-core material is present. So a class published in Random's Book of Stuff can't really get explicit support outside that book. Or at least it's more difficult than adding new material for core classes.

And personally, that policy is great (although sometimes frustrating). If book A assumes you have (non-core) book B in order to use most of its features, that's an issue. Plus that's a great way to get bloat and unforseen interactions. There's a reason they've reprinted stuff that only occurred in setting books before--so that they could add more support without running that risk.

[JonBeowulf]

I'm going full pragmatist here and say they got to publish something so we'll buy something so they can keep the train moving. It doesn't really matter what is being added because they gotta add SOMETHING.

Personally, I'd like to see a full-on variant sorcerer class and more fluff (and crunch) for warlock patrons.

[stoutstien]

One of the reasons I like the general mechanical idea of warlocks having 2 subclass options and the free form of invocations is it allows more room for adjustment compared to the the binary class/subclass style of design. Lots of room to add content. I'm kinda disappointed in the fact they haven't capitalized on this angle more.

One of the reasons I think the artificer is seen as such a good design(flavor aside) is it mimics that open ended customization and while they only have one subclass pathway they have complex party interactions to work with.

[MrStabby]

The issue with that is that WotC has made somewhat of a policy that new material doesn't assume that other, non-core material is present. So a class published in Random's Book of Stuff can't really get explicit support outside that book. Or at least it's more difficult than adding new material for core classes.

And personally, that policy is great (although sometimes frustrating). If book A assumes you have (non-core) book B in order to use most of its features, that's an issue. Plus that's a great way to get bloat and unforseen interactions. There's a reason they've reprinted stuff that only occurred in setting books before--so that they could add more support without running that risk.

Yeah, a good point. It leads to things like Tasha'needing to reprint core artificer to include a new subclass.

[PhoenixPhyre]

Yeah, a good point. It leads to things like Tasha'needing to reprint core artificer to include a new subclass.

Artificer wasn't core before. It still isn't. It used to be a setting-specific class (like bladesinger was as a subclass). Now it's promoted to a splat book, but not setting tied class.

And still, the next book can't include options tied directly to artificer without pain and being a waste/confusing to those who don't own Tasha.

[stoutstien]

Yeah, a good point. It leads to things like Tasha'needing to reprint core artificer to include a new subclass.

The artificer reprint did have some big changes from clarifying wording and making the action choice and scaling of the homunclus/steel defender cleaner to making the infusions less eberron dependent.

The homunclus being a second level infusion is a big change as well.

Proof that reprints are possible in 5e without the sky falling.

[MrStabby]

Artificer wasn't core before. It still isn't. It used to be a setting-specific class (like bladesinger was as a subclass). Now it's promoted to a splat book, but not setting tied class.

And still, the next book can't include options tied directly to artificer without pain and being a waste/confusing to those who don't own Tasha.

Yeah, sorry. Poor choice of words. By "core" I was meaning the class rather than subclass options.

I.e. in order to add a subclass in Tasha's they needed to reprint artificer.

[Pex]

I'm not so sure if an old Thing becoming obsolete is a bad thing. I get you don't want More POWR! to win D&D, but something being strictly better isn't inherently a bad thing. It would help if WOTC admitted a new subclass is intended to replace an old subclass for reasons. It's possible old ideas that used to work don't work anymore or there's a way to do it better or it didn't work in the first place. Why should it bother a DM if, as an example, no one ever plays a Champion again? If players don't like a subclass they don't like it, and there's nothing wrong with not liking it. It's not the DM's job to make players play it.

[Tanarii]

Replacing instead of enhancing is not only fine, it's the design working as intended.

The concern should be:
- subclass power creep, not subclass bloat. (E.g. Hexblade in Xanathars)
- other class feature replacement power creep (e.g. many replacement features in Tasha's)
- free "optional" class features (e.g. all non-replacement features and expanded spell lists in Tasha's)

[PhoenixPhyre]

I'm not so sure if an old Thing becoming obsolete is a bad thing. I get you don't want More POWR! to win D&D, but something being strictly better isn't inherently a bad thing. It would help if WOTC admitted a new subclass is intended to replace an old subclass for reasons. It's possible old ideas that used to work don't work anymore or there's a way to do it better or it didn't work in the first place. Why should it bother a DM if, as an example, no one ever plays a Champion again? If players don't like a subclass they don't like it, and there's nothing wrong with not liking it. It's not the DM's job to make players play it.

Because then you've got a trap option. And you've got the bad pay to win scenario: later books invalidate earlier ones, so you have to chase the treadmill to keep up. No thanks.

[stoutstien]

Because then you've got a trap option. And you've got the bad pay to win scenario: later books invalidate earlier ones, so you have to chase the treadmill to keep up. No thanks.

This is unavoidable to some degree unless a game is released in it's entity and never augmented or a game goes really heavy in setting specific material so you are dealing with a more simple base game with a bunch of different overlays. 5e is sort of the latter even if it fails to come out and says it.

[Lord Raziere]

Its an unfortunate part of DnD's "kinda-sorta flavorful" design choices. they make class options are halfway between being generic and being flavorful. this is because generally when designing options in an rpg you either go one of two routes:
1. you make everything so generic and flavorless that you can literally put any flavor on it and have it work, not attaching any descriptive fluff to it whatsoever

2. you design for the setting specifically and define options that only apply and have place within that setting, with in-setting explanation for all of them.

with DnD you don't get either. its semi-generic nature of having multiple settings means it can't fully go the 2 route because any of these options might get used in another setting. but they can't fully go the 1 route because its DnD and they must keep its identity as DnD distinct.

thus there is no actual logic for how subclasses, classes or anything get made and designed. if it was completely generic you could just forget the fluff and design purely mechanical subclasses that anyone could put anything they want on there and you could stop very soon as all the roles you'd want would be mechanically covered and you could leave people to do whatever they want with those.

While if it was completely setting based all the choices would be grounded in setting logic, thus any addition would be to fill it in, while not really allowing any room to refluff because the point would be to simulate being that thing specifically so you don't deviate from the fluff and do something else with it, thus necessitating options to specifically emulate another thing since you wouldn't be able to do that, unless you specifically build from the ground up to be that and fit it into the setting.

the way DnD does however is that provides some fluff that kinda is flavor, kinda explains what it is.....but always incredibly brief and general, making it shallow sort of fluff that is often unconnected anything else in the world with no real explanation as to how it fits into anything else. meaning you can kind of refluff it, but not fully. you can kind of try to take it as fluff to extrapolate on and embody in a more full form but you always end up having your own interpretation anyways, even though you can't truly change it. making it kind of frustrating in its wishy-washiness. you don't know fully what you can or can't refluff, because if you did, you won't need anymore options, because you can just refluff existing ones.

Worse, I think this half-hearted flavor choice is intentional so that WotC can make you keep buying books, not committing to simulation or mechanical purity, they keep the setting scope as wide as possible without providing you all the options to explore the possibilities on your own:
-If they committed to simulation and fluff that would drastically limit the setting you could play thus appealing to only a few fans at the expense of all the others. Fans would have to wait a long time to get their options to specifically play this or that. (for example you'd have a Forgotten Realms splatbook for Forgotten Realms ONLY classes and options only good for gaming in Forgotten Realms, none of which would make sense in say Eberron, which would need its only Eberron-only classes, including Eberron-only fighters and rogues and thus you wouldn't be able to play in eberron until Forgotten Realms was done first, unlike the current set up where the shallow fluff means you can easily play in one or the other without need specific things for them)
-If they committed to bland mechanical purity you'd only need a few books to as efficiently provide all the options you could possibly want to play DnD- lack of fluff would be your problem, this might turn off people who like flavor descriptions of things in books but more importantly it would mean the edition would be complete, and thus nothing would be left to sell as you would have all the options you could ever imagine in maybe three books max.

Thus WotC doesn't care about "bloat". They want you to buy more books, and since everyone loves character options, thats what they make. 5/6 people at the table are going to be players after all, so they're appealing to a majority of their fanbase because the game is designed that way. they don't care about DMs, by design they're in the minority.

[TrueAlphaGamer]

Because then you've got a trap option. And you've got the bad pay to win scenario: later books invalidate earlier ones, so you have to chase the treadmill to keep up. No thanks.

I can understand it being kind of lamentable that some options may be entirely eclipsed as new content is released, leading to the previous iterations seeming like a waste of effort or lost potential, but I don't really understand it as representative of things being "pay to win".

The game isn't really competitive, outside of some competition within optimization, as there aren't any strong instances of PvP, so having to 'compare' characters in terms of power level doesn't show up frequently (at least during play). Neither is the game played in a vacuum. You play at a table (or similar equivalent) with a few other people, as well as a referee, which can immediately veto mechanics if they seem broken or unfun. It's not really an MMO where you can solo for hours and pay for boosts/items/rewards. The DM controls the XP, the challenges, and the loot, so having a character be somewhat stronger than the others can be mitigated by a multitude of factors.

If stuff is released that does power-creep earlier content (which, if it does, is not likely to power-creep it to great extents, even among optimization circles), then it really won't be too hard to find it in this digital age. People will post screencaps of the book pages online, or it'll be available on Beyond, or show up in full on some forum post, or show up on one of the hundred other online resources for 5e. There is no real case where a person would be pressured to keep buying entire books solely for content relating to character creation.