Dungeons and Dragons Wiki talk:Categorical Balance Ranges

Ditching the SGT

I've talked about it before, but I'm actually going to propose something that we can hack away at (or totally discard if we think it is stupid).

First off, I think it would be good to define the goals of a good balance system:

• Allows categories for homebrew to fit in that fit certain playstyles for how strong characters are expected to be and what kind of capabilities they have.
• Is capable of being applied to any sort of homebrew that introduces character options and preferably can even be applied to things DMs use (like monsters).
• Is easy to use and, generally, won't cause huge arguments over what category a particular article should fit in.
• Can categorize utility abilities as effectively as combat abilities, because, really, a +100 to spellcraft should not be in the same category as a +2.

Just to emphasize why I think the SGT falls short:

• It isn't applied to items (or at least the current version isn't).
• The challenges we compare against are pretty much a categorical pass/fail. "Can I fly? Nope. I fail that one. Can I use a strong ranged attack? Yep. Okay I pass that one".
  • This can lead to a class (if we take it literally) where the class feature at 1st level is "flip a coin, if you get heads, you win the encounter, if you get tails, you lose". This would be considered rogue level. We have no base definition to say that this is a really stupid idea.
• The challenges are mostly combat oriented, but we try to throw in a few utility challenges to hopefully offset any classes with tons of utility but little combat.

My idea is to make a list of general traits that fit each balance level. A good article will be able to meet the traits for the balance level, but not have ANY in a higher balance level and try to have few in lower balance levels. Otherwise, we can call it poorly designed. Note: this means a lot of SRD classes would be considered poorly designed, which I think most people can agree on anyways.

I'm going to use our current naming conventions, but they are definitely not set in stone... The project page this talk page links to is basically the draft we can use to work away on this idea.

The main thing is that a class should fit for the most part into one category. It's okay for some overlap in definitions as well. I feel that once you cross a certain threshold upward, a class should be considered part of that balance level. And if it is generally monk-level, for instance, with a couple fighter level abilities, it probably should be either nerfed or beefed up.

Thoughts? --Aarnott 18:17, 17 April 2012 (UTC)

One thing to note. Feel free to come up with better definitions. This is a very rough draft. The main illustration here is how it would be implemented. --Aarnott 18:46, 17 April 2012 (UTC)

I think I like this better than what we have. With some clear guidelines, that can always be updated when edge cases show up and are argued, we could have a very solid way of defining classes without getting into the subjectivity of the SGT, items, etc. --Ghostwheel 23:29, 17 April 2012 (UTC)

I got asked to take a look at this and chime in on text. I don't really care, do what you want. Surgo 15:15, 18 April 2012 (UTC)

So are there any objections to moving away from the SGT and more towards a checklist system similar to what Aarnott put up? I like it a lot, and even if things don't fit perfectly, we can fit any specific material into the place where it mostly closely appears to fit. --Ghostwheel 23:32, 19 April 2012 (UTC)

I like this, except one thing which I think should be changed. Absolute statements like "no spells" and "no save or dies" should be eliminated and replaced with something a little more lax. Otherwise, there will be those who take those particular criteria as gospel, and stupid arguments will arise. EDIT: GW changed them. But it's something to keep in mind if we add criteria in the future.

Also, even with that change, arguments will arise. There should be a way of settling them. Maybe have some designated "experts" on balance. Balance would be determined by general consensus of users (including those experts), unless an expert feels that there is no informed consensus, in which case they make an official ruling. The distinction that the consensus must be informed is important, so that we can step in if a dozen people are commenting on a wizard clone and saying, "i tink dis is rogue;" sure, there's a consensus, but no justification is given. If experts disagree, the majority among them is what we go with. --—Preceding unsigned comment added by DanielDraco (talk • contribs) at

"Let the experts decide" is a poor way to establish a more objective definition of balance than the SGT. Given that classes, feats, items, and spells can (and should, in my incredibly humble opinion) do all sorts of totally freaky things, coming up with definitive and specific limitations is always going to result in a fair number of holes. This is why I feel that the SGT's flexibility, which lets the tester use tactics that befit the abilities of their creation, is a more appropriate fit for balancing things in a game as broad as D&D. Unfortunately, the SGT doesn't define any real method for picking terrains or enemy tactics, doesn't define how the classed enemies are built, completely ignores equipment, and has no way of determining balance points for items, feats, spells, or monsters.

As our balance points stand currently, the SGT is our last-ditch way to settle arguments, and every person has a slightly different set of "a little more lax" statements that they use to judge balance most of the time. --Foxwarrior 02:29, 20 April 2012 (UTC)

“

Given that classes, feats, items, and spells can (and should, in my incredibly humble opinion) do all sorts of totally freaky things, coming up with definitive and specific limitations is always going to result in a fair number of holes.

”

Yes, and that is precisely why absolute objectivity is an impossible and silly goal. And why attempting such an objective measure by using something like the SGT is just plain deluded. There HAS to be a judgement call. There is simply no other way to get a good measure. And if the community at large fails to generate a consensus through intelligent discourse, then some smaller group has to make a decision, and they should be people who have demonstrated themselves to have sound judgement. --DanielDraco 02:41, 20 April 2012 (UTC)

We had a group of experts once, and it ended badly. Any proposal to recreate a privileged group along those lines seems similarly doomed to failure in a volunteer community, and I'd strongly consider vetoing it. Due to the flux of participants, any disagreements need to be capable of being resolved by fair community discussion and vote or we may as well just use admin/bureaucrat fiat and call it such.

Also, and this is a totally selfish request, but can someone who cares about making this happen strip the proposal out and put it in a new page with some formatting? It's work to follow right now, and I really don't feel like giving it the read it might deserve. - Tarkisflux ^Talk 04:39, 20 April 2012 (UTC)

This is not the same. The idea of my suggestion is to have decisions primarily made by the community. The idea of "experts" is only to moderate it when that doesn't work out. The experts are secondary. The driving force is community decisions. It's much the same as how anyone on the wiki can make an edit, but the small group of admins monitors those edits to make sure things are going the way they should.

Really, the focus of my suggestion is to put this guidlines system in place, and have the categorization decided by community. I'm basically suggesting a decision-making process for these guidlines where it starts with the author setting it to what they think it is; if that fails, the community steps in and sets it to what they think it is; if that fails, we have a group of people to make a final decision.

All that said, I don't want to draw too much focus away from the topic at hand. It's just a possible specific implementation of the guidelines suggested. --DanielDraco 04:54, 20 April 2012 (UTC)

Ya know. If I could just add more than one Balance Point to an Article. It would really solve all of 'my' problems. IE: This item ranges in power from 'This Level' to 'This Level' depending on play style. I think the DM's who browse the page would appreciate that. Rather than hoping for a very specific level and having it misrepresented somehow. --Jay Freedman 11:11, 20 April 2012 (UTC)

Jay - Adding more than one is certainly a possibility. The issue is, do we really want to make that a sort of "encouraged" thing? The problem is, most of the time, if something is falling in several balance ranges, it is probably poorly designed. I can go into detail if you want me to enumerate the reasons why, but I'll leave it as an unsubstantiated statement unless you want me to talk in more detail on it.

That being said, it does handle a problem I've seen where people have two conflicting views of what a balance point should mean. On the one hand, there is the view that a DM should restrict anything of a higher balance point in order to avoid having players take options that totally dominate the game and make other players feel useless. On the other hand, there is the view that a DM should recommend players avoid anything of a lower balance point so that they can be sure the player's character will be able to contribute to the group at an adequate level. Having multiple balance points/ranges for an article could solve this problem in that both views can be satisfied. Restricting content to fall within a single balance point also solves this problem. I'm not sure which is better, but based on the design faults I usually see in articles that would fall in multiple ranges, I tend to prefer the latter.

DD - "... attempting such an objective measure by using something like the SGT is just plain deluded." That's exactly my main issue with the SGT. I'm hoping this idea is a good step away from that. I should clarify the main goal here: balance points are a categorical determination that we try to find a best fit for. I realized in my above post where I said there should be nothing stronger than the balance range is probably not a good requirement. A good article will do its best to avoid having capabilities both above and below, but it really is a judgement call if it is a good fit for a balance range. There is no objective scale such as "7 of the abilities out of 10 match rogue level". Instead, the categories just try to enumerate defining traits that the balance range should have and a best fit is picked.

This more open way of inclusion even goes a little against my own beliefs of good balance, but I think it works better to allow for the variety of beliefs that exist here. The fact is, my idea of rogue level is not necessarily the right way rogue level should be. I recognize that many people would feel that a UMD use of a staff of Irresistible Dance is rogue level. I don't believe this, but I'm cool having it show up with the same tag.

So, the main point of this idea is "can we come up with a set of definitions for each balance range"? If we can, then let's try figuring out which ones are crucial to each range and get them listed. --Aarnott 15:33, 20 April 2012 (UTC)

How would this affect Unquantifiable classes?--Ideasmith 03:31, 21 April 2012 (UTC)

Has anyone checked this against the existing hombebrew classes? I am guessing not, since it's a lot to do. --Ideasmith 03:31, 21 April 2012 (UTC)

→Reverted indentation to one colon

I would be one of those people who thinks that Irresistable Dance via UMD is rogue level, mainly due to opportunity cost and resistability reasons. It's much less awesome when it's something you have to invest in, can't easily swap out when you know you're fighting [Mind-affecting] immune foes, and can't boost the CL check when you hit SR. That this sort of thing doesn't really come out of the current lists is problematic I think, but I don't know how to add it in without moving to a setup like I suggested in response to the feat question.

Ideasmith - We have 9 unquantifiable classes, and they are those that generally depend on a secondary class choice. They're only unquant because they could be any of the ranges, depending on the class choice made. As such, this probably wouldn't impact them at all. As for checking this against existing homebrew, this looks to be more a codification of balance range views in a checklist setup. I don't see many reassignments happening, and don't think we should be worried about it as long as we stick with a similar set of qualifiers. - Tarkisflux ^Talk 17:34, 21 April 2012 (UTC)

Aarnott - I guess that makes sense. When someone adds a class to my campaign I generally look for the optimization they are going for. The build that makes that class really sing. I never expect a veteran player to goof-up their build or nerf themselves somehow. As a DM, I am always looking for the highest grade of power they class or item can offer. The best of the best. I don't worry when a pro picks a fighter build. I do worry when a noob picks a fighter build. So I guess it would make sense to look for the highest potential in an article when creating a Balance Point for it. If it falls short, then it's probably just player error or noob-ism. Thanks. ...Continue! --Jay Freedman 01:24, 22 April 2012 (UTC)

Feats

So... how do we do this for feats? Just ballpark it like we've done so far without real guidelines? It's just that feats range such a wide gamut of possibilities that it might be hard to give them benchmarks... --Ghostwheel 23:36, 20 April 2012 (UTC)

I think it's a good goal actually, and would help with a concern I've had a problem wording. With benchmarks for Feats (which likely apply to class abilities as well) and Spells (which likely apply to Powers and other per day things) and abilities in general, it's possible to define the ranges in terms of their components instead of their totals. You could get class balance out of the average or extreme outlier of it's components, like we more or less do already anyway, but we wouldn't have to worry about edge cases because we'd already have a higher granularity level to fall back on. Class ranges could be written in a more general way at that point, which would make me happier. The current lists feel pretty narrow, and I'd rather not append edge cases forever.

Actual benchmarks probably won't be that hard. Write up a list of abilities and resistances that the game hands out (I've probably got one around here somewhere), and write up a suggested level when it would be ok to get it on a per day basis and a level when it would be ok to get it on an at-will or per-encounter basis for each range. Then add things we missed from there. - Tarkisflux ^Talk 02:05, 21 April 2012 (UTC)

If every player got the same number of feats you might be able to add Balance Points to them. But some players get more feats than others, and that alone can push a player above the games level. Kinda like how a +1 sword in the hands of the Barbarian is very different in the hands of a Wizard. Base Attack Bonus and all. If feats were slotted and unique for each class this might work. But since any player can pick most any feat? And how some feats have prerequisite synergies and trees? This would be a very tough thing to do. --Jay Freedman 01:31, 22 April 2012 (UTC)

Don't be stupid. Barring bonus feats from classes (which factor into a class's power), characters of level X do all have the same number of feats. --Ghostwheel 03:15, 22 April 2012 (UTC)

Careless comment, I admit. After all, Humans and Flaw Feats, (if you use them at all), would only add an additional number of one or two. But feats still feel like puzzle pieces to me. Only in combination can you see the picture the player is trying to build. Even as a DM, I have to take each feat added to a character sheet in perspective. Especially in specific campaigns. And if I myself can't see how Balance Points added to feats would help me personally as a DM, I guess I can't recommend them to others. If you feel they would help you then forge ahead. By all means. But it won't save this DM any time at the table at all. Haha. --Jay Freedman 09:08, 22 April 2012 (UTC)

That's probably fueled in part by the fact that characters are stuck with classes that are not pre-optimizedand and have to take particular feats in order to be effective. That's a condemnation of the classes, not the feats themselves. Ideally, you could take whatever feats you wanted and still be largely effective in a game (case-in-point: wizards); this effectually establishes them as stand-alone pieces unless it's a key part of the class (like the fighter), and thus shows that they're deserving of their own Balance Points. - TG Cid 15:18, 22 April 2012 (UTC)

If a feat gave you a +2 to will saves, that's about the same as a class feature giving you a +2 to will saves. Class features give it out at a certain level, and feats make it available at a certain level based on prereqs (with minor interactions from bonus class feats and flaws). The important thing is that the character got a +2 to will saves, and you can determine the relative strength of that sort of bonus and determine when, if ever, it would be appropriate to receive in a game of X balance.

Now, that's a weak example, intentionally so, but it should illustrate that what we care about is the ability itself and the acquisition level. The actual method of acquisition is less relevant. It also illustrates the potential problem of determining stacking strength, but I'm willing to punt on that one and write it off for now. - Tarkisflux ^Talk 17:09, 22 April 2012 (UTC)

Feats do tend to be synergistic. But then again, so do class features. And items. And spells. It's a defining trait of third edition that you can take little abilities and put them together to create big benefits. But we wouldn't look a base class which is monk-level as a standalone class, then declare it wizard-level because an innocuous ability it gets at level 1 synergizes in some broken way with the abilities of some random PrC. We would call that combo broken, and take the class at face value. By the Pun-Pun Principle of "kobolds are not overpowered," we should look at each feat individually. --DanielDraco 01:14, 23 April 2012 (UTC)

...and you can determine the relative strength of that sort of bonus and determine when, if ever, it would be appropriate to receive in a game of X balance. - Tarkisflux. ...Your totally right. I myself judge the strength of a class feature or feat based on what minimum level it can be received and then compare it to a wizard's class features for perspective. This is how I determine 'Base Class Balance' for my campaigns. If did the same thing, but compared it to a Monk's, Fighter's, and Rogue's features for perspective. Then I could likely judge any feat or class feature for perspective in 'X Game's Balance'. Fascinating. ...Unfortunately, this means that many feats are giant variables-in-power based upon the specific campaign setting and it's included materials. Ex: Power Attack is great among Monk Level Games. Power Attack sucks among Wizard Level Games. Doesn't that give my 'Power Attack' feat or 'SorD' feat at least 4 levels of balance. One for each tier of play? --Jay Freedman 04:43, 24 April 2012 (UTC)