So I have a legitimate question, and I'm asking in good faith. Is there a reason why your posts seem to be angry and insulting? Is it intentional, or do you not realize it?
Because I have no problem with you (or anybody) preferring point buy. The standard array and point buy are essential approaches if you prefer to plan your character out ahead of time, especially if you've preplanned things like multi classing, feats, etc. If you are the type that likes character builds, it gives you a known foundation to work from. They are usually fast approaches as well, and ensure "fairness" when people are creating characters at home.
My reasons for preferring rolling are because my preferences differ from those and also in several areas you've stated, and for several reasons you haven't.
1. Not a primary concern of ours.
2. See below.
3. We are playing what we want to play - D&D. As for specific classes, we prefer that that certain classes are as rare in the campaign as they are in the game world, and develop rules for that purpose (part of the response for #2).
4. Any character that fits within the boundaries of the rules cannot be, by definition, overpowered. They can be more powerful than other characters. We're OK with that. If we think something is too much, then we alter the rules (boundaries) to fix that.
5. We allow folks to roll more characters for many reasons, the primary reasons, however, are not because the first 5 were bad.
6. We enjoy session 0 for character creation, along with many other reasons. If we were having a session 0 only to create characters and we were trying to save time at the table, then the players would create their characters at home. We've done that plenty as well (and still do), but we prefer to do it at the table (also part of the response below).
I have literally sat tat the table too and had people create characters in 10 minutes too. If speed is your primary goal, then great! That's not our primary goal in character creation.
As the home group came together, and I/we* started tweaking the 5e rules, character generation was naturally one of the areas that we addressed.
We didn't like that 5e felt "too easy" as written, and that the characters felt "too powerful" either. This is entirely subjective, and there are a lot of reasons we felt that way. In addition, the characters tended to feel very similar, again for various reasons. In regards to generating ability scores, we asked what everybody preferred. Without exception, they all preferred to roll. There were a variety of reasons, but that was the result.
So as I'm tweaking the rules, with that information as well, I started questioning what I/we liked (or not) and why.
1. We much preferred to generate characters together. The social aspect of it is a big part, but also because there are more people available for ideas, and we're also tying that character to the ongoing campaign. This is an important enough aspect for us that when somebody new joins, or when somebody is creating a new character, we either make time to get together, or we do it at one of the regular sessions, instead of playing.
2. We strongly disliked that with point buy or a standard array that it became difficult to create "interesting" arrays. That is, when you are deciding where to put particular scores you tend to default to what makes the most sense for a specific class. So characters of those classes were very similar stat-wise. If you don't have a class or goal in mind, it can be quite difficult to assign stats.
3. Point buy and standard arrays didn't provide any inspiration. It was OK if you were sitting down with an idea of what you wanted the character to be, but without any random generation of anything, there was no catalyst for an idea. On the other hand, sitting down and rolling (as I'm doing now) STR 13, INT 12, WIS 11, DEX 10, CON 14, CHA 13 might.
4. Those stats I just rolled are perfect, because they also play into the next point. We like the rules to support the game world. So one of our rules (rather than a mechanical one) is that at least two of every three characters is human. We're not trying to incentivize ourselves by making the mechanical aspects of the race in the game more attractive. It's just a simple demographic thing, 80% of the world is human, and we want that to be largely represented in the characters we make. In addition, we want the stat generation to reflect the general population. So we do have mechanical rules for level limits, qualifying for classes, etc. We like that it's hard to roll a paladin. That's the way we want it. We want the most common characters to be fighter and rogue, and we want that to be supported by the rules.
Part of that is that most people are average, or slightly above average in the area they focus. Our point-buy number is 69 - enough to make three 11s and three 12s. We didn't stick with a traditional method of rolling, because 4d6dL trends higher than we like, and 3d6 trends lower. So we came up (after testing the math) with 3d6 reroll 1s, once. It's not perfect, but it's much closer.
Because I/we aren't the only people that play in our games, and we recognize that we haven't set any boundaries for low/high rolls, we want to make it more feasible for other players who don't share the same thought process as us. In addition, we recognize that sometimes you just roll something bad that's not working for you. And really, our rules of "roll 6 sets and pick one" are to put something on paper. We really haven't limited people to 6, but on the other hand, if somebody is on roll #18, then they probably have different goals in mind. It's also why we allow the standard array or point buy too.
When you're sitting at my table, I encourage you to use the system that the table has chosen for their preferred method. But the other options in the PHB are open to you as well. Recognize that you will probably have higher abilities than most characters at the table (which is usually why they want it, since they are used to a point buy/array of 72 points or more). But that's fine, we can work with that.
--
So I'll respond more directly to your point #2. We aren't making stupid rules to get good rolls. We're making (in my opinion) well considered decisions to identify our goals, look at the process, and attempting to develop a process that meets our goals, while retains a core D&D "feel" and simplicity.
Roll 3d6, reroll 1s once, and record them in order (we still use SIWDCCha - I guess that's a nod to tradition, or more likely just the way I automatically say them after all these years). It's simple, and doable.
I don't think "tradition" is a good reason to do just about anything (and in a prior life, my job was specifically to challenge the processes we used in our business, and if "tradition" was the only reason we were doing it, it probably wasn't a very good process). On the other hand, whenever I'm making changes to the rules, I think it's quite important for them to still "feel" like D&D. Even though we have a lot of house rules, and can explain the reasons for them, somebody who has played D&D should be able to sit down at our table and agree that we're actually playing D&D and not another game.
I've come up with several approaches to the goals we have. One that I'm still considering is that you generate each stat as 6 + 2d6. That guarantees you'll never have less than 8 for a stat, and that would trend at a 12 for every stat, I think (72). Feels like it would trend higher.
Quick test: S18, I17, W16, D15, C12, Cha17. Yes. I just rolled that. A definite outlier, although not being a math guy, I'll have to go do some work to see if my 12 trend is right...it still seems like it should work!
Anyway, use what works for you.
*I'm pretty much the author of the house rules. But they are sparked by, tested, and approved by the table. Some of them are my idea, some are one or more of the player's ideas. No rule (official or house) survives at our table without the support of the players.
)
