D&D 5E Point Buy vs Rolling for Stats

[JesterOC]

Therefore, a statement like 'this method lets me play any concept I want' can only be objectively true if that method allows any combination of 3-18 for six scores to be generated by that method.

Point-buy does not allow that. Rolling does.

Interesting point. D&D was based on a concept that each stat was based on a independent bell curve for each stat. As Gygax was a insurance underwriter he loved his numbers and he developed a system that he must have felt approximated nature.

But the problem with his method is that it does not actually simulate nature when you take it down to the individual level. Look at Olympic level athletes. Here is where you find the 18s of the human race. But look what you don't see. You don't see a human that is the best at everything. You don't even find many humans that are best in more than one thing. A weight lifter is not as fast as a fencer. The fencer can't beat the marathon runners, the Marathon runners can't even beat sprinters.

Because in real life individual bodies at better at one thing than others. You build fast twitch muscles, then you lack endurance. You have twice the muscle mass than the average, you loose flexibility.

So independent die rolls don't reflect reality. But a point buy system can. Mainly because you can approximate getting pretty good at something, or being well rounded but you can't make a superman.

 

[Tony Vargas]

I'm not getting hung up on numbers. I'm just saying there's a big difference between being extremely clumsy(4-5 dex) and a bit clumsy(8 dex). Any concept involving being extremely clumsy isn't going to happen with point buy or array.

Sounds hung up on the numbers, to me.

This translates into, "Yes, but no.".

Yes, but in context. Can you play /anything/ you want? No. You cannot play Thor. You cannot play the "best" character - a generation system like array or point-buy delivers some balance, that way, letting everyone play the character they want, without also letting everyone stomp all over eachothers' concepts, too.

If I'm invited to a game and I bring character concepts with me, only to find out when I get there that it's point buy, I cannot play what I want.

Only in the tautological sense that what you want is random generation. If you walk into a point-buy game with a concept, you can build to it, you'll get closer the less your concept requires undercutting the concepts of others, but you can do it. If you walk into a game using a standard array or random generation, you can still arrange stats to best fit your concept, but you're limited to shifting them around, if your concept calls for something other than exactly that standard (or exactly what you rolled) too bad.

 

[Hussar]

So I'm going to assume your not dumb because at first glace your answer appears to be leaving out so much information that it's hard to believe it's an attempt to answer my question.

So I assume your Ranger is using a primary stat of Dex and that you dumped Con. So yea he definitely is an example of a character that doesn't put his 2nd highest stat into ac or con. However, the question was "how often do you do such things". Cool example though.

For your fighter example: A fighter that Dump statted Dex means he was a Str fighter. I can't tell if this character also dump stated con or whether it was his 2nd highest stat. I don't want to assume here so I suppose it's best to ask? I guess we will have to come back to him.

Anyways these first 2 examples even if they both pan out as true examples of what I am asking about, that still doesn't tell me in general how often you build characters that way.

You do make 1 general observation though. That a lot of characters you have seen have low dex. I have 2 questions about them. Do they obtain AC from some other source like heavy armor? What kind of con would you generalize these low dex characters to have?

To be fair, we are pretty fighter type heavy, so, it's not a surprise to see Con being the second highest stat. I mean, between 3 campaigns over the past few years, we've had 20 ish characters, and of those, easily half and probably more than that have been fighter types of one flavor or another. So, yeah, Con is a pretty heavily bumped stat. No shock there.

And, again, to be honest, up until very recently, I almost always DM'd, so, can't really say I have a "general" way of building characters. I've only played 3 PC's in the past four years (ish), and prior to that, I hadn't played a PC in over a decade.

I was just saying that it's not easy to generalize. But, in my current Thule campaign, we have 6 PC's and no core casters (for reasons). 2 rangers, a paladin, a fighter, a monk and a rogue. Con is a the 2nd highest stat for one PC (a ranger) and tied for 2nd for one PC ( the other ranger). For 3 PC's (the fighter, the rogue and the monk) it would be 3rd highest and the paladin has a 10 Con. So, pretty wide range there, even if it is a very, very small sample size. Note, all these were built on point buy.

To be fair though, pretty much everyone went Dex builds (heh). Strength is the dump stat for 4 PC's and 2nd lowest for the other 2. This has to be the physically weakest party I've ever seen.

 

[Arial Black]

Interesting point. D&D was based on a concept that each stat was based on a independent bell curve for each stat. As Gygax was a insurance underwriter he loved his numbers and he developed a system that he must have felt approximated nature.

But the problem with his method is that it does not actually simulate nature when you take it down to the individual level. Look at Olympic level athletes. Here is where you find the 18s of the human race. But look what you don't see. You don't see a human that is the best at everything. You don't even find many humans that are best in more than one thing. A weight lifter is not as fast as a fencer. The fencer can't beat the marathon runners, the Marathon runners can't even beat sprinters.

Because in real life individual bodies at better at one thing than others. You build fast twitch muscles, then you lack endurance. You have twice the muscle mass than the average, you loose flexibility.

So independent die rolls don't reflect reality. But a point buy system can. Mainly because you can approximate getting pretty good at something, or being well rounded but you can't make a superman.

The question is not how accurately the 3d6 bell curve models a real population; the question is which method models it better: rolling or point-buy.

You use the Olympics to show how inaccurate the 3d6 bell curve is, but fail to mention that point-buy fails even harder!

First, people can be generally 'sporty', being above average in Str, Dex AND Con. Olympians, for example.

Second, you absolutely cannot infer that these sporty Olympians MUST have lower than average mental scores! There is no 'zero-sum game' in real life. In point-buy, there is. Point-buy is a less accurate model there.

And third, the fact that the Olympics exists shows that some people really are 'Faster, Higher, Stronger' than others! Point-buy would not have Olympiads because everyone is created equal! That is so not true in real life populations!

Point-buy is a much, much poorer approximation of a realistic population than that obtained by the 3d6 bell curve.

 

[Hussar]

Again, unless you have an Einstein out of every 200 people, and an Olympic level weght lifter out of every 200 people, and a Ghandi every 200 people, no, the 3d6 bell curve most certainly does not even come anywhere remotely close to approximating a realistic population. Not even remotely. You're seriously going to claim that in every village of 200 or more people, you'd have one of each of these people?

I'm sorry but, can we please shelve these "realism" arguments? D&D is unbelievably bad at trying to approximate anything even in the same zip code as anything even remotely associated with realism, regardless of which system you use.

 

[Arial Black]

Again, unless you have an Einstein out of every 200 people, and an Olympic level weght lifter out of every 200 people, and a Ghandi every 200 people, no, the 3d6 bell curve most certainly does not even come anywhere remotely close to approximating a realistic population. Not even remotely. You're seriously going to claim that in every village of 200 or more people, you'd have one of each of these people?

First, you can imagine that Einstein et al have stats higher than 18, or maybe higher than 20. Einstein would be a high level Physicist, and his adventures would have given him ASIs and Tomes and so forth.

All the 18 says is that you are (statistically) the strongest in a room of 216 people, not that you are the strongest man on the planet!

This was even addressed by Gygax himself in AD&D 1E, when he mentioned that the hypothetical strongest man in the world wouldn't just have 18/00 Str, he'd have 18/00000 Str (I can't remember the exact number of zeroes; it was nearly forty years ago!). Even so, a guy with 19 Str would be stronger.

Second, it's not trying to model real-life populations, but realistic populations for fantasy worlds. Here, the bell curve is looser than the real world.

I'm sorry but, can we please shelve these "realism" arguments? D&D is unbelievably bad at trying to approximate anything even in the same zip code as anything even remotely associated with realism, regardless of which system you use.

Some are much, much better at approximating realism than others. It's a valid thing to desire in a character generation system. Balance between players may also be desirable. How much priority we put on each quality varies from person to person and table to table, which is why some of us like rolling best and some prefer point-buy.

I've found that there are several ways of adjusting rolling to make the results more balanced between players. I can't recall point-buy being adjusted to make it more realistic.

 

[Maxperson]

Sounds hung up on the numbers, to me.

Nope! Wrong again. Fluff and crunch just have to match or there is a disconnect. I don't hang with that new fangled "My dex is 20, but I'm clumsy." crap.

Yes, but in context. Can you play /anything/ you want? No. You cannot play Thor. You cannot play the "best" character - a generation system like array or point-buy delivers some balance, that way, letting everyone play the character they want, without also letting everyone stomp all over eachothers' concepts, too.

Nobody is asking to play Thor. We're discussing characters that are acceptable within the default game rules. If you house rule out rolling, you eliminate a lot of concepts.

Only in the tautological sense that what you want is random generation. If you walk into a point-buy game with a concept, you can build to it, you'll get closer the less your concept requires undercutting the concepts of others, but you can do it. If you walk into a game using a standard array or random generation, you can still arrange stats to best fit your concept, but you're limited to shifting them around, if your concept calls for something other than exactly that standard (or exactly what you rolled) too bad.

There are still more concepts available to me via rolling than there are via point buy or arrays. I don't have to have all of them available on every roll for that to be true.

 

[Maxperson]

Again, unless you have an Einstein out of every 200 people, and an Olympic level weght lifter out of every 200 people, and a Ghandi every 200 people, no, the 3d6 bell curve most certainly does not even come anywhere remotely close to approximating a realistic population. Not even remotely. You're seriously going to claim that in every village of 200 or more people, you'd have one of each of these people?

How do we know that it doesn't? Being an Olympic level athlete is more than just having the highest ability. You also have to have the motivation to train hard enough to hone your ability, the means to do so, knowledge that your ability is good enough, and not have a bad day at the Olympic trials. That eliminates boat loads of Olympic level people.

I'm sorry but, can we please shelve these "realism" arguments? D&D is unbelievably bad at trying to approximate anything even in the same zip code as anything even remotely associated with realism, regardless of which system you use.

Why would we shelve realism? We get that realism does not equal "mirrors reality exactly" like you try to intimate whenever it is brought up. D&D is full of realism all over the place, the 3d6 bell curve just being one example of it.

 

[HomegrownHydra]

If I'm invited to a game and I bring character concepts with me, only to find out when I get there that it's point buy, I cannot play what I want.

You've previously said that you come up with far more character concepts than you'll ever get to play. Not even one of those concepts can work with point-buy?

 

[JesterOC]

Point-buy is a much, much poorer approximation of a realistic population than that obtained by the 3d6 bell curve.

True. The point buy is a representation of Olympians not the general population. And Point buy does not relegate you to having a low intelligence score. In a better point by system you could break up the points between Intelligence and Physical Attributes.

But one could argue that natural stats have very little to do with Olympians. Your DNA provides a solid foundation. But training (class levels) is what gets people to the top of the stats. No one gets to be an olympic weightlifter without YEARS and YEARS of experience lifting weights. You can be a natural 15 perhaps. But training gets you the rest of the way. In that way the current point buy system is better at representing the best of the best.