Question to those who don't like Point Buy...

193 posts / 0 new
Last post
...because I keep seeing examples of the bad things that result from Point Buy, namely hugely skewed and min/maxed stat lines.  Something like 18/16/15/7/7/7, or along those lines.

I won't dispute that that is annoying, horribly min/maxed, and very likely outside the "spirit" of D&D.

But, and I don't have all my old books to confirm, I don't remember a Point Buy system that even allowed anything like that.  4e, for example, gave you an initial stat array of 10/10/10/10/10/8, with no means of lowering that 8 at all.  You could the 10's to 8, but the points gained wouldn't be enough to raise a high stat even higher, so at worst, people kept the 10's as 10's.

And my recollection of 3e was that the system was somewhat similar, if not identical.  Can anyone confirm?

Because, based on my memory (which could be wrong, I'll admit), putting forth the horribly min/maxed example above is a type of strawman.
Essentials zigged, when I wanted to continue zagging. Roll dice, not cars.
The method in the 3.5 DMG lets every stat begin at 8, but they could only be lowered through racial ability penalties.  A half-orc monk with Int 6 and Charisma 6 was both entirely reasonable, and mechanically encouraged by the system.

Second Edition Skills & Powers had a point buy method where you had roughly 75 points to spend amongst your six stats, and little reason to not go 18/18/18/7/7/7.  Skills & Powers was not exactly known for its balance, though.

There are also other point buy methods, rather than just the ones published in a few specific books.  Point buy systems have a wide variety of implementations, many of which do allow you to "sell back" stats, or which start at the equivalent of 3 across the board.  I believe that the Pathfinder system lets you start at 10 and sell back as low as 7.

The argument is only a strawman if it is specifically supposed to be an argument against one particular point buy method where it is not the case.  If you're just talking about point buy systems in general, then that is indeed one of their common (though not universal) faults.
The metagame is not the game.
Thank you for the clarification.  I was getting increasingly concerned that I was missing some vital piece of information.

So, that beng established, is the 4e version of Point Buy (no stat lower than 8, and mathematically untenable to lower another below 10) less ornerous, or more palatable to those who don't like point buy in general?

I guess what I'm trying to get at is whether people who object to point buy (in general) do so because of the propensity for min maxing, or some other reason.  If the min maxing could be systematically be mitigated, would the objection be dropped, or at least lowered?
Essentials zigged, when I wanted to continue zagging. Roll dice, not cars.
Many of us like the spontaneity, unpredictability, and originality of dice-rolling for ability scores.

A better system might be for a combination of dice-rolling and point buy.

Roll 3d6 for each ability score.
Example:























12
12
14
5
8
8

Total: 59

Player is then given the difference in total value from a 75 point buy.  In this case, an extra 16 point buy is given.  No score can be improved beyond 15 using this point buy.

It does not unbalance the game and adds a level of unpredictability.  Perhaps someone might create this character.























15
15
15
10
10
10


Another option would be to allow the character to improve each score 1 point at a time until all points are spent and not able to improve a score until all other scores have been increased as well.  So we might end up with this.























15
15
17
8
10
10
Thank you for the clarification.  I was getting increasingly concerned that I was missing some vital piece of information.

So, that beng established, is the 4e version of Point Buy (no stat lower than 8, and mathematically untenable to lower another below 10) less ornerous, or more palatable to those who don't like point buy in general?

I guess what I'm trying to get at is whether people who object to point buy (in general) do so because of the propensity for min maxing, or some other reason.  If the min maxing could be systematically be mitigated, would the objection be dropped, or at least lowered?


Maybe. Because theres two reasons people use rolling instead.
1. They hate min/maxing
2. They just like rolling

I think a fix for min/maxing might win back crowd no.1, but the people who just plain prefer rolling will probably keep on rollin'. 
My two copper.
Rolling is just a lot more fun from my group's perspective, but if other groups want to use point buy that's fine. We personally like the chaotic / organic development of our characters; we even recently started to limit keeping the stats in the order you rolled them, and only switching one score with another (we usually just roll them and put them wherever you want). I can see how that would infuriate other players, but we really do enjoy when a wizard has 15 strength, and suddenly you get the idea that maybe that wizard does a lot of labor on his farmstead or is just naturally ripped. Some of our most memorable characters have had sub-par stats. In a way it almost makes them more heroic. But I really do think that there should be a point buy system for those that have a strong character concept in mind, don't like rolling in general, or even for those groups that like to optimize their characters; but it just wouldn't feel like d&d to me if I didn't roll my stats. Tongue Out
...because I keep seeing examples of the bad things that result from Point Buy, namely hugely skewed and min/maxed stat lines.  Something like 18/16/15/7/7/7, or along those lines.

I won't dispute that that is annoying, horribly min/maxed, and very likely outside the "spirit" of D&D.

But, and I don't have all my old books to confirm, I don't remember a Point Buy system that even allowed anything like that.  4e, for example, gave you an initial stat array of 10/10/10/10/10/8, with no means of lowering that 8 at all.  You could the 10's to 8, but the points gained wouldn't be enough to raise a high stat even higher, so at worst, people kept the 10's as 10's.

And my recollection of 3e was that the system was somewhat similar, if not identical.  Can anyone confirm?

Because, based on my memory (which could be wrong, I'll admit), putting forth the horribly min/maxed example above is a type of strawman.



3e started you at straight 8's.  But yes, thankfully most Main-line D&D has never allowed the kind of sell offs that let point buy minmazing get truly insane (See: Tri-Stat, which functions on a vastly different DM/player social contract because of its loose and easily breakable system)

The issue (for me at least) is that the Spirit issue maintains even if you get fewer points and start with 10's.  It's an issue of Creative Sterility.  If you use nonrandom generation methods, there WILL be a mathematically 'perfect' execution that gets used ad nauseum.  In 3e it's dependant on a lot of factors, but in general you'll see a high attack stat and constitution, a completley ignored Charisma (unless it's the attack stat, in which case Wisdom is ignored and the Force of Personality feat used to compensate for the will save), Completley ignored strength unless it's the attack stat (Encumbrance is stupidly easy to avoid if it's even enforced, and classes that don't use str to attack either don't melee or use touch, with which rolling to hit us usually more of a formality), and remaining points spent on a +Dex dependant on armor type worn, enough int to get your "essential" skills, and Wisdom for a better will save.

The general formula is really complicated, but in practice the execution is not: you just have to pick a class first.  Assuming you're playing a (human) Wizard with the standard 25 points, you buy an 8 STR and 8 CHA (0pts) every time.  You buy a 14 Con (6pts) and 16 Int (10pts) to cover your HP and Casting, and a 14 Dex (6pts) for AC in the early game and better hitting with rays later on and drop the remaining 3 points into Wisdom for a 11 Wis.   Playing a human fighter? You buy an 8 CHA and 8 INT because they do nothing for you, 16 Str, 14 Con, 14 Wis (you need that will save) and 11 Dex: this is the build for weilding a two-handed weapon: it gives you plenty of HP, damage, and the ability to power attack.  Your des suffers, since you can't use a bonus in full plate anyway.

Long story short: In a deterministic system, you can with a particular input predict, exactly, what the output will be.  I, personally, find that to be not in spirit with D&D, and furthermore that having a mathematically solved system erects a psychological barrier to doing anything OTHER than usint the solve, dampening (not eliminating, just dampening) creativity.  Not everyone has the same reaction ,and if you can innovate under point buy more power to you.

Some Comments on Rolling Stats and Common Misconceptions Thereof: An Aside

I also find that these debates get marred a LOT by bad definitions.  Just as many random-roll players forget that selling down stats has never really been a thing in D&D, I feel that people tend to labor under some... odd ideas about rolled stats.

1) "My rolls suck! Now I'm stuck with a lousy character".  This is a bit of a holdover from old editions: In 3e (the WotC edition that promoted random rolling as its primary generation method), the PHB itself says that any stat block that does not contain at least one score of 14+ or where the sum of the ability bonuses does not total to at least +1 can be dropped entirely.  Scratch out those six numbers and reroll them.  WotC understood that there needs to be a competency floor, and installed one from Day 1.  As a random roller, I see no conflict with this: the purpose of randomization is not power-level-lottery, it's forcing me out of my box a little.

2) "His rolls were better, therefore his character will always be better".  This would be true if all other factors were equal.  Which, in Dungeons and Dragons, they basically never are.  When was the last time you saw a campaign where two characters were exactly the same in every way (at the same time, not just Dead Guy Mk II)?  For me, the answer is "Never".  It's rare as hen's teeth to see two members of the same class in the same game.  Add WotC-Era build options into the fray?  It shouldn't ever happen.  So what really occurs when somebody hit the stat jackpot and somebody else just squeaked in above the competency floor?  You find out that above a certain threshold, stats (especially secondary stats) actually don't tend to do that much for you.  A fighter with (18, 17, 15, 14, 12, 11) and a rogue who rolled (15, 13, 11, 11, 10, 8) are both baseline playable.  The rogue will have to make some harder choices with skills at level 1 than he might like to, but in the end they engage with the campaign in different ways, and since the rogue still has the minimum stats to do his thing, he gets to do his thing.  The fighter on the other hand, will have more toys than he would otherwise have, but at the same time, his toys don't take away from the rogue's toys because they're looking for different things out of their stats and build.

3) "More power = More Fun.  Therefore, To have equal fun we need equal power."  At best, this is a personal opinion.  I contend, though, that Fun is not a direct function of power, even relative power in an extent.  You do want to be in roughly the same band, but an artificial floor combined with the natural bell curving and cap handle that already where stats are concerned.
aside to an aside: The First Time I played D&D...
I played a wizard who won the stat lottery bigger than any character I've seen since.  An 18 floating about, a 17 in that Int, and not a stat below 15: That's rolling 3d6 and placing them in order.  I clearly did enjoy that game, seeing as I stuck with the hobby, but I have to say it wasn't my favorite PC to run, all things considered.  The stats really had little to do with it.  The fighter of the party?  She had pretty average stats: a positive sum of bonuses, I think, but nothing to write home about.  Loved every minute of it, and wasn't nearly as satisfied with technically "Better" characters in later games.



Hopefully we can agree that...
a) Point buy is a valid system
b) Rolling for stats is a valid system
and
c) D&DN should let groups choose which they want to use. 

As long as there's a choice present, I don't particularly care which gets top billing.

"Enjoy your screams, Sarpadia - they will soon be muffled beneath snow and ice."
On Worldbuilding - On Crafting Aliens - Pillars of Art and Flavor - Simulationism, Narritivism, and Gamism - Shub-Niggurath in D&D
THE COALITION WAR GAME -Phyrexian Chief Praetor
Round 1: (4-1-2, 1 kill)
Round 2: (16-8-2, 4 kills)
Round 3: (18-9-2, 1 kill)
Round 4: (22-10-0, 2 kills)
Round 5: (56-16-3, 9 kills)
Round 6: (8-7-1)

Last Edited by Ralph on blank, 1920

Well traditionally their are four reasons I know of to dislike pointbuy.

1. You're one of those obnoxious people who never rolls below 16.
2. You dislike the min-maxxing
3. You think there's something special about being forced to think about your stats instead of just placing them to max out your build
4. You don't like the extra math

Nobody outside group 1 likes them so forget'em.

Three is variable, sometimes it's fun and engaging, especially when you don't have a really solid character concept before you start. When you do wanna play a specific character it gets really annoying if the dice don't agree. This could especially be a trap in 3e where the class balance was so borked up since you could roll low mental stats forcing you to choose between playing a weak class that matched your stats or a strong class that didn't. 

The fourth reason is a bit meh, but I'm a bit of a math snob so I'm biased.

 Two is what you're addressing. I don't know how you plan to mitigate the min-maxing though.

Anyway I think a superior solution would be a synthesis of point-buy and rolling. You roll each stat and the result costs a number of points, or gives a number if sufficiently low. So using the 3.5 system a roll of 15 would cost 8 points out of 25, and a roll of 6 would give 2 points back. Once you've rolled the final stat you then add or remove stats as if using the point-by system until you balance to 0.

So assuming the example character rolled 15 str, 6 dex, and then 8's accross the board, he would have 25-6+2 (21) point buy points remaining with which to alter his stats as per the normal point buy rules, with the limitation that he can only add points if his pool is positive, or subtract them if his pool is negative.
The specific example you gave is factually incorrect, yes.  No edition of D&D has allowed it. (except maybe S&P, I don't remember that being legal by the S&P rules, but I could be wrong.)


However, it's more of a mistake than a strawman.  The spirit of the example and of the argument is absolutely true.  It is mathematically a bad idea to have a 4E character who has more than two stats with significant point investment (and in some cases, more than one stat).  You basically don't see more than two stat arrays, with a third one popping up in some weird cases.

That fact is what I object to about point buy; with the way D&D stats are relevant to classes, it's just a bad idea to have anything other than the mathematically superior array.
The difference between madness and genius is determined only by degrees of success.
Actually witht he way 4e defenses are set Most builds I see at least start with 3 solid stats. At higher levels yes the applicable  2 tend to dominate.

Thank you for the clarification.  I was getting increasingly concerned that I was missing some vital piece of information.

So, that beng established, is the 4e version of Point Buy (no stat lower than 8, and mathematically untenable to lower another below 10) less ornerous, or more palatable to those who don't like point buy in general?

I guess what I'm trying to get at is whether people who object to point buy (in general) do so because of the propensity for min maxing, or some other reason.  If the min maxing could be systematically be mitigated, would the objection be dropped, or at least lowered?



I don't object to it's presence,  but I won't use it given a random die choice,  so I'll try to answer your question.

I have several specific issues with point-buy.

-Point buy as you note,  encourages min/max play,  often resulting in situations where every adventurer is effectively a super-human in some area,  and no one is just above average.

-Point buy also tends to generate "Clone wars" type play,  where nearly every character of a certain class  is identical to the next guy,  in all but name.  Especially once people start "Net decking" their characters and picking feats based on a guide. 

-Point buy also removes the attachment to the Character.  If he dies,  a Clone can just rise up and take his place.  Part of the excitement of the game is fighting to save a beloved character knowing you're unlikely to get someone identical to him.  Point buy removes that.

-Point buy removes the ability to have an exceptionally good character,  and removes the challenge of having an average character.

-Point buy discourages Roleplaying.  Since Characters tend to be clones,  Character personalities tend to be clones.  You would have to actively try to have a character with some disability that would encourage you to roleplay something off the wall,  a great example is Raistlin from Dragonlance.  His personality was born from someone Roleplaying his constitution.  In a point buy system,  only the most dedicated of Roleplayers is going to create someone with a drawback like Raistlin had.

-Point buy also tends to be the sanctuary of some of the most notorious of gamers,  Min/Maxers,  Munchkins,  and Power Gamers.  That said,  point-buy will also be the haven of dedicated Roleplayers,  since they may be able to create the specific character they have in mind (Which is why I heartily agree that point-buy should be an option)

-Point buy encourages System-mastery over gaming,  the above groups use their system mastery to make the most optimal stat choices in order to eke out every last drop of performance.  Without taking the risk of receiving something that wouldn't be considered optimal.

Point buy really facilitates alot of negative forms of gaming,  but it also really facilitates some very positive forms of gaming in the right hands.

I choose not to use it because I prefer the chance of an exceptional or average result,  I try to roleplay the character generated,  and I prefer the attachment to the character and the excitement in trying to keep him alive.
I guess what I'm trying to get at is whether people who object to point buy (in general) do so because of the propensity for min maxing, or some other reason.  If the min maxing could be systematically be mitigated, would the objection be dropped, or at least lowered?

I dislike point-buy for much the same reason I dislike 4d6 drop the lowest and assign to taste- because it raises the bar for expected competence.  I prefer more down-to-earth characters, with significant but not overpowering strengths and real below-average flaws.  I want someone with strength 14 to feel strong, which is hard to do when every other character has strength 18; I find someone with strength 18 and nothing below 8 to be really hard to relate to.

I would probably have more fun with a very low point buy (if it allowed stats in the 5-7 range) than a 4d6 drop the lowest roll, but if the system is designed around the expectations of everyone having good scores in their main stats, then it's hard for someone with a twelve in a few stats to feel as competent as I want them to be.

I also enjoy the fun of starting with the random distributions and developing the character from that, rather than starting with a character and trying to express that in game terms, but that's just personal preference.  

 

The metagame is not the game.

Thank you for the clarification.  I was getting increasingly concerned that I was missing some vital piece of information.

So, that beng established, is the 4e version of Point Buy (no stat lower than 8, and mathematically untenable to lower another below 10) less ornerous, or more palatable to those who don't like point buy in general?

I guess what I'm trying to get at is whether people who object to point buy (in general) do so because of the propensity for min maxing, or some other reason.  If the min maxing could be systematically be mitigated, would the objection be dropped, or at least lowered?



I don't object to it's presence,  but I won't use it given a random die choice,  so I'll try to answer your question.

I have several specific issues with point-buy.

-Point buy as you note,  encourages min/max play,  often resulting in situations where every adventurer is effectively a super-human in some area,  and no one is just above average.

-Point buy also tends to generate "Clone wars" type play,  where nearly every character of a certain class  is identical to the next guy,  in all but name.  Especially once people start "Net decking" their characters and picking feats based on a guide. 

-Point buy also removes the attachment to the Character.  If he dies,  a Clone can just rise up and take his place.  Part of the excitement of the game is fighting to save a beloved character knowing you're unlikely to get someone identical to him.  Point buy removes that.

-Point buy removes the ability to have an exceptionally good character,  and removes the challenge of having an average character.

-Point buy discourages Roleplaying.  Since Characters tend to be clones,  Character personalities tend to be clones.  You would have to actively try to have a character with some disability that would encourage you to roleplay something off the wall,  a great example is Raistlin from Dragonlance.  His personality was born from someone Roleplaying his constitution.  In a point buy system,  only the most dedicated of Roleplayers is going to create someone with a drawback like Raistlin had.

-Point buy also tends to be the sanctuary of some of the most notorious of gamers,  Min/Maxers,  Munchkins,  and Power Gamers.  That said,  point-buy will also be the haven of dedicated Roleplayers,  since they may be able to create the specific character they have in mind (Which is why I heartily agree that point-buy should be an option)

-Point buy encourages System-mastery over gaming,  the above groups use their system mastery to make the most optimal stat choices in order to eke out every last drop of performance.  Without taking the risk of receiving something that wouldn't be considered optimal.

Point buy really facilitates alot of negative forms of gaming,  but it also really facilitates some very positive forms of gaming in the right hands.

I choose not to use it because I prefer the chance of an exceptional or average result,  I try to roleplay the character generated,  and I prefer the attachment to the character and the excitement in trying to keep him alive.



+1
I don't object to it's presence,  but I won't use it given a random die choice,  so I'll try to answer your question.


I have several specific issues with point-buy.


And I'm not going to call you wrong for having these issues, but rather, I will offer some counterpoints, or ask follow-up questions, in the spirit of open discourse.

-Point buy as you note,  encourages min/max play,  often resulting in situations where every adventurer is effectively a super-human in some area,  and no one is just above average.

Isn't this actually due to point buy totals, and not point buy as a system?  I mean, I could effectively get the same thing (consistently super-powered PCs) if I used a roll system like "Roll 6, keep the 3 best, for each of the stats".  In turn, I could create a point buy system that generates only barely above average PCs if I simply lowered the amount of points to work with.

-Point buy also tends to generate "Clone wars" type play,  where nearly every character of a certain class  is identical to the next guy,  in all but name.  Especially once people start "Net decking" their characters and picking feats based on a guide.

Certainly, this is more likely than in rolling (because of the element of player control or agency), but I ask if it is really an actual problem.  First, does it really matter if a player is using identical stats if he's still roleplaying the character in a fun and interesting way?  And second, is it really a given that just because the point buy system is in effect, players will make "clones"?  My personal anecdotal evidence tells me otherwise.  Players make the characters they want to make.

-Point buy also removes the attachment to the Character.  If he dies,  a Clone can just rise up and take his place.  Part of the excitement of the game is fighting to save a beloved character knowing you're unlikely to get someone identical to him.  Point buy removes that.

Does this really actually happen enough to merit complaint.  Isn't peer pressure enough to prevent a person from making "Bob the Fighter II"?  And is it really so wrong that a player may want a certain experience from the game, one delivered by a particular character?

Also, I differ pretty strongly on the statement that point buy removes attachment to the character.  I'd argue that because each point invested represents a deliberate choice to create a specific character, point buys actually enables and reinforces attachment.  A point bought character is a creation, and not happenstance.


-Point buy removes the ability to have an exceptionally good character,  and removes the challenge of having an average character.

No argument there.  Point buy certainly normalizes the experience of the game, at least when it comes to stats.  I take that as a positive, but I cannot object to your statement on a factual, objective basis (assuming, of course, we are using a "standard" point buy pool; a larger or smaller pool of points would naturally result in characters with higher or lower base competency, respectively).

-Point buy discourages Roleplaying.  Since Characters tend to be clones,  Character personalities tend to be clones.  You would have to actively try to have a character with some disability that would encourage you to roleplay something off the wall,  a great example is Raistlin from Dragonlance.  His personality was born from someone Roleplaying his constitution.  In a point buy system,  only the most dedicated of Roleplayers is going to create someone with a drawback like Raistlin had.

I see no practical evidence that point buy discourages roleplaying.  First, I see no direct correlation between stats and personalities.  Identically statted characters need not have the same personality.  Second, I see no evidence that rolling for stats precludes the chance of players making characters with identical personalities.  If a player rolled, by chance, the same stat line, would he be obligated to have the new character have the same personality as the old one?  Why would buying stats be any different?

Furthermore, does point buy actually create characters without weaknesses?  Raistlin's personality emerged in part by roleplaying to his weak CON stat.  Why would this be any different if the player purposely chose to have the weak stat?  In fact, what if the character, gifts and flaws, were conceived before determining stats?  Wouldn't rolling risk upsetting that attempt at character conception?



-Point buy also tends to be the sanctuary of some of the most notorious of gamers,  Min/Maxers,  Munchkins,  and Power Gamers.  That said,  point-buy will also be the haven of dedicated Roleplayers,  since they may be able to create the specific character they have in mind (Which is why I heartily agree that point-buy should be an option)

I saw plenty of min/maxing, munchkining, and power gaming in 2e with rolled stats, so I'm inclined to believe that problem players will be problem player, regardless of stat determinism.  People who are obsessed with having a maxxed primary stat will re-roll and re-roll until they get a desired result, for example.

-Point buy encourages System-mastery over gaming,  the above groups use their system mastery to make the most optimal stat choices in order to eke out every last drop of performance.  Without taking the risk of receiving something that wouldn't be considered optimal.

That begs the question: is that actually a bad thing?  Optimization doesn't mean "not roleplaying".  It just means optimization.  It just scales up capability.  The only time I see optimization being a real problem is when it is not group-wide.  And really, how is one player optimizing any different (in end result) to one player rolling ridiculously well?

And as a follow-up, does point buy actually encourage system mastery any moreso than rolling?  From where I stand, it takes great system mastery to make the most of what you may roll.  The system mastery may manifest in a different way, but it is still present.


Point buy really facilitates alot of negative forms of gaming,  but it also really facilitates some very positive forms of gaming in the right hands.

True.  I could easily say something similar about rolling for stats.  It facilitates imbalanced and uneven gameplay, making a baseline hard(er) to guage, but can also facilitate outside-the-box experiences in the right hands.  Or something to that effect.

I choose not to use it because I prefer the chance of an exceptional or average result,  I try to roleplay the character generated,  and I prefer the attachment to the character and the excitement in trying to keep him alive.


And that's really the crux of it, ain't it?  You generate stats and design a character to match.  I conceive a character, and generate stats to match.  You prefer the excitement of keeping your character alive because you may never see him again, whereas I prefer the excitement of keeping my character alive because I have designed him (and his story) and want to see where he goes (with that story).
Essentials zigged, when I wanted to continue zagging. Roll dice, not cars.
I think they're on the way of solving the problem of the "dump stats" in D&D Next with all stats being a potential saving throw (is it even a problem?).

All they need to do is make sure that a reasonable number of monster attacks target something else than AC and all stats become important.

Another idea would be to limit the ability score range with point buy. Say 8 to 16. That way, you can't build a PC with an ability score above 18 after adjustments. You can either secure a solid 16 in your primary ability score or take your chances.
IMO best solution is controled rolling.

most likely number to get while rolling 4d6 drop lowest is 13.

So 13×6 is 78.

Then total of abilities is set at 78.

then you roll stats with 4d6 and then you add or subtract from rolled scores until you get to 78.

optionally, you can say that first you must bump up all scores atleast to 8 or even 10.
...because I keep seeing examples of the bad things that result from Point Buy, namely hugely skewed and min/maxed stat lines.  Something like 18/16/15/7/7/7, or along those lines.

I won't dispute that that is annoying, horribly min/maxed, and very likely outside the "spirit" of D&D.

But, and I don't have all my old books to confirm, I don't remember a Point Buy system that even allowed anything like that.  4e, for example, gave you an initial stat array of 10/10/10/10/10/8, with no means of lowering that 8 at all.  You could the 10's to 8, but the points gained wouldn't be enough to raise a high stat even higher, so at worst, people kept the 10's as 10's.

And my recollection of 3e was that the system was somewhat similar, if not identical.  Can anyone confirm?

Because, based on my memory (which could be wrong, I'll admit), putting forth the horribly min/maxed example above is a type of strawman.



 It was probably me who posted that. In PF I think you can get a 7 and I have played around 4-5 d20 veriosns and they all have differnet methods of point buy and in 3.5 you can get down that low.

 I don't like point buy becuase it leads to min maxing. A fighter with 15 charisma can be rolled depending on if you assign how you like or whatever. In 4th ed it was more like 8,8,8,18,14,10 or whatever the numbers added up to. I have been using point buy for yearsand have started playing 2nd ed again and discovvered that rolling is more fun. 

 IN D&DN point buy does not work well with the human. It makes them really overpowered instead of just a little. It should still be in the core book as it takes up very little room.
...because I keep seeing examples of the bad things that result from Point Buy, namely hugely skewed and min/maxed stat lines.  Something like 18/16/15/7/7/7, or along those lines.

I won't dispute that that is annoying, horribly min/maxed, and very likely outside the "spirit" of D&D.

But, and I don't have all my old books to confirm, I don't remember a Point Buy system that even allowed anything like that.  4e, for example, gave you an initial stat array of 10/10/10/10/10/8, with no means of lowering that 8 at all.  You could the 10's to 8, but the points gained wouldn't be enough to raise a high stat even higher, so at worst, people kept the 10's as 10's.

And my recollection of 3e was that the system was somewhat similar, if not identical.  Can anyone confirm?

Because, based on my memory (which could be wrong, I'll admit), putting forth the horribly min/maxed example above is a type of strawman.



 It was probably me who posted that. In PF I think you can get a 7 and I have played around 4-5 d20 veriosns and they all have differnet methods of point buy and in 3.5 you can get down that low.

 I don't like point buy becuase it leads to min maxing. A fighter with 15 charisma can be rolled depending on if you assign how you like or whatever. In 4th ed it was more like 8,8,8,18,14,10 or whatever the numbers added up to. I have been using point buy for yearsand have started playing 2nd ed again and discovvered that rolling is more fun. 

 IN D&DN point buy does not work well with the human. It makes them really overpowered instead of just a little. It should still be in the core book as it takes up very little room.



Re: the bolded part: In creating this thread, I went back and looked at the rules for point buy in 4e, and there is no way to "buy back" points, RAW, in 4e.  You start with 5 10's and 1 8, and you can only go up.  And since there are no racial stat penalties, it is impossible to have a stat line like the one above, at least in 4e.

Moreover, even if it was possible, that would a be phenomenally poor strategy, from an optimization standpoint.  The NADs plus the strong effect secondary (and sometime teritary) abilities have on character effectiveness, combined with the diminishing returns on raising your prime stat during character creation, made starting with an 18 often wasted.  It wasn't always true (Wizards for example tended to not have any consistent secondary or tertiary stat), but it was true often enough that the best optimization typically fell around having a 16 in your prime, with racial bonuses bumping you up to 18, or boosting your secondary AND teritary scores.

But yeah, humans suck in D&DN.  Not "ineffective" suck, but "way too good" suck.

That reminds me: even if I was a stat-roller, I'd use point buy for playtest purposes for the exact reason some don't like it: it normalizes the stats, and thus the experience.  Having point buy reduces the variables in play, helping me, as a playtester, narrow down what works and what doesn't on account of the actual design, and not some freakishly high or low rolls.

Essentials zigged, when I wanted to continue zagging. Roll dice, not cars.
 I'm just saying most of my point buy has been outside 4th ed. Similar probelms in all versions of D&D and in 4th ed as long as youhave your 3 NADs covered its fine. Str, Dex Con fighter dump the rest. Sme of the Char Op boards did have builds for 4th ed with a 20 at level 1. 

 THe overall effect is the same whether you get 8/8/8/12/14/18 or get your scores down to 6 or 7 in PF/3.5 via racial mods (PF may be able to get a 5). Its not a 3.5/4th ed thing as we used point buy in both. 4th ed kind of incentivised 3 stats alot of classes in 3.5 you only needed to feed 2 stats or even one. I've used point buy in 3,rd,4th,SWSE and PF and I'm returning to "lets roll". If you get a screwy character oh well lets play.

 We've started playing 2nd ed again after a 12 year break and the players enjoyed rolling 4d6 frop the lowest (I allow some fudging). 
I don't see the point here; you can roll for your stats your you can point buy; there is no reason to exclude one or the other.

Also, I think rolled stats are an antiquity kept for the sake of nostalgia and tradition, but that is simply my opinion.

Supporting an edition you like does not make you an edition warrior. Demanding that everybody else support your edition makes you an edition warrior.

Why do I like 13th Age? Because I like D&D: http://magbonch.wordpress.com/2013/10/16/first-impressions-13th-age/

AzoriusGuildmage- "I think that you simply spent so long playing it, especially in your formative years with the hobby, that you've long since rationalized or houseruled away its oddities, and set it in your mind as the standard for what is and isn't reasonable in an rpg."

I'm almost 100% positive they will have both options like they have since 2000 or so (or was 3.5 point buy?).
Point buy discourages Roleplaying.  Since Characters tend to be clones,  Character personalities tend to be clones.  You would have to actively try to have a character with some disability that would encourage you to roleplay something off the wall,  a great example is Raistlin from Dragonlance.  His personality was born from someone Roleplaying his constitution.  In a point buy system,  only the most dedicated of Roleplayers is going to create someone with a drawback like Raistlin had.

I see no practical evidence that point buy discourages roleplaying.  First, I see no direct correlation between stats and personalities.  Identically statted characters need not have the same personality.  Second, I see no evidence that rolling for stats precludes the chance of players making characters with identical personalities.  If a player rolled, by chance, the same stat line, would he be obligated to have the new character have the same personality as the old one?  Why would buying stats be any different?

Furthermore, does point buy actually create characters without weaknesses?  Raistlin's personality emerged in part by roleplaying to his weak CON stat.  Why would this be any different if the player purposely chose to have the weak stat?  In fact, what if the character, gifts and flaws, were conceived before determining stats?  Wouldn't rolling risk upsetting that attempt at character conception?



The issue is that Point Buy allows a player the option to avoid having a bad score in Con. Considering how important Con is to survivability, it is unlikely that anyone, even a die-hard, would deliberatly make such a character. I did so once on a dare (2E) and enjoyed it until an AOE caused damage to the entire party. Even with my save I couldn't survive the attack, even though the attack was reasonable for our level. Since your character's lifespan is almost entirely at the DMs whim, no one would willingly choose this.


Current rolling methods prevent this as well, since you choose where your rolls go. If anyone rolls a really bad score, they almost universally put it in Charisma unless they are a Charisma based character.



Point buy also tends to be the sanctuary of some of the most notorious of gamers,  Min/Maxers,  Munchkins,  and Power Gamers.  That said,  point-buy will also be the haven of dedicated Roleplayers,  since they may be able to create the specific character they have in mind (Which is why I heartily agree that point-buy should be an option)

I saw plenty of min/maxing, munchkining, and power gaming in 2e with rolled stats, so I'm inclined to believe that problem players will be problem player, regardless of stat determinism.  People who are obsessed with having a maxxed primary stat will re-roll and re-roll until they get a desired result, for example.


Every group I have played with required you to roll your ability scores in public, thus preventing re-rolls.

I have always had a dislike of point buy, but I see why it is an attractive option (and I mean I can see why for players other than those that are interested in min-maxing or what have you). It's about control over your character and never having that moment where the dice dole out scores that could be totally interesting and entertaining to you... if you hadn't already decided you wanted to play a character nothing like those scores indicate, such as wanting to play a charismatic and studious wizard and rolling a character that would be a perfect, albeit dimwitted, warrior.

The reasons why I dislike point buy, however, will likely seem very strange to anyone having just read that:

1) Dice provide a suggestion as to the sort of character to play, including sometimes unique sorts of combinations like a particularly muscular wizard - point buy rewards playing as stereotyped of a character as is possible.

2) Dice provide a way to set the "balance benchmark" of the game at the most statistically common range of ability - this is the exact reason that D&D prior to 3rd edition had such a wide central range of scores that had next to no benefit or penalty - while point buy either forces all character above that "balance benchmark" in their highest scores or raises the benchmark to match an assume bought score that makes builds other than those min-maxed to match the benchmark less viable.

3) Nostalgia - I learned D&D when the standard methods all involved dice rolls and was introduced to my first build point system with Skills & Powers, which was an amazingly poorly thought out system to attach to a game that was perfectly playable if every character had average rolls in every ability score because it had absolutely no drawback built in for choosing to simply have the best stats possible in half of your ability scores.

4) Kind of an elaborate continuation of 1 that I acknowledge makes me a "jerk DM" - I find it to be extremely boring to DM for a group of players that play the same types of characters al the time, which point buy allows to go without any form of check other than me saying "Really, another Elven Wizard?" while dice provide some suggestion to the player as to what to play that might actually inspire them to an idea they haven't already thought up, and encourages the player to play a more varied repetoire of characters.

An important note: I am not saying that un-filtered rolling of scores is the only, nor best, way to handle things. In fact, the system for character creation that I find the most rewarding and balanced is one that contains the following steps:

1) Roll each score in order on 3d6
2) Check resulting scores to make sure they are not too low (of course, too low is defined much lower than most people I know would define it - at least one 13+ and not two scores 5 or lower)
3) Arrange scores however you like - OR - swap only two scores and recieve a bonus - OR - keep the scores where you rolled them and receive a larger bonus.
4) Adjust your ability scores with the points used to purchase other aspects of your character if you want to, with diminishing returns on your investment based on whether you are raising a low score (less than 9), a moderate score (10-15) or a high score (16+)

It allows the dice to make a suggestion that might lead a player to playing a character they wouldn't have otherwise thought of, but guarantees that you can instead play a character at least something like the one you had in your head if you so choose.
Careful, man. That much logic might be illegal on the internet. - Salla
I haven't really liked random rolling since the late 80's.  An illustration of why happened when a very good friend of mine who just so happened to have horrible luck with dice decided to roll up a ranger using the Unearthed Arcana rolling method (an insane method that involved rolling 3d6 for your dump stat and increasing numbers of d6 for more important stats in the hopes that you would meet the class minimums) but was still unable to meet the min's for his chosen class.  

If I want to make a character I want to have control over the design of that character not be at the whims of the dice gods.

If your table prefers to roll for stats (using whatever your prefered method is) more power to ya.

No reason these methods can't all be alternatives.

 

The one place where I've looked at point buy and said "I really have to do it that way" is in Earthdawn, for a few reasons.


The races in Earthdawn are designed as having either a lot of abilities or a lot of attributes. The other major balancing factor from race to race was what they called a karma die, which is basically a die you can add to any roll by spending a point or more of karma. So a race with more minuses than bonuses to their attributes would have a karma die of d10 while another race would have a lot of bonuses but a karma die of d4.


Other abilities figured in, like one race had the ability to fly and some bonus to their physical defense but had a messload of minuses, but really the karma die was the mechanical fix to the system.


Thing about Earthdawn's races is they're actually quite balanced in spite of the fact that their bonuses vary hugely, but that balance hinges on the point buy system delivering a normalised range of attributes. The moment you start rolling, the attribute bonuses become too strong and the penalties too severe. When you're buying stats from a finite pool of points the attribute modifiers balance really well with the karma die because the range is reliable. The other impact assuming a point buy does for Earthdawn is it let them be really heavy handed with the races and each race feels unique in a way that D&D doesn't even try.


So that's an example where purchase points are vital to game design. Nothing stops you from rolling up stats in Earthdawn, but doing so often creates a huge imbalance from race to race.


Other thing about this game is it's a heavy fluff, concept-driven sort of game which purchasing attributes is absolutely necessary, so I don't think the min/max argument is particularly effective in this discussion. Yeah, you can do it but you can do it with rolling as well.

I tend to prefer arrays.

With Point buy and Rolling, I tend to see stereotypes. After you put your best roll/purchase to your primary ability score and your second best to CON, you rarely have anything left for thematic placements.

Woohoo! I rolled 16/14/12/11/10/7. Blargh 16 to Str/Dex/Int/Wis. 14 to Con. 11 to Charisma. I be the charmer now.
At least with array, players know they have 3 decent rolls to create characters.

Orzel, Halfelven son of Zel, Mystic Ranger, Bane to Dragons, Death to Undeath, Killer of Abyssals, King of the Wilds. Constitution Based Class for Next!

I tend to prefer arrays. With Point buy and Rolling, I tend to see stereotypes. After you put your best roll/purchase to your primary ability score and your second best to CON, you rarely have anything left for thematic placements. Woohoo! I rolled 16/14/12/11/10/7. Blargh 16 to Str/Dex/Int/Wis. 14 to Con. 11 to Charisma. I be the charmer now. At least with array, players know they have 3 decent rolls to create characters.



in all of my custom excel character sheets I have a list of all of the arrays and I let people choose from them.  if they want a specialist array with an 18 or 20 cool.  But I have all of the arrays listed by the bonuses they give.

 While an 18 or 20 might look promizing 16,14,12,12,12,10 with 8 points look more appealing when you see how much more pluses you get.

Or you can get your randomization and make a table with all of the arrays on it and roll for them. 

Play whatever the **** you want. Never Point a loaded party at a plot you are not willing to shoot. Arcane Rhetoric. My Blog.

If given the opportunity, I tend to min/max to build the most powerful character possible. This tends to put other characters in the party at a marked disadvantage since they are generally less analytical when creating characters and put together combinations which sound fun but don't work out mechanically.  Rolling dice forces me to build around a set of scores which was not planned for and breaks patterns which optimize a character.  One difficulty which often presents itself is that most groups want the same character creation method for everyone in the party. They want all point buy or all dice-rolling and they want to force their method on everyone else.
When I am forced to use point buy, I design a character with single attribute dependency and mix and match race-class-feats-etc. until I arrive at what I consider to be the superior combination.  That is different from the player who always wants to play an elven ranger. They want point buy so they aren't forced to play as a wizard or cleric.
People who play with point buy, don't want to sit at table with random rolled characters that have better scores.
Random rolled characters should be forced to subtract points if they roll to well
I just created a character in my current playtest campaign with an 18, 2 17s, and a 16. I rolled in front of everyone and the dice were impossibly lucky to me.  We had no mechanism for subtracting from scores to balance them out.
My suggestion is to mix random generation with point buy with adding and subtracting to balance final scores.  Scores should be rolled in order but one pair of scores may be switched.
Some people may like point buy but don't assume that everyone at the table does.  I find that people who say that everyone in their group agrees with them often does not know their group. They think they speak for everyone when they are really voicing their own opinion and drowning out the voices of discontent.
One of my DMers in my group just likes rolling for the sake of rolling so he had us roll our dice. The other two who DM will often either just say we are doing x point buy or let us decide which ends up being point buy. I'm pretty indifferent to both really. All rolling means is that I must create my character at my friend's since that is where my dice are kept. We also were allowed to use an array this time as well but I didn't because I felt my rolled stats were better and well....... I just rolled them. 
IMAGE(http://i5.photobucket.com/albums/y152/RockNrollBabe20/Charmed-supernatural-and-charmed_zps8bd4125f.jpg)
Isn't this actually due to point buy totals, and not point buy as a system?  I mean, I could effectively get the same thing (consistently super-powered PCs) if I used a roll system like "Roll 6, keep the 3 best, for each of the stats".  In turn, I could create a point buy system that generates only barely above average PCs if I simply lowered the amount of points to work with.



You could get the same output from some rolling schemes,  true.  But Point Buy consistently generates super-human characters,  whereas die-rolls inconsistently do.  With die-rolls,  even with 6d6,  there's still a significant chance that I could roll a score of 9-12,  or even roll all scores of 9-12.  With Point Buy,  you have to actively decide that you're going to play with low scores spread,  because otherwise it will always generate a super-human. 

If you try to reign in the super-human characters,  you pretty much have to either say "18 isn't allowed",  or start everyone at 12 and give them 4 points.  As otherwise,  someone's just going to max their prime stat if you give them enough points to do it. 

Likewise,  in Point Buy,  you have to decide that everyone is going to be exceptional,  exceptional doesn't just happen. 

Certainly, this is more likely than in rolling (because of the element of player control or agency), but I ask if it is really an actual problem.  First, does it really matter if a player is using identical stats if he's still roleplaying the character in a fun and interesting way?  And second, is it really a given that just because the point buy system is in effect, players will make "clones"?  My personal anecdotal evidence tells me otherwise.  Players make the characters they want to make.



Depends entirely on where you play.  If you're playing with friends,  sure it may be less of a problem.  If you play in some form of organized play,  it may be a significant problem,  especially if there are rewards attached to it (A D&D form of Mtg organized play for example). 

If there's any kind of organized play,  especially with any kind of reward system be it social or material,  you'll end up seeing this in spades.  If WOTC ever does a D&D version of MTGO,  you'll see this become extremely prevelant.

Does this really actually happen enough to merit complaint.  Isn't peer pressure enough to prevent a person from making "Bob the Fighter II"?  And is it really so wrong that a player may want a certain experience from the game, one delivered by a particular character?

Also, I differ pretty strongly on the statement that point buy removes attachment to the character.  I'd argue that because each point invested represents a deliberate choice to create a specific character, point buys actually enables and reinforces attachment.  A point bought character is a creation, and not happenstance.




It's just the fact that if my mage with an 18 intelligence is threatened with death,  does it really matter?  The DM's going to let my create another character,  and I can just create an exact copy of that character,  so is death really a threat? 

How does point buy reinforce that attachment if I can simply just do the same thing over again?

I see no practical evidence that point buy discourages roleplaying.  First, I see no direct correlation between stats and personalities.  Identically statted characters need not have the same personality.  Second, I see no evidence that rolling for stats precludes the chance of players making characters with identical personalities.  If a player rolled, by chance, the same stat line, would he be obligated to have the new character have the same personality as the old one?  Why would buying stats be any different?

Furthermore, does point buy actually create characters without weaknesses?  Raistlin's personality emerged in part by roleplaying to his weak CON stat.  Why would this be any different if the player purposely chose to have the weak stat?  In fact, what if the character, gifts and flaws, were conceived before determining stats?  Wouldn't rolling risk upsetting that attempt at character conception?




Identically stat'd Characters don't need to have the same personality,  but there's no immediate feedback to the Player to suggest that this is a different character if the only difference is the name.  Whereas with a Rolled system,  one character might have rolled a 15 for Int and another a 9,  which signals to the Player that there is a difference and they should be played differently.

Buying stats is different because the only way there's variation within a class is through a choice the Player made in creation.  A player doesn't get a character with average intelligence,  or average strength,  the player actively has to choose that character.  That takes away the nudge to have Character A and Character B roleplayed differently.

Point-buy does remove characters without weakness unless the Player actively chooses to have a weakness.  Very few players are going to intentionally create a character with a serious drawback unless they're strong roleplayers.

I saw plenty of min/maxing, munchkining, and power gaming in 2e with rolled stats, so I'm inclined to believe that problem players will be problem player, regardless of stat determinism.  People who are obsessed with having a maxxed primary stat will re-roll and re-roll until they get a desired result, for example.



You're right,  those behaviors do occur in any system,  but point-buy facilitates it.  With Rolling,  the DM just has to put a finite line on re-rolls,  such as no rerolls if you have two scores > 14.  With point buy,  it's always going to happen,  the system always facilitates those kinds of behaviors without debilitating limits.

That begs the question: is that actually a bad thing?  Optimization doesn't mean "not roleplaying".  It just means optimization.  It just scales up capability.  The only time I see optimization being a real problem is when it is not group-wide.  And really, how is one player optimizing any different (in end result) to one player rolling ridiculously well?



Optimization at this fine level of control causes *enourmous* problems.  The largest of which is a very rapid power-creep.    Let's say I've optimized the next edition of D&D rules around the idea that a 15 in a prime stat is the average character with a normal curve,  and the majority of the Players find a way to consistently have 18's in their prime stats.

The most immediate effect is that all of my level design is now aimed at a significantly weaker character,  my Kobold is no longer a level 1 challenge,  my Ogre now is.  Soon,  Players will start to complain,  so in order to keep my game healthy,  I release material to try and correct that.

Soon a character with 15's becomes unplayable,  because now my content is aimed at a character with 18's. 

I didn't play 4th edition at all,  but isn't this something 4th edition had a major problem with?  I've had the impression from posts here that a very similiar scenario actually happened with 4th edition.   

And as a follow-up, does point buy actually encourage system mastery any moreso than rolling?  From where I stand, it takes great system mastery to make the most of what you may roll.  The system mastery may manifest in a different way, but it is still present.



I would argue that it does,  with rolling,  you make the best of what you have.  But with point-buy,  it's a careful calculation of where to put each point in order to obtain the best effect,  to the point where it can significantly effect game balance over time though it's consistency and power creep.

      

    

In the games I run I let players do what makes them happiest. Some roll, some use an array. Those who use an array only do so because it makes creating the character go a little faster.

As for myself I prefer to roll. I enjoy the greater randomness it allows for. I choose my class and roll and take what I get. I do not choose class based on attributes but rather abilities. 

I would argue that it does,  with rolling,  you make the best of what you have.  But with point-buy,  it's a careful calculation of where to put each point in order to obtain the best effect,  to the point where it can significantly effect game balance over time though it's consistency and power creep.

      


    




And I would argue the whole argument is moot.  Allow those who want to roll their stats to do so and those who prefer point buy to use that system.  Everyone's happy.
 

At the end of the day, I like that we've got a lot of choice. Determining attributes is the one place where D&D is consistently successful in giving us a range of methods and allowing those methods to function at the same table within reason.


Put it this way: if I've got one player with a character they used points for and another who determined randomly, then what I've got the same potential variance as if both rolled randomly in terms of game balance. If the game is designed with the idea that some or all of the players will be rolling their attributes up then the game should take most of that into account and everyone's happy.


I played a lot of neverwinter nights and one thing I noticed with the point system there is unless I really only wanted one stat, I never got the 18. In fact, I'd rarely push my attributes over 14 and my characters were decent in spite of the fact that I didn't have more than one 16 and no 18's at all. To me, that says that the real fight is how attributes are perceived rather than how they're determined.

-Point buy as you note,  encourages min/max play,

-Point buy also tends to generate "Clone wars" type play,  where nearly every character of a certain class  is identical to the next guy,  in all but name.

-Point buy also removes the attachment to the Character.


-Point buy discourages Roleplaying.


-Point buy encourages System-mastery over gaming,



Yeah, I think I can safely ignore what anyone against point-buy has to say. Thanks for the ammo.
I am amused by the system mastery comments, as if encouraging you to learn and get better at the game is a bad thing.

Take a game like Chess.  All those complicated rules with different pieces having different movement, tons of tactics and openings and fianchettoes and all that nonsense, all that requirement for system mastery just gets in the way of the gaming.

Yeah.  Right.

"encourages system mastery over gaming/roleplaying/realwayplaying" is the excuse scrubs tell themselves to make themselves feel better about not feeling bad that they're not better.
D&D Next = D&D: Quantum Edition
Thank you for the clarification.  I was getting increasingly concerned that I was missing some vital piece of information.

So, that beng established, is the 4e version of Point Buy (no stat lower than 8, and mathematically untenable to lower another below 10) less ornerous, or more palatable to those who don't like point buy in general?

I guess what I'm trying to get at is whether people who object to point buy (in general) do so because of the propensity for min maxing, or some other reason.  If the min maxing could be systematically be mitigated, would the objection be dropped, or at least lowered?



I just like rolling for stats.  The random factor appeals to me, so I simply won't use point buy or arrays.  The closest I ever came to using an array was with a DM who had 6 players and we each rolled 4d6 once and used the same 6 numbers.  I ALMOST didn't play in that game, but since they were all good buddies and it was rolled, I stuck it out.
With the exceptions of the most extreme examples of each genre - roll-in-order for rolling and crazy permissive point-buy options that let you plummet stats down to 3 or something - the differences betwen rolling and point-buy are pretty insignificant. Rolling essentially is little more than point buy with a bit of noise thrown in there. That's because when you let people roll, they just arrange the stats more or less as close as they can get to a normal-looking point buy anyway. Rolling doesn't magically make charisma a good stat for fighters. It just gets the lowest score that someone happened to roll instead of the eight that a point buy would put there, and under most rolling systems, that's going to be something a lot like "eight". A rolling system could produce a fighter with fifteen charisma, but only if the fighter rolled at least fifteen six times. With all but the most absurd rolling systems, the player is more likely to choke on the dice while rolling than to roll fifteen+ six times in a row.

Rolling doesn't do anything particularly meaningful to make characters more distinct from each other than point buy does, with two exceptions:

- A character who gets better rolls will just be linearlly better. Yay?
- If you imagine that characters walk around with copies of their character sheets floating over their heads, than the minor bits of noise introduced by rolling will superficially make the characters appear different, although the actual impact of their stats on the game world - you know, the things that anybody inside of the game can actually see without the magical charsheet hats - will be largely the same anyway.
Dwarves invented beer so they could toast to their axes. Dwarves invented axes to kill people and take their beer. Swanmay Syndrome: Despite the percentages given in the Monster Manual, in reality 100% of groups of swans contain a Swanmay, because otherwise the DM would not have put any swans in the game.
- If you imagine that characters walk around with copies of their character sheets floating over their heads, than the minor bits of noise introduced by rolling will superficially make the characters appear different, although the actual impact of their stats on the game world - you know, the things that anybody inside of the game can actually see without the magical charsheet hats - will be largely the same anyway.

There is some truth to this, though.  Even in real life, if you think that you're more charismatic than average, then you're more likely to present yourself in such a way as to make that apparent; if you think that you're dumber than average, then you're less likely to look for a clever solution to a problem, because you don't think you'll be able to find it.

Even though the mechanical effects are fairly small, a minute difference can change your approach to playing the character, thus magnifying the impact of that difference.  Even though someone with Int 12 is only +2 to Int checks compared to someone with Int 8, the smarter character is more likely to put herself into situations where Int checks will be called for.

The metagame is not the game.
I like roll in order, switch any two, best out of all the methods.

My least favorite method is roll place anywhere. To me it combines the possible imbalance which is the drawback of rolling with the predictabity of where stats are assigned of point buy. Ugh.

Point buy falls somewhere in between those two. It has its good points, but one thing about it annoys me-- the math. Not the 1st level math. The fourth level math and eight level math when stat bumps occure that I should plan for at 1st level when doing point buy or else I'm wasting points. Having to plan levels ahead annoys me. But that may just be me. 
IMO best solution is controled rolling.

The best solution is to reach a consensus at each individual table.

As a side note, I'll point out that the RPGA pretty much has to use Point Buy or some other non-random method exclusively.