How did 32 point buy become standard instead of 25?

[Endarire]

Greetings, all!

The 3.0 and 3.5 DMGs say for point buy that 25 points is the default. Nowadays, many people assume 32ish is the default. How did that happen? How does this change game balance?

[eggynack]

Dunno how it became standard, but the balance impact tends to be positive. Low tier classes are simply more MAD on average.

[ExLibrisMortis]

Eh, stat inflation happens in a lot of games. Probably because people like to have power and options, and easily meeting cool-but-nuisensical requirements (Dodge and Combat Expertise for Karmic Strike - 13 int and dex for a str/con-synergistic feat) is a balance bonus, as eggynack mentioned.

[Inevitability]

As Eggynack says, it's fairer. High-tiers can typically get by with one high stat. Low-tier classes, however, need to boost anywhere between three to five stats.

There are some exceptions to these rules (Duskblades and Healers) but generally heightening point buy is better for the classes that need a boost. Not to mention that it's easier on the players.

[Eldariel]

It's just a bit more freedom in character creation/more tools to play with. Besides, 25pb is based on erroneous math by WotC anyways; accounting for the reroll rules, the average result of 4d6b3 is closer to 28. But yeah, lower point buys just make playing anything not SAD rather inconvenient, so people have shifted to the more enjoyable format. Mind, I've played 25pb games; generally I just roll a spellcaster and put 18 in my casting stat and some minor stuff in others. If I play a warrior or something though, I'll be looking at a line of 14s if I want any kind of versatility.

[CharonsHelper]

Dunno how it became standard, but the balance impact tends to be positive. Low tier classes are simply more MAD on average.

That's why I actually prefer using an array (potentially 2-3 options), though I can't usually get my group to agree.

[icefractal]

Is 32 standard? I tend to think of it as 25 = low, 28 = standard, 32 = high. I usually use 32 when running, though.
As for reason, for me it's two things:
1) MAD classes are generally not stronger than SAD classes, so penalizing them is not a desirable effect.
2) With 32pb, people have more freedom to take a high stat for RP reasons (Strength on a Wizard, say) without being mechanically crippled by it.

For that matter, I find that for many campaigns, you can just let people choose their stats and they'll be completely reasonable with it. That works better with an existing group that knows each-other though.

[Elkad]

My last game I gave my players 35 points.

They all like MAD (and lower tier) character types, no reason to not encourage them.

[Zanos]

For what it's worth, the 3.x cRPGs both used higher point buys. NWN 1 used 30, and NWN 2 used 32. That might have had some influence on it.

[Thurbane]

I've wondered this myself. TBH, it doesn't really impact our games, since we pretty much always roll anyway.

Some good points about it helping low tier MAD classes, though...

[Tvtyrant]

I use 18/16/14/12/12/10 as an array for the same reason, the lower the points buy the weaker mundanes get but casters don't care. Especially Warlocks and Dragonfire Adepts, they can be decent with no good stats at all.

[weckar]

I never was the hugest fan of "Heroic" arrays. I like my 8s and even my 6s as I find weaknesses can define a character more than strengths.

[Gwazi Magnum]

Personally for me it was because I entered 3.5 after having played Kotor for years.
And Kotor always used 32 Point Buy so it's just what I naturally gravitated towards.

[Jowgen]

I personally operate under a 28 pb, partially because of the average 4d6b3 being closer to 28, also because I find it works out quite nicely in stat arrays (e.g. so I can pump by base main to 18, or go for 2 16. Hmmm...)

Not a fan of 32 myself. Too much discrepancy with your average NPC/Monster stat array (be it elite or non-elite).

[Sandsarecool]

People like power. A few more points doesn't blow a character's overall power out of the water, but it gives them an edge, which players like, and helps deal with aforementioned prerequisites and dealing with MAD characters a little better.

[Fizban]

In a metagame where everyone assumes you have an 18 in your main stat and 14+ con, a high point buy is required to actually get that. I would say the trend originated with people using ridiculously powerful made up rolling methods, which causes a significant inflation of expectations. And don't be surprised if the same people that invented those rolling methods turn up their noses at point buy for being "overpowered."

The Neverwinter Nights games may have also had an influence: NWN2 gives you 30 or 32 point buy.

[Khedrac]

I found these boards after 32 became standard so I cannot comment on that.
Our games use 28 primarily because that was what Living Greyhawk used (as a 'high' point buy).

I do find it interesting how everyone says that the higher arrays help the weaker classes more. While this is true, it is actually only true once one gets to higher levels and over 28 point buy.

The reason I think this? - Well (nearly) everyone on the boards (nearly) always assume that the primary spellcasters start with an 18 in their casting stat. This can be done at 28 point, but involves sacrifices, it is much harder at lower totals.
And spell DC rocket tag is one of the main ways primary casters utterly dominate the high-level game. In the low-level game boosting those save DCs is harder, and one of the main ways is to start with an 18 in the casting stat - something that the higher point buys makes automatic (or above 18 with races with a state boost in the right place).
Yes clever casters don't rely on spells that give their opponents a save, but when you have to you want that save DC in the stratosphere.

Case in point: the playground talks a lot about "Save or Lose" spells and how useless fireball is as all it does is damage.
If you opponents have a 25% chance of making their saves then the Save or Lose spell is clearly better, but what about if they have a 50% chance? At this point the effects of a spell on a successful save start looking much more important. This is how high-level AD&D played - high level opponents were going to make their saves so you used spells based on what happened when they made not when they failed - hence fireball was a good spell, it did something when they saved.
Now back to low level D&D 3.5 - from 16 to 18 in the primary stat is only +1 to the DC (though some melee clerics might even start with a 14), but on a 2nd level spell that is taking the DC from 15 to 16. If you assume an opponent with that as a poor save (so say +1) you have just gone from a 65% success rate to a 70% - not a huge difference. Against an opponent with a good save of +4 (because sometimes one has to) that goes from 55% to 60% - while still only a 5% increase it is in some ways a more important 5% because of the long run effects:
3 casts a 60% (40% to save) gives a 6% chance the opponent will make all 3; 55% gives 9%. Over multiple casts that extra +1 to the DC makes quite a difference.

[eggynack]

The reason I think this? - Well (nearly) everyone on the boards (nearly) always assume that the primary spellcasters start with an 18 in their casting stat. This can be done at 28 point, but involves sacrifices, it is much harder at lower totals.
It's not that much harder. 28 point buy allows 18 intelligence, 14 constitution, and 11 dexterity. 25 means dropping the dexterity but keeping the intelligence and constitution. Even, say, a 15 point buy will only mean something like 16 intelligence and 13 constitution, which isn't a huge loss. So, you can maintain these main stat scores with really low point buy. Because you can afford the loss of secondary stats. Because you're SAD. This in and of itself invalidates a lot of your following arguments about the value of a single point of DC, because you're not necessarily losing that point. Of course, as you yourself note, a caster of any type can easily avoid spells that rely fully on saves if necessary. Y'know, spells like haste, silent image, or grease. Notably, while those latter two spells have saves associated, they still do cool stuff even if your opponent has high likelihood of success against those saves. You also have stuff like stinking cloud, which has impact when it fails, which hangs around doing its damage in the longer term, and which spreads its success chance across multiple enemies.

Along those lines, one other cool element of casters is that they don't really need stats at all. Give a caster a single 11, after race, and they'll still have full access to their spells at every level except level three. Give those stats to a melee class of any stripe and they won't accomplish much of anything, but all those spells that don't rely on stats, which is a lot of them, will operate just fine. Give those stats to a druid, and they'll do all that casting while also running an animal companion with stats that're just fine, as well as personal melee through wild shape. Hell, that's still the case if you give the druid all 3's, though at that point the casting part is likely gone.

And that's all stuff I'm noting while barely touching on how the melee classes are operating. If the caster is losing a point in their main stat, then how do you think the fighter is doing? Or, hell, the monk. A wizard has the ability to do things that ignore this minor additional failure chance. A fighter cannot. All the combat oriented stuff a fighter does is going to rely on strength in a way that only some of what a wizard can do has that sort of reliance. And the fighter touches on that stat possibly multiple times a turn. It means nothing that you can point to a disadvantage striking the wizard if I can point to a greater one striking melee classes.

[Gemini476]

I found these boards after 32 became standard so I cannot comment on that.
Our games use 28 primarily because that was what Living Greyhawk used (as a 'high' point buy).

I do find it interesting how everyone says that the higher arrays help the weaker classes more. While this is true, it is actually only true once one gets to higher levels and over 28 point buy.

The reason I think this? - Well (nearly) everyone on the boards (nearly) always assume that the primary spellcasters start with an 18 in their casting stat. This can be done at 28 point, but involves sacrifices, it is much harder at lower totals.
And spell DC rocket tag is one of the main ways primary casters utterly dominate the high-level game. In the low-level game boosting those save DCs is harder, and one of the main ways is to start with an 18 in the casting stat - something that the higher point buys makes automatic (or above 18 with races with a state boost in the right place).
Yes clever casters don't rely on spells that give their opponents a save, but when you have to you want that save DC in the stratosphere.

Case in point: the playground talks a lot about "Save or Lose" spells and how useless fireball is as all it does is damage.
If you opponents have a 25% chance of making their saves then the Save or Lose spell is clearly better, but what about if they have a 50% chance? At this point the effects of a spell on a successful save start looking much more important. This is how high-level AD&D played - high level opponents were going to make their saves so you used spells based on what happened when they made not when they failed - hence fireball was a good spell, it did something when they saved.
Now back to low level D&D 3.5 - from 16 to 18 in the primary stat is only +1 to the DC (though some melee clerics might even start with a 14), but on a 2nd level spell that is taking the DC from 15 to 16. If you assume an opponent with that as a poor save (so say +1) you have just gone from a 65% success rate to a 70% - not a huge difference. Against an opponent with a good save of +4 (because sometimes one has to) that goes from 55% to 60% - while still only a 5% increase it is in some ways a more important 5% because of the long run effects:
3 casts a 60% (40% to save) gives a 6% chance the opponent will make all 3; 55% gives 9%. Over multiple casts that extra +1 to the DC makes quite a difference.
Fireball isn't useless because all it does is damage, it's useless because it does 1d6/level against enemies that used to have 1d6hp/HD (OD&D) or 1d8hp/HD (AD&D) but now have (1d8+CON)hp/HD... while the fireball damage stayed the same. Also, some of those have differently-sized hit dice or more hit dice per CR - zombies have 2d12 for every hit die they had in life, for instance.

And since the DM metagame isn't to have fifty kobolds any more, or 40-400 orcs, the massive AoE is less relevant. And, well, the system as designed keeps the saves at roughly the 25%-50% level rather than going down to AD&D's 95%.


Similarly, Turn Undead has been massively nerfed over the years... probably because zombies got more hit dice and others started to get Turning Resistance because apparently you can't just have the Cleric turn the Lich.

[Zanos]

Case in point: the playground talks a lot about "Save or Lose" spells and how useless fireball is as all it does is damage.
That's bull, the prevailing opinion is that SoL spells are bad, specifically because they do nothing on a failed save. I've seen over and over that people recommend BFC spells without saves, like fogs and walls, or spells that still have a nice effect when people make their saves, like Grease. In fact, in Treantmonk's original "God" Wizard guide, which isn't from the playground but has a lot of influence on the prevailing opinion on wizards, he rags on SoD spells for about a paragraph.

SoL spells are only considered good on the playground if they effect a lot of targets, require the target to pass multiple saves to be safe, make an area too scary for the enemy to go in to, or have a very good effect even when they pass their save.

Fireball is considered bad by most Playgrounders because it has to go against basically everything. 1d6 damage per level is not particularly good. Reflex half is not particularly good. SR: Yes is not particularly good. And fire resistance/immunity are more common than any other type of elemental defense on monsters. I personally still think it's a good spell...to prepare maybe once. If your DM likes large numbers of weaker enemies and you can bunch them up, fireball is pretty good. In pretty much every other circumstance, it's eclipsed by it's other 3rd level brothers, particularly Haste and Slow. And your argument for it being good is pretty terrible, if your spell save DCs are low, that hurts fireball a lot. It's damage when enemies pass their saves is absolutely pitiful.

[Exocist]

Similarly, Turn Undead has been massively nerfed over the years... probably because zombies got more hit dice and others started to get Turning Resistance because apparently you can't just have the Cleric turn the Lich.

Kind of off-topic, but I would disagree with this one.

Divine Metamagic is a pretty good class feature :P

------------------------------------------------------------

As for the 32PB, with that you can put a 16 in main stat (10pts), put a 14 in 3 other stats (18pts) and leave the last 2 stats at 10 so you don't get penalised.

What does this mean for casters? Pretty much nothing, they're mostly just going to go 18 (16pts), 14 (6pts) and 16 (10pts) and dump the other 3 stats because [Primary Casting Stat], Initiative (DEX) and CON are the only stats that matter for them.

But for martials, it means you can put a 16 in DEX or STR (whichever you're more reliant on), have a 14 in both CON and the other of DEX or STR, as well as putting a 14 in another stat in case you're like a Monk where you need a mental stat for class features. Leaves you with 4 points left over, so you can either shore up those 8s, or boost one of your 14s to a 16.

4 stats at +2/+3 modifier means that your MAD isn't as pronounced, but the Wizard/Cleric/Druid/Whatever is just as SAD as they are at 25 points.

[CharonsHelper]

Fireball isn't useless because all it does is damage, it's useless because it does 1d6/level against enemies that used to have 1d6hp/HD (OD&D) or 1d8hp/HD (AD&D) but now have (1d8+CON)hp/HD... while the fireball damage stayed the same. Also, some of those have differently-sized hit dice or more hit dice per CR - zombies have 2d12 for every hit die they had in life, for instance.

Mostly this. The spell has remained the same, but the rules have shifted to make it do a much smaller % of a foe's life damage. In Pathfinder you can actually make it solid again by being a wizard Evoker who dips a level in crossblooded sorcerer (orc/draconic) and takes the trait to up your caster level +2 (max of character level). In total you're a wizard level down, but each die of fire damage is d6+3. So (for example) at 8th level your fireball does 8d6+24 damage, which is much better than the 8d6 it would otherwise be.


And since the DM metagame isn't to have fifty kobolds any more, or 40-400 orcs, the massive AoE is less relevant. And, well, the system as designed keeps the saves at roughly the 25%-50% level rather than going down to AD&D's 95%.

I think that the #s varied a bit - but it not being too viable is pretty much a result of how the 3.x combat system works. As a GM trying to keep track of scores of foes isn't really viable and would slow the game to a crawl. Even 4e minions had limits, though that was one of the few things which I did like about the system. I suppose it might work better in 5e since they don't use a grid, but it would still be trickier than early editions.

[Fizban]

While I'm hesitant to continue the derail, I'd like to point out that "no one" who casts Fireball ever does it for base damage thanks to metamagic and all sorts of other boosters. Damage spells require some build investment like they should. If you increased the base damage it would only make the uber-metamagic builds that much worse. Contrasted with every other type of spell that works without any investment at all, it's obvious what the problem is.

Exocist gave the breakdown on what sort of arrays you can make with 32 points and pretty much sums it all up: you can get multiple stats high enough to matter without completely dumping everything else, which is nice. I really, really hate having to leave even an 8 in any stat and will gladly forgo an 18 in order to avoid doing so, optimization be damned.

[Pugwampy]

Everyone wants two 18,s on their abilities so I say give it to em . Bad rolls can affect the whole game so i wont force anyone to dance with the dice gods for good stats . Most dont mind trying their luck .

They can either choose to roll for stats or get points .

In my case its heroic rolls . 4 d6 take away lowest yadda yadda plus a D20 hail mary roll to swop out with their one stat . If its a 19 or 20 roll , they get an 18 .

I dunno if my home brew point buy has a name . Basically all stats are set to 10 and they get an extra 20 points to p... away as they want to a maximum 18 of course . Thats a good double 18 and one 14 , if they wanna squeeze .

I usually suggest to players wanting an AB hog class to roll for it .

[Aetis]

I still use 25pt buy!

Dumping to 6, 4, or even 3 ability score(s) is not uncommon.

[CharonsHelper]

I still use 25pt buy!

Dumping to 6, 4, or even 3 ability score(s) is not uncommon.

Point-buy doesn't normally allow that.

[Aetis]

Point-buy doesn't normally allow that.

I guess that makes me a cheater. :smallamused:

[tsj]

I would suggest the following :

The class chosen determines what stats may start at 18 and the remaining is point buy?

In order to help weak classes...

Example... fighter

Str, Dex and con start at 18
The remaining 3 would then have 25/ 2 point buy = 12 points.

Or 25/6 * 3 = 25 / 2

[Aetis]

I think it just depends on what philosophies you play with.

Do you want less power gap between the classes? (to justify giving extra points to certain classes)
Are the PCs special snowflakes in your world? (to justify giving more points to PCs over monsters/NPCs)

And so on.

[Zaq]

When we're talking about baseline stats (and therefore numbers that have a long-term effect on gameplay and which come up again and again over the life of the character), penalties hurt more than bonuses help. It's not so much a question of whether the Wizard starts with 16 INT versus 18 INT; rather, it's a question of whether the Rogue can have enough DEX to sneak around reliably and enough INT to fit something close to as many skill points as they need in and enough CHA to be the smooth talker they envision the character to be.

To a full caster, being able to fit in a few more points on the highest stat means an extra spell per day (nontrivial, but not a dealbreaker if you can't get it) and a +1 on your spell DCs (again, very nice, but the character is far from unplayable without it). Whereas a Rogue who has to cut things to the bone and who can't afford a good investment in INT after putting in enough DEX to be roguish (it's likely their attack stat, a major source of their AC, and a source of Sneak Attack in the form of initiative—and of course many iconic Rogue skills are based on it) and enough CON to keep their aggro-attracting d6 HD butt alive, well, they're going to be running a skill point deficit (even more than usual—I know full well that there's no such thing as "enough skill points"), so they might not have access to a whole branch of abilities they were hoping to have. That's a disproportionate cost.

And as others have stated, there's a lot of feats that require investment in stats that don't seem to necessarily fit the theme of the feat. The difference between high point buy and low point buy might not just be a +1 to hit here and a +1 to saves there—it might be the difference between being able or being unable to perform your desired combat style at all. Even the feats that require stats that make sense can sometimes be pretty pricey (it's not always trivial to hit 17 DEX for some of the higher-up archery or TWF feats, especially if your GM is overly harsh about not letting items help you meet prereqs).

Speaking as someone who's played with Iron Chef a lot, there's also plenty of obvious and not-so-obvious skill point costs hidden throughout the system, as well. Maybe a character with a base class that offers a low number of skill points aspires to get into a PrC that requires a little bit more investment in skills than they can normally handle (either the PrC requires skills that they wouldn't normally take, or it requires more skills than they normally can afford, or whatever). The difference between having to dump INT and getting to invest even a 10 or a 12 in INT might mean the difference between getting into that PrC on time and not.

Sure, not every character concept is going to be completely screwed over by having low stats. But plenty are, and that's frustrating.

The other side of the coin is that up to a point, throwing additional stats at characters who already have enough stats to hit their baseline goals doesn't do a hell of a lot. I mean, everyone benefits from having higher saves, or better to-hit, or extra skill points, or a bigger bonus to throw at some skills. We like having good stats. But it's a much smaller benefit than the difference between getting a feat and not getting a feat, or getting into a PrC and not getting into a PrC. If a Fighter wants to do Improved Trip tricks and needs Combat Expertise, that Fighter is going to be better off with 16 INT than with 14 INT just because everyone can use a spare skill point, but that's a much smaller benefit than the one gained from getting 14 INT instead of 12 INT (which is, of course, the difference between Combat Expertise and no Combat Expertise). (Even if the Fighter throws those extra points into STR or CON or whatever, it's still a disproportionate benefit.) It's not going to break anything to allow already good numbers to become a little better, but it might cause some problems if we can't allow those numbers to ever become good.

(I will emphasize that this is up to a point. If we allow buying up way beyond 18, or if we allow ridiculous point-buys that are way the hell beyond the scale of the normal game, that's another story. But yeah.)

Now, of course, this is a different question than the one you asked. I've been defending high point-buy over low point-buy, but I haven't addressed when or how or why that became standard. And the answer to that is that I don't know when that shift happened. I know that I'm happy with the result, and I hope I've given an idea of why I prefer it this way, but I don't know where the shift occurred.

[LoyalPaladin]

Our group went from rolling stats (3d6+1), to 34 point buy, to 32 point buy, to 36 point buy. So we've never been "standard". Recently I've seen a few PBP games on here that feature a 42 point buy system. Possibly to accommodate for the more MAD classes.

[J-H]

Most people play D&D partly because it lets them do Cool Things. It's easier in D&D 3.5e to do More Cool Things if you have higher base stats.

I think this holds true unless you have a group whose idea of Cool Things is sneaking by and barely surviving the apocalypse.

[Gruftzwerg]

I personally operate under a 28 pb, partially because of the average 4d6b3 being closer to 28, also because I find it works out quite nicely in stat arrays (e.g. so I can pump by base main to 18, or go for 2 16. Hmmm...)

Not a fan of 32 myself. Too much discrepancy with your average NPC/Monster stat array (be it elite or non-elite).

imho THIS.
I have been playing d&d p&p & video games since about 25y. And most of the video games use 32p buyout (if there is a buyout system). I think it's because of solo class balance.
It's as others already pointed out, T1 & T2 are SAD, while the weaker classes are MAD. So you just hurt the weaker classes and increase the gab between them.
a 32p fighter is at least for some part of the 20 lvls not complete garbage compared to a wizard. While a 25p fighter will have problems to shine at all (high str, or some defense dex/con, or some int for feats?). He will at least either lack offense or defense.
Or try to build a good monk with 25p, its impossible.

[bekeleven]

For those wondering:

If you roll 4D6b3 and throw out arrays with no scores above 13...

The median array is 16,14,13,12,11,9 = 29 point buy.

However, the average (mean) array is 15.90, 14.34, 13.08, 11.96, 10.49, 8.56.

It just so happens that due to the shape of the distribution, 4 of these numbers "round" up when selecting the median (including one that's under 10½), and 2 round down. Rounding just the 10.49 down brings the point buy to 28. Rounding all of them down or up gives a point buy range of 24-32.

If you run the same simulation without dropping arrays with no numbers above 13, You get the following median: 16,14,13,12,10,9. In other words, it's exactly the same except the number that was previously on the cusp moved down 1. 28 points.

The average, however, moves to 15.67, 14.18, 12.95, 11.75, 10.41, 8.50. If you round these to 16,14,13,12,10,8, you get 27 points. Rounding all of them up or down gives a range from 23-31.

I can't fathom how they came up with 25 points as standard.

[Name1]

Personally I believe that you should never play in a PB that makes it unable to gain at least the Standard array of 11,11,11,10,10,10. I played in a game where everyone had 8s across the board, and it wasn't that great.