The Rise of Point-Buy and Joys of Min-Maxing

By Stephen Radney-MacFarland (or Creighton's Sinister Nemesis)

Last week my dear friend and brilliant designer Creighton Broadhurst unleashed a diatribe against the point-buy method of generation of ability scores and min-maxing. I’m here to tell you he is dead wrong. And I believe I can prove it.

Tumbling Dice by Justin D. Russell

March of History

The generation of ability scores has a fascinating history that slowly creeps toward choice rather than randomization. Consider the first method to determine ability scores found in Men and Magic, starting on page 10. It’s a straight 3d6 in order, let the chips fall where they may. And comparatively, you didn’t get much from your ability scores, except prime requisite scores, Constitution, and Dexterity, what a high ability score meant was almost only descriptive and sometimes incredibly vague. Take for example one of the “benefits” of having a high Intelligence or Wisdom.

“[The ability score]…will also affect referees’ decisions as to whether or not certain actions would be taken.” (Men and Magic page 10)

The remainder of the history of the game is a steady march toward ability score definition and blunting the effect of randomization in their generation. Greyhawk and Eldritch Wizardry gave more for abilities to do, and as early as the AD&D Dungeon Master’s Guide (and likely even earlier in Dragon Magazine) many and much more lenient ability score generation methods saw print. Creighton’s favored method is a more stringent version Method I in that book, which is the least lenient method presented in that tome. Why did Gygax abandon the first, more stringent method, still used in the D&D game? He  offers this explanation:

“While it is possible to generate some fairly playable characters by rolling 3d6, there is often an extended period of attempts at finding a suitable one due to the quirks of the dice. Furthermore, these rather marginal characters tend to have a short life expectancy – which tends to discourage new players.” (AD&D Dungeon Master’s Guide page 11)

The first systems have a hard time just making viable characters, and softening the curve when it comes to the random generation of ability scores is only progress of a kind. While it is less likely to create a truly crap character, it is indeed possible to do so. On the other end of the spectrum, it is possible to create very over powered characters, which I think is at the heart of all nostalgia to roll for ability scores. It’s the gambler’s dream. This time I’ll win big, but it is more likely to live with disappointment, mediocrity, or mere contentment. Its harsh reality is the day-to-day malaise of the casino: big winners are a rarity, and long-term winning comes only in a lifetime of dwarfed peaks and shallow valleys. Creighton even expresses such doldrums with the following statements:

 “I might not get the stats I need to play the character I want to play. Such is life. I’ll struggle through.”

Instead, my worthy opponent has contented himself not with the character he wants to play, but one that he now intends to play because he is forced to by the “capricious gods of chance.” And like some shrill parent imploring us to eat our blanched and limp boiled cabbage. But like those parents just attempting to inject some scant amount of vitamins in our system, Creighton has reasons that he thinks are good for us, but all of them talk about the creativity of absence rather than the creativity of options. While it is true that creativity can be given a nudge though limitation or randomness, those should be exercises rather than the main impetus.

Enter Madness

Creighton then argues that you should place the ability scores in the order that you rolled. His scheme is Method I from the AD&D Dungeon Master’s Guide without the flexibility of putting the scores where you wish toward the end of reducing min-maxing, but many players will just make an elf rogue instead, and min-max without the satisfaction of creating the character they originally wanted to play. He says this in an effort of destroying the hated min-maxing in the game and forsaking the love of the mechanical parts of the game, but good luck with this. As with life, it’s within everyone’s self-interest to capitalize on what we do well and to avoid what we do least well or find some way to shore up those weaknesses. In short, he is asking players to go against their gut toward the ends of creating a “real person.”

Of course, when creating a fantasy RPG character, we are not creating a real person. We are creating a hero in a story that is larger than life. Min-maxing is a way to do this. You create a hero that achieves great heights and possesses a set of weaknesses by comparison.  From Homer to Martin, this is the formula of the genre.

Countering Madness

But enough of merely bludgeoning my friend’s arguments. It’s easy enough to tear something down, but that should not be a genuine test of my argument. My argument is that point-buy is the best method for the game in general, and while I have softened the contrary opinion, I have not yet argued that point.

First off, Gygax in his opening before unveiling a host of rather liberal generation methods in the AD&D Dungeon Master’s Guide provides one argument for leniency while determining ability scores. Imagine yourself as a ripe youth, playing your first game of D&D. You follow the rules outlined in Men and Magic, or either of the Basic Sets (up to Mentzer’s Red Box) chances are you’re going to roll an abysmal character. Not only that, when you realize that, you’re going to roll again. It’s a trick that more experience players have figured out, and roll with relish. After all, none of the rules say you have to play the first character that you roll, and even the more liberal AD&D Dungeon Master’s Guide methods seem to embrace and put limits on such practices. Eventually, you’re going to get a character you’re happy with, and it’s going to have over average scores. Even Mentzer’s Red Box embraces this to a lesser extent, instructing you to reroll any character with a high score of less than 9 or one that has two scores less than 6.

These games have depth and can seem and often are quite intimidating at first blush. The new player is going to do a lot of things “wrong” be the mistakes are real or imagined by the crunch monkeys. Starting out with either substandard or meh stats is the wrong way to go. If anything, you would want to give new players above average stats. Consider it a handicap of sorts.

Why is this important? First impressions are important. RPGs are about fun, and the less fun you have your first time, the less likely they’re not going to come back. I know RPGs can be a cliquish tribe, and some folks out there are saying something akin to, “good, we don’t want those mouth-breathers playing,” but shut the hell up. Games, especially RPGs create fun. They are not nor should they be treated like some elitist country club. Some folks may not find them their cup of tea, but we should never passive-aggressively kill enthusiasm.

I submit that, short of using some handicap scheme for new players, a point buy method with clear guidance to the new player is a better way to go toward that goal.

My second reason may seem a bit more obscure but is near and dear to my heart. Over the design of 5th Edition D&D, you heard a lot about bound accuracy. There’s a good argument that in 3.5, 4.0, and Pathfinder as a character reaches higher levels, accuracy gets very swingy. The problems start to manifest as low as 3rd level and get worse with each level. This has everything to do with the bonus economy in the game and a lack of good design principles on the game’s back-end. While there are many factors to the issue, one of the main issues involves the core percentage spread intrinsic to ability scores and the fact that your highest one will continue to rise, and your lower stats stay stagnant. The 5th Edition’s solution is shallowing the spread of bonuses outside the ability scores and to throw magic items to the wind and submit their balance is DM’s job.

Believe it or not, point-buy methods help to reduce this problem. Given that most point buy systems start at the average and add points to it, it shallows percentage divides coming from Ability scores. You may find that “unrealistic,” but I’ll counter that realism and RPGs don’t mix well. Don’t we get enough realism in our life? It’s heroic fantasy for [censored] sake.

Which brings me to my last point: fantasy games are escapism. We escape our normal bounds and play a hero or villain that might be somewhat similar to us in personality, but is capable of things that we are not. Creighton’s argument boils down to if you can’t have the one you love, love the one you’re forced to be with. This argument is nothing more than the grumblings of a grognard shaking his fists at the kids for not walking to school in the pouring rain and driving rain because the school buses saw invention and implementation.

It’s obvious that the momentum is moving in the opposite direction of Creighton’s desires and has been for about 30 or so years.

Path to the (Far) Future

This summer sees the release of Starfinder, a new RPG that takes the Pathfinder franchise into the realms of science fantasy. In that book, the main method for generating stats is point-buy with optional rules for rolling presented as a variant. But the point-buy method used in that game takes another shot at solving the problems that the divergence of ability scores causes at higher levels, but you are going to have to wait until this summer to see how.

Time will tell, but I think as these games progress we’ll all find that this is the best way to create the character that you want to play and flex your RPG creativity to new and greater heights.

A Note From Creighton

Thanks very much to Stephen for taking the time to putting the counter argument. As always, a thought provoking read from my yank chum.

But what do you think? Let us both know, in the comments below!

Creighton is the publisher at Raging Swan Press and the designer of the award winning adventure Madness at Gardmore Abbey. He has designed many critically acclaimed modules such as Retribution and Shadowed Keep on the Borderlands and worked with Wizards of the Coast, Paizo, Expeditious Retreat Press, Rite Publishing and Kobold Press.

Please note: I reserve the right to delete comments that are offensive or off-topic.

Leave a Reply

Your email address will not be published. Required fields are marked *

18 thoughts on “The Rise of Point-Buy and Joys of Min-Maxing

  1. I think that your esteemed opponent either fails to understand, or intentionally misrepresents, the argument you presented.

    Having gone through many years of min-maxing versions of the Great Game, I am personally happy to return to 3d6 in order, creating a far more diverse-seeming group of characters than other methods I have used create. All too often, PCs made from point buy systems tend to be extremely similar to one another, like Sontarans created from the same clone batch.

    There is something to be said for systems that allow you to craft a specific character exactly to your desires….but 3d6 in order creates a stronger sense of the fantasy world being a “real” place, in which the PCs happen to dwell, but which PCs are not necessarily the focal point of, whereas the other tends more toward a sense that the world is built around the PCs. It is noteworthy that games with strong “build” structures also tend to be the ones that are most railroady in execution….some even suggest that Finding the Path is the point in their names!

    One must also take the effort to create a character into account in a game system, because the greater the effort, inevitably, the more the system encourages the GM to protect that character from death. After all, if it takes hours to craft a character, having that character die means that you are out of the game for those hours. In other games, you can be ready to go in only a few minutes, meaning that your survival really is more on you than on “plot” (read: character generation effort) protection.

    All in all, I will take 3d6 in order. Working within limitations has a far greater chance of spurring creativity for most people, IMHO and IME, than working without limitations at all. Making yet another clone is easy. Creativity is hard. If there is an easy way, most people choose it, most of the time.

    YMMV, and, if it does, play what you want. In fact, play what you want regardless! Just don’t imagine that this article actually refutes the previous post.

    • I do understand it and don’t misrepresent it. In fact, I talk right to the subject in the Enter the Madness section. You are not creating a “real” person. You are creating something you want to play. Letting the dice decide that for you can be an interesting exercise, but it’s a lazy mechanic.

      As for the assertion that it creates clone characters, I’ve never seen that happen with giving players choice. When I moved the Living Greyhawk campaign from magic item certs to an adventure record system, a similar argument was the hue and cry of many detractors, and it never manifested. In fact, the opposite flourished. Diversity in character builds and how they were played greatly increased. Pathfinder Society is currently (if not historically) the largest organized play campaign going, and we are in the midst of a large play survey concerning character builds. While the survey is not over, and we have to rigorously categorize the results, we are seeing that build diversity is even greater than what we thought it would be.

      So, in short, I’ve seen this argument in many forms, but it’s obviously counterintuitive at best, small-sample anecdotal at its mean, and overly nostalgic at worst. Given that my friend’s argument is mostly about how he feels on the subject and not the history of the game or a look at anything approaching the greater stage of actual play, I think is very telling. I do think it’s a generational thing. You are more likely to think that rolling ability scores is the way to go if you’re e a gamer of a certain age (as I am), but younger generations of gamers tend not to truck with that…hmmm…outdated convention.

      • I think you are absolutely correct on one particular point. I think the stat generation method you choose is very much a generational thing. Obviously, that’s a generalisation, but I mainly play with people my own age, and I don’t think we’ve used point buy in any of the games we’ve played. I’m not sure anyone has ever suggested we use it (but I may be wrong).

        Most of us are thus “old farts” and used to doing it a certain way. Even in my own group, though, I don’t demand they roll 4d6 in order and keep the scores where they land, although a player is welcome to do that if they desire. (One of the downsides of this method is that it’s quite hard to come up with a balanced party this way.)

        I did very much like this link a reader posted in the comments only original post: http://www.dandwiki.com/wiki/Point_Buy,_Generate_Random

        It’s a great mix of point buy and randomness which I thought was rather spiffing.

      • AFAICT, the original argument doesn’t require you to be creating a “real” person….Which is exactly what I mean. You may have an argument about why you prefer point-buy, but your argument doesn’t seem to be addressing Creighton’s argument.

        Just went back and re-read Creighton’s argument to be certain. Whatever you are countering, it is something else.

  2. Well said throughout!

    The ‘point’ of point-buy rules are just to give every character an equal shot at being uber enough to have some awesome aspects without so many as to be Chet Awesomelaser and break the system. It promotes party building by encouraging different characters to have differing strengths. It simplifies character generation so people can get on with it to make a character in a straightforward manner and start playing the game. And you can make the character you want to play.

    Is it necessary for every game table? Not at all – that would be like saying every table has to use the encumbrance rules straight out of the book. Or that all Paladins must be Lawful Good. Etc. etc.

    Personally, at my table, everybody rolls two sets of stats (4d6, reroll 1’s, take highest three dice, arrange as desired) and chooses which one they want to use. Having to take stats in order without rearranging would wound me as deeply at having to take disadvantage on my CON save to avoid Sewer Plague from the bite of a giant rat in a swamp.

  3. There are good arguments on both sides. I just recently had an interesting experience I want to share. I built a homebrew post-apocalyptic game, originally with random stat generation and random distribution of powers/ mutations. I have, in other genres within the same homebrew rules construct, allowed for strictly point-build. For a specific campaign I was running, I moved away from the random stats, mainly because it made it very difficult to plan the campaign with what was likely to be wildly differing “power levels” in terms of stats.

    Once that was established, I set a maximum point value the characters could have to start, and a portion of my players decided to roll randomly for powers, and then build characters around what they got for powers, and then flesh everything else out, up to the max point value, with skills of their choosing. A number of players really enjoyed this as they had no real concept of a character in mind before playing.

    On the other hand, at least one of my players had a fully fleshed-out concept in his head before beginning, so I allowed him to fully point-build, no random powers at all. He squeezed every bit of utility out of his “build” that he could and enjoyed the challenge. Later when the game got underway, it was clear one player’s character was unable to do the job the player wanted for that character because the character’s powers and statistics just weren’t up to the task, so my “build” player offered some suggestions and I allowed the previously-randomly-rolled character to change out some abilities to be better able to do what was designed.

    That player seemed fairly pleased with the changes, but one of my players, who had rolled all powers randomly, was somewhat offended by the idea of changing up the character in play that had been previously rolled randomly (though the character had never used any of the changed abilities as it was still very early in the campaign). Granted, he ended up with a pretty solid set of powers when rolled randomly, so he had little to complain about on his character. But the other player was quickly becoming frustrated. So, it seemed that the players who had little to no concept of their character in mind before starting were fine with what they got and used that to inspire the character concept, which they were then largely pleased with, while players that had a specific concept in mind preferred to fully point-build their characters while rolling for little or nothing.

    Since all the characters had the same total point value (due to being able to use excess character points to buy up skills and stats once all rolled powers were “paid for”), the overall power level was balanced, but there was still a little tweaking required to make them balanced in terms of utility. Either way, though, they were balanced enough to make the campaign challenging but not impossible.

    So, in summary, random rolling seems to be okay for players with no concept in mind for their characters, especially if there are some things that can be added to make them fully fit whatever role the player ends up seeing for them, while players with a solid concept in mind are going to be unhappy with rolling anything randomly, almost by definition. Different strokes for different folks. While Pathfinder and its kind are fully embracing point build, games like Dungeon Crawl Classics, with an entirely different philosophy of character design (check it out if you haven’t ever looked at it- the “character funnel” concept is a hoot), remain quite popular in their own circles. So, it seems as though there is no “One True Way”!

  4. I like point buy. The only time I ever started a campaign rolling for stats I got a character who’s highest stat was 11 while the Rogue had a 19, two 18s and his lowest was a 16 (I remember calculating if he had a point buy it would require like 72 points).
    All well and good to play a more realistic low fantasy campaign. But all players should be on equal grounds and I don’t see the appeal of my character not being able to do anything while my friend gets to do all the social, exploration and combat.

  5. There’s a lot of nostalgia for dice rolling ability generation lately. It’s right up there with OSRIC style adventures with the black and white line style of artwork, as older gamers view their past experience with the rose-tinted glasses of age.

    I’m not exactly a spring chicken at 56. I’m not one of the super grognards, I started relatively late in 1980…. that’s 37 years ago this summer. I also remember when people started leaving D+D in droves because of it’s rigid systems and the other games such as GURPS and HERO and Champions that ran on a point buy system. Granted some of them were crazy such as how people could game the system with flaws to pad their point budget for abilities and powers. But none of that equaled the insane amount of rampant cheating I saw when people were rolling their 4d6 drop the lowest, even given that variant’s affect on the 3d6 curve. Not that I can entirely blame them… because attributes gave you shit unless you had them at 15 or higher, and some classes had insane levels of requirements for entry. And of course there was the 10 percent experience bonus for having high prime attributes as well. So the incentive, the urge to cheat was very common. In fact some DM’s would abet or urge this cheating themselves because they wanted robust characters for them to throw abuse at.

    I will never ever again run a campaign on anything other than point-buy.

    • We started gaming around the same time, and our experiences are the same. I think the last time I GMed a game using random character generation was maybe… 1987? There might have been a brief period between the time I first saw GURPS and the time I was all-in for point-buy, but it wasn’t very long.

      One interesting thing I’ve observed with versions of point-buy — if anything, the more striking tendency is for a significant number of point-buy characters to have bland stats because they don’t want to be bad at anything. (This may be because none of the systems we use look anything like D&D.) A pair of friends came up with a homebrew system that gave you a baseline — with four stats, you had to be great in one, terrible at another, and average for the other two. I’ve been excited about this ever since learning it, though I admit I don’t have enough long-term experience with it yet to know if it is really better or not.

  6. I find rolling stats to be more fun. The 3e PHB method being the one I prefer. (4d6, order as you wish, reroll if no score of 13+ or total modifiers 0 or less.

    I don’t need high stats to play a hero.

  7. Wait! People still play RP games with stats? Joking.

    As a GM, I prefer using point buy to make the group more playable and balanced. While I may not have a large number years participating that many people do, I believe I can beat most for hours. I have played with grognard groups, kid groups (usually the most fun), and everything in between, but I feel most point buy players enjoy their characters more and rarely hear grumbles about stats. Whereas in a groups using point buy, people tend to play their crap stat characters more brashly, and often with a death wish, while a character with good stats by the same player is careful.

    I like rolling stats, but 3d6 is just the worst. 2d6+5 is the only way to go…

  8. My favored method is still 27-25-23:

    Roll 3d6. Subtract from 27. These are two of your ability scores*.
    Roll 3d6. Subtract from 25. These are two of your ability scores*.
    Roll 3d6. Subtract from 23. These are two of your ability scores*.
    * Reroll any time you’d end up outside the range of 3..18.
    Add +2 to any one score that is 16 or lower.
    Arrange to taste.

    You will always have an ability score total of 77. You have about an 80% chance of a 17 or 18 (if you roll 4d6c3 you actually have a slightly lower chance of getting one of these values). The scores are slightly better than the common 4d6c3, but don’t have nearly the same variation in total value. I find they tend to feel much more organic, I’ve never had a set of rolls that felt boring (I once rolled all 14s and 15s for a character, and it was not at all interesting).

    It also has the happy benefit — for how I was doing things — of having three odd and three even scores. At the time I started using 27-25-23 I had modified the ability score advancements to giving +2 to a single ability score on levels 2, 6, 10, 14, 18, and +1 to all ability scores on levels 4, 8, 12, 16, 20 (and no enhancement bonuses or wish-based bumps).

  9. I don’t think that randomizing stats is always the way to go, but as others have said, it hoghlights the odea that your character isn’t the center of the universe. There are other ways to do it, like small pool point buy, but it works.
    You could roll stats, then fill some sort pf luck pool based on how low your stats end up, this giving low stat characters equal power in the game, albeit flavored as some sort of divine favor.

    The main issue I’d like to address is that “fun” is not synonymous with ease. In high level game design, as for video and board games, it has been found that fun is arises from challenge. The less the world is catered to you, the more fun it is to succeed. RPGs have an escapist element, but that is not all they are about.

    Immersion comes not from escaping from our world, but investing in another one. Now, not everyone seeks immersion, but a game that bends itself to the players tends to be more casual on the front end. It does RPGs a disservice to lower the emotional investment required to play them. Characters are on a spectrum between avatars of us (or others) in the reality of the game world, and simple playing pieces. For me, anything that makes a character more of a playing piece is to be avoided.

    Still, the RPG world has moved to point buy. I’d be interested to see other options, like trait auctions similar to Amber, or some sort of role “drafting” mechanic, where there is some stategy, but indetermnate outcomes.

    • I agree in principle… but I have to admit that one greatly appealing element of Swords & Wizardry Light is the very quick character generation. Roll six stats. As long as you’ve got at least one that is 15 or higher (which is worth +1) you’re good to go. Assign, pick a race and a class, grab some gear, and go.

      Complex or specialized ability score generation… I can appreciate the attraction, but sometimes it’s nice to just get on with it.

  10. The problem with ‘Min-Maxing’ isn’t so much a flaw of a point buy system, but a flaw on the part of a player. Regardless of what system of generating a character is used, a min-ax player will still min-max to the extend they are allowed. Rolled a 9 strength and an 18 Wisdom? Well, so much for that Barbarian idea; the player is going to switch to a Cleric (or Druid depending on the Con roll). A player whose mindset is geared towards optimization will always optimize so much as the rules allow. It’s not about creating a character to inhabit for a Min-Maxer, it’s about getting a massive bonus to something that will allow them to be a star in some area, almost constantly. The way to solve this is to figure out ways in the story to allow character to have the spotlight without focusing on optimization of some aspect.

  11. Interesting note. At least to me. Way back in the day, 70’s something TSR put out what was a “Rogues Gallery”. In the very back we’re some NPC’s described in some detail. Such names as as Mordenkainen, Tenser and Robilar just to name a few. ALL of them had quite high ability scores despite this being the time of 3d6 take them where they land?