Point-buy Play and Design

Point-buy is a common mechanical conceit in tabletop roleplaying games. It’s pretty omnipresent in traditional games in minor ways, plus there’s also a hard core of games that elevate the point-buy to a central principle, funneling everything in the game through the points-counting in one way or another.

This is all pretty bizarre, and I don’t think it’s at all a given that a hobbyist understands what games like e.g. GURPS, the crisis of infinite superhero games (“crisis” is the collective noun for superhero products, in case you didn’t know), Sorcerer and whatnot are even trying to do. Let’s see if I can shed any light on the phenomenon.

Defining the concept

Let’s start with a definition, people who read theory articles like this seem to like those. Point-buy used to be pretty common, and most old-timers probably know what it is without having me explain, but better safe than sorry – people have great capability for recycling terminology, so for all I know “point-buy” means micro-transactions at your local table, with the GM selling bonus dice at $ a pop.

Technically speaking point-buy as a game mechanic is simply the extended general case of multiple option choice: instead of making just one pick off a list of several choices you have X picks, with each individual choice costing Y picks to make. Maybe every option costs the same – one pick – as a simplified case, like e.g. in Otherkind mechanics, or maybe you have a full-blown point-buy situation with every option carefully priced for its inherent attractiveness. From this perspective point-buy is quite common in roleplaying games, you don’t need to play GURPS to encounter the basic mechanic. D&D has had point-buy as an ability allocation option since the early ’80s.

Traditional issues

Point-buy is interesting because of how popular it is in tabletop rpgs despite the fundamental issues the structure involves. I’ll mention a few core design issues that are outright ordinary for games that elect to use point-buy, you can see for yourself how peculiar an institution it can be:

Optimization is encouraged but not taught: Point-buy inherently encourages the player to make optimal choices (that’s sort of the point here, giving the player room to make their own choices), which in itself does not need to be a problem, but what is an issue is that the game and the player often have vastly different understanding of what is desirable in character creation, and how to achieve it. A common consequence is a “min-maxing” heuristic where the player recognizes a single character trait as desirable and uses the point-buy freedom to spend all points in that one area. The results may range from powerful-but-completely-unrealistic to laughably dysfunctional depending on the game and the specific min-max; either way, the point-buy experience is not very satisfactory because the player obviated their opportunity to make complex trade-offs in favour of identifying a single concern – their “max” – to pursue. This can be deadly for a game that would actually prefer to have the characters be thoughtful, serious, perhaps realistic, perhaps dramatically interesting. Besides, in many games min-maxing actually produces weak characters; the player tried to optimize, but failed because this simple decision-making heuristic led them astray.

Random access interface chokes up: Any game where you have to make point-buys off a long list has the potential to be a boring piece of trash simply because the amount of work involved in making an informed choice off a list grows linearly with the length of the list itself. To pick between 20 options in an informed way, you have to read through 20 options. If the importance of the choice is not in proportion to the number of available options, the player is forced to obviate the spirit of the point-buy by making a random choice, or bore themselves to death poring over the options. Some very common games have major issues with this, e.g. modern D&D: picking spells and feats.

Point-pricing is arbitrary: This should be obvious, but the specific history of traditional tabletop roleplaying has been rather successful about pretending that roleplaying games have some inherent universal balance that justifies the point-buy pricing in whichever game as anything more than arbitrary. This is not necessarily a problem, depending on what you’re actually trying to accomplish in a given game, but for games that pretend the point-buy to represent some sort of strategic balance or spotlight distribution between player characters or whatnot it’s a reality that you’re building on feet of clay by pretending that the point values do that thing you think they do.

Flaws are externally motivated: Your typical point-buy game allows the players to balance their points-budget by taking on “flaws”, character qualities that are considered undesirable (in some arbitrary way that lives in the trad gaming burrow and may or may not be intelligible to a sunlight person). You get more points to buy desirable qualities by accepting undesirable ones, and “undesirable” can be anything ranging from “has a spouse” to “doesn’t have legs”, really – when I say that the concept of “flaw” is quite arbitrary here I’m not kidding. The problem here relates to the arbitrary pricing issue in that you’re either paying the player to do something they would have done anyway, or giving them an external reward for accepting an intrinsic penalty, which is the exact opposite of what you want in a creative exercise: you shouldn’t train players to accept a flaw that they don’t want in exchange for some ultimately meaningless external reward. The relationship you get between a player and the flaws they pick is often a minimizing one: the player tries to pick flaws that don’t matter, so as to get points without having to deal with the flaw in actual play. A random fobia unlikely to ever trigger, anyone?

Most point-buys are dull: This is something of an emergent consequence of the fact that point-buy is a generalized version of choice-making. While a game designer or GM is perfectly able of providing an interesting specific choice that provokes thought and entertains the player, generalizing the choice matrix makes it geometrically more difficult to maintain the excitement simply because the number of potentially cross-conflicting values represented by the choice grows so large that the player no longer recognizes the choice’s tactical or thematic significance. It definitely does not help that in dedicated point-buy games a more universalized choice matrix is usually further detached from the game’s ostensible subject matter to better enable the generalization of the point-buy: the outcome is that something that could have been a half dozen exciting choices about the game’s fiction becomes a spreadsheet about some very non-fictional, generic mechanical choices. If you will, compare (early) D&D’s unpretentiously fictional and sharp “do you want to be a Fighter or a Wizard?” with the much more diffuse GURPS equivalent “which of these 200 special abilities would you like to have, and to what extent?” The latter simply engages the actual subject matter of the game much less, which can be a serious problem in building up to a functional creative activity. It does not let you fall in love with the idea of playing an elf, because it never gives you the choice, preferring to subsume that choice into generic “you can be anything you want!” rhetoric. Freedom without structure.

This is not to say that point-buy is only problematic, but I do think that in the big picture it’s something to be leery about. I’d prefer it if making it the central feature of a game was truly special, a choice justified by an unique necessity. I don’t think that the traditional use cases like superhero games, generic systems and trad skill list hellholes qualify as a matter of course: none of those topics ipso facto justifies the use of point-buy.

The Royal Conundrum

I reserved one particular problem of point-buy for separate treatment because I consider it both the most subtle and most significant of the potential problems that a point-buy game faces. This is the royal conundrum of point-buy:

The royal conundrum cuts to the core of the critical design issue when considering a point-buy game design: considering the potential problems that point-buy has, as outlined above, what justifies the use of point-buy in your game? Why does it exist? What would happen if you simply… didn’t have it? Maybe there is something that point-buy actually does for you. Even if there is, could there exist a solution with less overhead?

I personally believe that there are many roleplaying games that fail to answer the royal conundrum satisfactorily, which essentially means that many point-buy games are badly designed: they do not justify the use of a rather specialized, relatively clumsy game mechanic. Some examples of games that as far as I understand it prominently suffer from this issue:

Call of Cthulhu: Or any other BRP game, or any of their multitude of descendants, really. The conceit in this type of “skill list + points” game is that you define a character’s skills by distributing points between skills off a long, long skill list. There’s typically about a hundred skills. Now, CoC is an atmospheric horror game with a heavy adventuring bias (the PCs form a party that goes on quests), and the whole skill point distribution rigamarole has nothing whatsoever to do with the core activities of any kind of CoC that I have ever seen. Even the practical consequences of having whatever kind of skill distribution in the party are minimal, as can be ascertained by the fact that CoC is the classic grandfather of railroad gaming: it simply doesn’t affect anything substantial in the big picture whether you picked up some specific skill or not. The character creation that otherwise focuses on being simple and character-oriented screeches to a halt when you grant the player their several hundred points and tell them to start slogging through the skill list to distribute points that are only marginally meaningful later on during play. There is no creative value in deciding whether your character should have “Driving” at 50% or 70%, yet the game expects the players to make literally hundreds of choices of this kind to create one pesky Investigator. As far as I understand CoC, it fails the royal conundrum completely and thoroughly, and would be a superior game if it produced these skill lists (or better yet, abandoned them) in some other way.

Modern D&D: Charop games like this feature a particularly vile institution in the form of a supplement treadmill that the players use as part of their character creation efforts. The more expansive games might feature literally dozens of books, all of them full of lists of alternate character elements such as spells and feats, most entirely accessible to any character who really wants to have them. The so-called system-mastery is a meaningless trainspotting exercise in poring over these lists and picking character elements that synergize with each other. If this is what you like to do with your spare time, all well and good, but do remember that this is a game that claims to be adventurous, fun, exciting and intended for casual gamers who simply aren’t that into massive point-buy optimization exercises. D&D only succeeds in answering the royal conundrum insofar as it decides to throw out the initial premise of being an adventure game: the point-buy is entirely justified if the game you play is about creating and testing out character builds in the skirmish combat system.

GURPS: What you think of when thinking of point-buy games is of course an ’80s generic point-buy system like GURPS or Hero System or one of their many copycats. The point-buy in these games is a bit more justified in that the game revolves so thoroughly around its point-counting aesthetics that this really is the game. However, even with these games I have to admit that the why, the royal conundrum, escapes me: what, exactly, does the point-buy get you in GURPS? It’s not the “universal generic” appellation, certainly? I mean, the fact that GURPS has stats for swords and laser pistols both doesn’t derive from the point-buy, it could have those things just as easily without it. It can’t be balancing effectiveness between characters, as the point-buy has such a wide scope that it’s very easy to spend your points more or less effectively. It can’t be strategizing the spending choices, as you create characters before any kind of hypothetical wargaming scenario is presented – and GURPS certainly doesn’t smack of being a challenge-based game in the first place. I do have my own opinions about what GURPS (or Hero System) is good for, but for our purposes here it suffices to say that the game text of either thoroughly fails in offering a clear justification for the elaborate point-buy.

Many, many roleplaying games are in some ways similar to the above three, and therefore may have difficulties justifying their point-buy schemes. Food for though: what is the answer to the royal conudrum for your favourite point-buy game? Would the game be any worse off if it didn’t have the point-buy?

A few remarkable point-buy games

Next, I want to draw your attention to a couple of the most exceptional point-buy games that I am aware of. This isn’t a full review, I’ll just explain why the following games are noteworthy targets of study for anybody interested in understanding the full picture here.

Amber Diceless is an early ’90s point-buy game with some seriously traditional structures in that regard: there’s a GM and a bunch of character players, with the latter being granted a hundred character build points to design their characters. These points are then expended in the main through an innovative attribute auction that bids the players against each other in determining who gets the most desirable character qualities. The remainder of the points are then spent on more traditional buys from a limited list of character qualities like various kinds of magic and such that you might desire your character to have. There are even a few ways to gain more points, as you’d expect of a point-buy system.

Amber has a few remarkable qualities for what qualifies as a point-buy game, rather than a game that merely uses point-buy. The most important one beyond doubt is that the players do not have to spend the whole points budget, and they can go to the negatives if they want to: this remainder of their character design effort is then stored as good or bad karma for the character, to be used during play in various content framing and resolution procedures.

The second noteworthy point-buy quality of Amber’s is that it features a remarkably constrained and carefully considered set of character options that are available for purchase with the points. It’s the very opposite of the generalized power lists that you’d see in a typical point-buy game like GURPS. All the options have immediate significance for the game’s fiction, so there’s none of the “effect first” thinking that you usually get in point-buy games. The outcome of the PvP bidding affair and the carefully fiction-tailored point-buy options is that the character creation system, properly applied, oozes charisma.

In the interest of full exposure I’ll note that Amber fails to answer the royal conundrum in an explicit way, as do most point-buy games, but that’s mainly because it suffers from what I consider a creative agenda incoherence in the game text: if you know that your Amber campaign is a ruthless PvP game of thrones (one of the implied purposes of the game), or that it’s a fantasy family drama (about equally likely, really), then the point-buy will justify itself one way or another, with either the attribute bidding or the karma system having pride of place. The game’s something of a moving target in this regard, requiring specific work to nail it down.

Sorcerer is an equally meritous rpg from the late-’90s with some considerable point-buy elements to its design. The point-buy is not at its core like it is with Amber, which is in itself remarkable for how Champions-inspired the game is in many ways.

The important point-buy in Sorcerer resides in its extensive demon-creation subsystem. The overall game is about morally ambiguous sorcerers with the capability of summoning and commanding supernatural entities. The player gets to outright design a demon at character creation, and should the character summon more demons during play, the GM designs them using the same rules. The point-buy is much more discrete than in many dedicated point-buy games, with the demon-building budget ranging from a couple of points to about a dozen, but point-buy it nevertheless is, with an explicit list of power options and even “effect first” thinking.

What Sorcerer does right in its point-buy – and this largely helps the game avoid the pitfalls I discussed earlier – is that it strives to construe the demon design in terms of a simple list of “what can this demon do, exactly”. The points aren’t actually a currency so much as they are a limit to the length of the list: you can get X different extraordinary things that this demon can do. The power options are not priced, they all cost one pick. Each pick represents a single dramatic empowerment, a single plot twist where the demon reveals one more way in which it breaks the rules (when compared to a human, animal, object, whatever their physical chassis in this natural world is).

Sorcerer also, remarkably, answers the royal conundrum: the reason for why it uses a point-buy is that it’s important for the demons summoned in the game to align to the power of the summoning ritual, and therefore there needs to be a distinct budget constraining the demon’s design. It’s interesting and necessary that the player character’s actions caused a 3-Power demon specifically and not something else. Furthermore, because generating the demonic mythology occurs in dialogue between the entire group, rather than it coming from some sort of pre-existing monster manual, it’s necessary for the GM to have a “language of design” for mechanically defining all kinds of wildly unexpected supernatural critters with the rules. The system is point-buy, but it’s very strongly “fiction first” in that you first have a circumstance and inspiration for bringing in a new demon, and only then you go into the power lists; you don’t pore over the list (which is very boring in itself) like a kid in a candy store, picking up cool powers, so much as you go in with an idea for what you want to buy, and then identify the abilities that enable the demon to do what you envision it as doing.

Recommended solutions

All right, the is the actual gist of this article: I wanted to share a few ideas that I find important to successful use of point-buy, and successful point-buy games in particular. The above is largely just background for the following observations that are pretty simple, but also pretty specific to this field of study. My points are very much applicable to most pre-existing point-buy games, so don’t hesitate to hack them to fix the most egregious issues!

Use flexible budgeting: Like Amber, above, you should allow players to both not spend available points, and to spend in deficit. The remaining balance is a very obvious mechanical hook that can be used for all kinds of interesting purposes, whatever aligns with the wider goals of the game. In Amber a character’s “good/bad stuff”, their outlying points balance, is used as a generic luck attribute, for example; in my own Solar System deficit-purchased character traits carry a penalty die until paid for. The truly important advantage of flexible budgeting, however, is that it frees your point-buy character build scheme from the tyranny of the spending pivot: you can now modify characters with all kinds of advantages and disadvantages when the fiction suggests doing so, instead of having to wait for the game to allocate you more character points (usually through a XP-like reward scheme).

Use fiction-first purchase logic: Whether your game does “effect first” or not, it’s no excuse for allowing the creative focus of the players to descend into a degenerate, speculative study of lengthy power lists. There are various techniques for avoiding this, such as a “concept first” approach, where you first figure out what you want, and then search the available options for something that fulfills your idea. In character development context I recommend that most games adopt a fiction-based precedence rule: to purchase power X, attain power X in the fiction of the game. To learn the feat, learn the feat. This principle moves attention away from the frankly secondary rulebooks and into the actual events of play, where it belongs; you need to train players to play the game, not the speculative white-room exercise of poring over options that should never have been presented as “being in the game” in the first place.

Use structured purchase procedures: Get away from massive point-buy systems and emphasize multiple-choice situations. Having a player choose which of three mysterious wizards their character picks as their mentor is much better than having them spend 200 character build points for various magical powers; it’s better because the human mind finds it more meaningful and exciting to compare three package deals with their various advantages and disadvantages, than a long list of arbitrary options. In Hero System or GURPS we would say “rely on templates”, for instance: develop your campaign’s specific character creation so that you expect the players to spend like 80% of their points budget by purchasing fictionally flavourful, significant package deals, with a few points left over for personalization (or left unspent, as per earlier advice). The players should actually rarely interact directly with the raw, “stupid” interface of the power list itself. A generic hint here: you can turn typical point-buy chargen into something much more powerful with some well-chosen lifepath systems quite easily; not a trivial amount of grunt-work, but if you love your game you’ll want it to be the best it can be.

Use dramatically activated flaws: The old Champions-style “accept a disadvantage to get more points to purchase an advantage” thing is almost universal in point-buy games, even the otherwise excellent Amber has a particularly dysfunctional form of that. There exists a better way: construe character flaws as relatively cheap or even free character features that, when relevant in play, grant the player with a reward (like more points, this being a point-buy game). By moving the reward away from the initial choice to accept the disadvantage we help players to not claim flaws unless they want to play them; the flaw will only grant you external rewards when it’s used in play, which instead of rewarding the player for avoiding the flaw encourages them to feature it in play where otherwise appropriate. Tell me this isn’t strictly superior to the old-style flaw mechanic.

Treat the point-buy chassis as a work in progress: Rules-as-written attitudes do not work well with the most expansive point-buys for many different reasons. The complexity of modular interactions is too great for the game to constantly work well, and the contextual effects of having the game be played by this particular group, over this particular subject matter, mean that getting whatever it is that you care about out of the point-buy system will probably require constant fine-tuning. Embrace that! Change the point values when you realize that something is under-priced or over-priced; maybe settle the difference with players who already have bought that option, maybe not, depends on the overall game which is better. Hold seminars with the other players, let them decide how much things should cost and whether some options should be available or not; it’s their game, too. Reward players for writing up new fun options; you’re playing a point-buy game, so presumably you like having a lot of options.

Use the point-buy through the game: Some point-buy games suffer from front-loading the point-buy: you have this complex universal balancing chassis (or whatever it is that you imagine the point-buy boil to be doing for your game), and you have to teach the players to use it, but then after you finish character creation (what, something like the first two or three sessions of this exciting game?) it’s discarded and hardly used anymore later on. A big reason, of course, is that the point-buy may be inherently irrelevant to the ostensible topic of the game; if it’s an adventure game where the player characters aren’t generally reconstructed for every adventure, and the GM doesn’t use the points at all because he just puts whatever he wants into play, then of course you don’t really interact with the points-counting system much. Maybe you should develop the game further towards a “let’s build things!” concept where the game activities naturally come back to the point-buy rules again and again? Sorcerer, for example, knows where it’s at: the game is about summoning demons, so you can bet your best athame that the demon-construction subsystem will come into play regularly.

Elevate the points as a formal currency: I’m getting into royal conundrum territory here, but it’s a fact that in many point-buy games there is an implicit underlying secret unspoken reason for the obsessive points-counting: the game is supposed to formally measure significance of a thing by its points value: this 100 point mecha is important to our space opera story, sure, but this here 300 point mecha is even more important. In the most generic terms, without commenting upon the game’s creative goals, this “importance” means: what do we pay attention to during the game? We ignore the things that do not cost points, and the more points something costs, the more of our attention it should grab. What the points are spent on supposedly helps the game mechanics reflect this underlying importance of attention, but there is no guarantee that a game actually succeeds. When designing (or fixing) a point-buy game, make it so that the points align with importance: otherwise the fact that more points equals more busywork in budgeting doesn’t make sense. Grant points to things based on their importance, and have rules that prioritize the things with more points over those with less.

Answer the royal conundrum: Even with everything above, I am still sceptical and need to be convinced: why are we counting points here when we could instead be establishing game fiction? The points-budget and the various point-prices, and the economy they form, are not inherently useful procedures for most kinds of tabletop roleplaying games. They are not framing procedures that help in introducing content and situations; they are not resolution procedures that help action move forward; what does knowing that Strength score of 14 is equivalent to 5 character points get you in this game, exactly? Answer the question (I think there are answers, but they’re frankly somewhat esoteric and even surprising, and very specific to individual games), and I believe that your game will improve simply because you can direct your efforts better instead of doing spreadsheet math on a complex point-buy just because the rulebook tells you to.

Of course there are special cases and exceptions, but by and large I think that the above suggestions are strictly improvements on how point-buy is used in tabletop rpgs. I’m interested in the exceptions, but to me it seems that these best practices are usually better than their alternatives, and therefore something that point-buy games should bring into wide-spread use without delay.

OK, so maybe I have to try to answer the royal conundrum

I was initially planning to simply not answer the royal conundrum for any single game at all, but I guess it’s a bit weird to write an article about the challenges and solutions of point-buy design without even trying to answer the core question. That question being, why does GURPS (or Hero System, equivalently) have that remarkable point buy chassis? Even if we agree that having generalized power lists is great fun and everybody should be doing that (questionable), why do you have to price those powers out carefully and build characters by budgeting the things you want? Why can’t you just give the character the powers they should have (as per your character concept)? Presumably you’re following some sort of build plan, after all, so you already know what you want the character to have. What do you gain with the pricing and budgeting?

An answer that I would characterize as naïve would be to say that you have to have a points budget or otherwise the players will create characters that are too powerful for the planned campaign. Or what if one player creates a character that’s more powerful than the rest? This basic answer is faulty, I think, because it does not rise from a place of artistic strength: it presumes that the group cannot coordinate and restrain themselves without having a rulebook on the table (for the GM to hide behind, presumably) reminding them to practice moderation. Adults (and why would be presume that children would play something like GURPS) should be perfectly capable of coordinating the creation of a bunch of appropriate heroes for their particular flavour of adventure story without pricing their ideas in terms of points. It is possible in many other roleplaying games, after all, and those aren’t played by remarkably wiser people as far as I know.

I’m no veteran of GURPS or Hero System (which doesn’t mean unfamiliarity with point-buy, because the frigging thing is everywhere), but I did formulate a bit of a theory about GURPS after playing a handful of sessions a couple of years ago. I noticed that the game’s most delightful aspect was the way it attempted to marry the formal character description (stats, skills, etc.) with organically reality-based adventure situations. You might expect a massive point-buy game to simply say that you have X points invested in a Jumping skill, and therefore if you encounter a pit with X points of width, you can jump over the pit. But no, pits don’t get points in GURPS (unlike e.g. vehicles, which certainly do), pits get meters of wideness, and you get this cute little formula for calculating how far your character can jump based on their various characteristics.

(Hero System is a bit similar, by the way, but generally slightly more “cartoony”. It does have jumping distance calculations, too, but I wouldn’t personally be very surprised if I encountered a pit trap statted up with X points of wideness in that game. At least I can say off the top of my head that you could calculate the value of “wideness” in points by figuring out how much jumping strength it takes to triumph against so much wideness; superior amounts of points should have the advantage.)

Aside from that, though, the points-counting business… I don’t know, I think that the pricing and budgeting is just a metastasized cyst, a remainder of when this stuff had some relevance early in the game’s development history. What I see in GURPS is the attention-significance phenomenon I mentioned earlier as an underlying feature of point-buy systems: having points implies mechanical complexity (more powers) and more agency, more “power”. Maybe, just maybe having players create characters on identical budgets helps ensure that those players have similar amounts of agency in the game. But of course this is another naïve illusion, any experienced gamer knows it to be a truth that your agency as a player is, and always shall be, a feature of how you play the game. GURPS cannot give you agency by giving you points.

I should note that I’m very on board with the formal mechanization agenda that GURPS has: it wants me to “build characters” by defining their statistics and skills and powers, using pre-existing building blocks. Combine that with some setting that I’m interested in, and I’ll be there with bells on. For example, I would find it entertaining to stat up the Fellowship of the Ring from LotR in GURPS, and put them through their paces against some adventurous challenges. The more of Middle-Earth stuff we’d build for the campaign, the more interesting it would be to use the various characters to benchmark each other and produce an usable playset. Maybe something more primal would work even better, like I don’t know, He-Man and the Masters of the Universe. We could debate how many Strength points He-Man should have for hours. But where does the points budgeting enter into this?

I imagine a GURPS campaign where there are no preset budgets, and perhaps we wouldn’t even care about pricing (how much this or that costs). Does that not work in some way? If it does work, then does that mean that there is no answer to the royal conundrum and GURPS would be better off without the pricing and the budgets?