Mini-maxing

[jdagna]

It's my general opinion that min-maxing in a well-designed system is a natural and healthy thing to do.  In a poorly-designed system, you can run into a lot of problems.  Some are just plain design errors, others stem from incoherence.  In a homebrew, I'm betting that part of the problem lies in poor design.

Problems with min/maxing in good systems seem to fall into two categories:
1) G/S conflicts
This is a case where Simulationist goals get subverted by Gamist elements in character design.  The conflict goal may have Drifted into a coherent design or be present in a hybrid Sim/Gam system.  I probably don't need to say any more, since others have done this justice. The Gamist players drive the story through sheer force while the Sim folks scream about "roll-playing".  Generally, I suggest that people find a game with either a more coherent character creation mechanic though it can sometimes be solved at the social contract level.  This seems to be the biggest problem with min/maxing in D&D.  At this point, it's essentially a Gamist design, but many continue to strive for Sim play out of it.
2) Player/GM conflicts
If one player wants to blow stuff up while everyone else wants a cloack and dagger spy mystery, the group already has problems.    However, it'll sometimes show up when the first player builds a combat monster who can blow up anything and is worthless at anything else.  In GURPS, he'll usually take "disadvantages" that require his character to fight at the drop of a hat and reinvest the points in combat abilities.  Either way, the GM is now forced to either add some more combat in, or ignore the odd player and design for the majority... in which case the combat monster stomps all over every clandestine meeting with his +2 minigun of ignorance.  Either way, he's forced the group to cater to his tastes with a character that needs and creates combat situations.  People often blame this problem on min/maxing or on the system, when the reality is a conflict of interests at the social level.

[M. J. Young]

If it's not supposed to be part of the game, why is it there?

I had been thinking I had at least part of the answer when I saw that

What if it is the rules that allow the players to feel cool and clever?

E. R. Jones had a variant in his character creation system in D&D. He had people roll 3d6, and then reroll any die that rolled a one or a two. I quickly observed that it would be more efficient and achieve statistically identical results to roll 3d4+6--the curve is identical. He actually became angry that I would suggest this as an alternative to his method. He didn't care whether the results were identical. What mattered was that letting the players pick up those dice and throw them again until they got at least 3 on each gave the players the feeling that they could get a better character out of the system.

I think that systems are designed to permit min/maxing specifically so that players feel like they have more control over what kind of character they produce--they are not as much at the mercy of the system during character generation. If they want to make the combat brute for survival's sake, or the superintelligent character for its special abilities, they can.

Perhaps the telling feature of min/maxing, though, is the degree to which specialists are better than generalists. This may be a very important aspect in system design which is frequently overlooked. We know that the superintelligent magic-user is better than the average magic-user, and the musclebound fighter better than the average fighter, at that at which they are specialized. Class systems tend to encourage specialization, because they limit the degree to which you can benefit from generalizing. That is, in earlier editions of D&D, if you want to be both magic-user and fighter, you either have to go for one of the hybridized classes (Paladin, Ranger, Wu Jen) or you have to be a multi-classed non-human. That means there is very little advantage for a fighter to be smart, and if you can trade away smart, wise, and charismatic for strong, tough, and dextrous, you're in better shape.

The way to break min/maxing in design is to create a system in which the benefits of being a generalist are at least equal to those of being a specialist.

You still will get some min/maxing if the game has a party play emphasis. Statistically, it's probably better to have five guys with a weak chance to pick a lock than one guy with a strong chance to do so, but it will always feel to the players like they're better off with one great thief, one great fighter, and so on, rather than five guys who are average at everything. On the other hand, if you can gear the game such that player characters have the impression that each character is going to have to be independent of the others, and survive and prosper based solely on his own abilities, you again push the advantages toward the generalist.

Maybe that helps?

--M. J. Young

[Piers]

Maybe it is useful to distinguish between the effects of Minmaxing on the relation with a) other players, and b) the GM.  

In the first case, the major problem seems to be the potential deprotagonization of the other players.  They feel less 'cool and clever' that the minmaxed character.  

In the second case, but also maybe as a whole, minmaxing causes a breakdown in consensus as to what is _possible_ in game.  That's to say, minmaxing causes disagreements about what the players can and should be doing.  The Vincent Baker suggest (with the 'lumpley principle') that the rules are simply a means of achieving consensus about the game world, but in the case of the minmaxer the gameworld he or she sees described is very different from that seen by the rest of the group.  This can be as much use of the system as minmaxing in character generation.

Personally, I find that often there is a large disparity between what many systems imply about the gameworld and what the GM and Players think they imply.  It often takes someone who has paid a great deal of attention to the rules to rub their faces in the fact.  

The fact that both of these influences are often in action simultaneously complicates things.  In trying to understand what is going on, pulling them apart can be very useful.

[Jack Aidley]

To my mind it is entirely credible for a system to say: here's the rules, we know they can be min-maxed to make stupid characters, we designed it that way because we don't think it's up to us to place arbitary limits on what you can create, or play silly arms races with the power games. No. Instead we leave it up to the intelligence of the players and the GM to create characters that make sense.

[Jack Spencer Jr]

I think we're devolving a bit into finger pointing a bit. It's the stupid designers for not knowing what the heck they're doing in designing a game. It's the stupid players who make characters that don't make sense.

I'm not the sort who thinks that no one is to blame, can't we all be brother, yaddah, yaddah. I'm thinking everybody is to blame, at least when it doesn't work.

[John Kim]

To my mind it is entirely credible for a system to say: here's the rules, we know they can be min-maxed to make stupid characters, we designed it that way because we don't think it's up to us to place arbitary limits on what you can create, or play silly arms races with the power games. No. Instead we leave it up to the intelligence of the players and the GM to create characters that make sense.

OK, this corresponds to some degree to what I said -- namely that making the rules abuse-proof is more trouble than its worth.  However, this implies that mini-maxed characters are stupid and don't make sense. I completely disagree with this.  Mini-maxed characters are more powerful than naively-generated characters, but I think that's about all you can generalize about them.  Most of my PCs have been pretty heavily mini-maxed, but they also have had interesting personalities.  

The problems that I have seen with mini-maxing have usually been with new or non-rules-saavy players.  I have tended to solve this by helping those players mini-max their characters.  If their concept isn't mini-max-able, then I'll request or make rules changes to make it so.  On the other hand, people have different feelings on this.

[Ben Lehman]

To my mind it is entirely credible for a system to say: here's the rules, we know they can be min-maxed to make stupid characters, we designed it that way because we don't think it's up to us to place arbitary limits on what you can create, or play silly arms races with the power games. No. Instead we leave it up to the intelligence of the players and the GM to create characters that make sense.

BL>  I apologize for my tone here.  Please don't think that I'm angry with you -- rather, this is an idea I often hear repeated and it is very frustrating.

I think that this method, blaming the players for using perfectly reasonable methods of increasing their credibility, is fatally flawed and results in criminalizing both those who listen to it and those that ignore it: Those that heed this message with most game systems will build characters who, quite frankly, do not function; whereas those that ignore it and minimax are villified for doing so.  This is a horrible Catch-22, and I think the common presence of such statements in RPG books amounts to an Impossible Thing Before Breakfast.

The only possible way for this sort of statement to make sense is in a very heavy simulationist mode, in which we're saying "these are 'real people' and thus should be made however the 'are,' rather than however they are most effective."  However, this discounts basic human behaviour in-game.  If we are, in fact, in a very heavy simulationist mode, the rules of the game are the rules of reality, and characters should be expected to be reasonably intellegent and not make dumb personal choices.

I think that that was a little unclear, so I give an example:

(this is stolen from somewhere on RPG.net, but I honestly cannot remember where.)

Imagine a system where it is cheaper to buy skill with related "weapon types."  For instance, skill with a Battle Axe makes skill with Hand and Throwing Axes cheaper.  Now, imagine a warrior character under this system.  The clear minimax choice is to use all the same type of weapon (sword, dagger, throwing knife) rather than a mix (axe, mace, dagger, sling).  Some people might say that such a character is blatantly minimaxed, and that character is.  But, from an in-world perspective (remember, heavy sim here), any seasoned warrior would notice that his comrades that use similar weapons are much more likely to survive than those that use a variety.  People would pick up on this rather quickly (warriors are good at that sort of intuitive jump) and, pretty soon, everyone would use all straight blades, or all axes, or the like.

In short, that stance essentially says "players, and their characters, must behave like morons, otherwise they are ROLLplaying and not ROLEplaying."

This is lame.

yrs--
--Ben

[Marco]

I think there's at least some confusion on my part as to what min-maxing means.

If it means a warrior specializes in one (or maybe 2) weapon skills--I don't think that's what people are talking about.

If it means exploting breakpoints religiously, well, maybe that's getting closer (what's the extra point for an INT of 14 *really* worth in Hero?).

If it means building characters designed to exploit the rules in such a way as to create logical contradictions (the GURPS character who sold DX to buy a higher weapon skill) then I think we're dead on.

Min-Maxing is not only a reasonable one in zero-sum games like the Stock Market or Black Jack it's the only one anyone'd expect you to use. No one takes a hit on a 19 to give the dealer a sporting chance.

But what does it really mean in RPG's--what's the gradient.

I think Ben's example is perfectly logical. I think some character concepts I've seen (champions characters who sold BOD to buy up defenses with 1 of the ponts and STN with the other pt) are in the more extreme category.

And while I'm not 100% convinced that a game system shouldn't at least consider edge conditions, I definitely fall in the camp of the GM and Players having a say as to who they game with--which really takes care of all the nonsense right there.

-Marco

[jdagna]

Marco, min/maxing is simply the technique of maximizing character effectiveness using the available (limited) resources.  It gets its name because most systems reward characters who specialize in certain aspects while lowering their abilities in other aspects (the dumb fighter, the weak wizard, etc) - raising one ability to its maximum by reducing another to its minimum.

Thus, all of your examples are min/maxing at work.

[Ben Lehman]

I think there's at least some confusion on my part as to what min-maxing means.

If it means a warrior specializes in one (or maybe 2) weapon skills--I don't think that's what people are talking about.

If it means exploting breakpoints religiously, well, maybe that's getting closer (what's the extra point for an INT of 14 *really* worth in Hero?).

If it means building characters designed to exploit the rules in such a way as to create logical contradictions (the GURPS character who sold DX to buy a higher weapon skill) then I think we're dead on.

BL>  All of the above are minimaxing.
As for the last, I don't see how that is any different.  GURPS, apparently (I don't play it much myself) is a system in which the most effective warrios is highly trained and clumsy.  Thus, GURPS models worlds in which the most effective warriors are highly trained and clumsy.

If this is not a feature that you want in your worlds, change your system.  But it is important to point out that this is the system's fault, not anyone at the table.

And while I'm not 100% convinced that a game system shouldn't at least consider edge conditions

BL>  I think here we have the real root of all the ranting about minimaxing -- it is a cover for poorly designed games.  Blame it on the players -- not the designer who made a system which makes no sense and is mathematically nonfunctional.

yrs--
--Ben

[Jack Spencer Jr]

After thought, I have decided I don't agree with Mr Jack.

Instead we leave it up to the intelligence of the players and the GM to create characters that make sense.

What "makes sense?" If it's in the rules, then fine. If not, then I tend to side with Ben, it's a fault in the rules.

I would temper Ben's blame a bit, perhaps focusing again on the purpose of the Forge. Since we all hope to design games, some for commercial release, we need to be mindful of the asumptions we are making about our audience, and also learn to listen when we get it back or else we'll just make Greedo shoot first.

[Jack Aidley]

Hmm, I think I should clarify my position a bit. You probably still won't agree with it, but hey. To my mind things suchs as the GURPs Dx problem are broken, the system rewards the wrong strategies. That IS a problem. Similarly I consider currency and breakpoint issues to be a problem.

However, suppose I have a skill-based system (I tend not to use stat+skill, just having skills) and I give the players 40 pts to spend on skills. I expect the players to create characters that are well rounded and make sense regarding their backgrounds. I certainly don't expect to see a character with: Sword 30, Health 10. But at the same time I'm not going to create a complex set of rules saying exactly what is and isn't allowed. I firmly believe creating characters should be an intelligent debate between the player and the GM, not just on stats/classes/whatever but also on background and personality.

[Fletcher]

I will respectfully attempt to give some of my thoughts on min/maxing with the help of this community.

A few examples, definitions, anecdotes of min/maxing have just been given above which have illuminated to myself that this one of those nebulous things that all of us are intimate with while role-playing until actually pressed to give a textual definition. One might even say that all these explanations are at once both correct and incorrect. Indeed, I cannot be more succinct than the next person so I can only offer to dialog in the realm of what my favorite philosophical term calls "intellectual honesty"—if you can see that I am making a point within your own thought process then go with it; otherwise I am sorry.

I dare say this "min/maxing" business goes part and parcel with this gamist approach to role-playing. And if this is so then min/maxing can only be bad when there is general incoherence in the group or when the system fails somewhere in its design for gamism.

Now, I'm afraid that the above paragraph is all I can say about the GNS matter. I must spend a lot more time on Ron Edwards' essays and the GNS specific forums to better wrap my grey matter around the whole point because with all of my lurking I cannot have but the most basic intuitive grasp of the matter. If I do have a point though, and I have not posited a heinous thread drift then I would much appreciate someone else to continue on with the point.

To add something else to the discussion though, I would like to offer up an intuitive example from my own experiences.

There seems to be a problem in games where people can play significantly different types of characters or archetypes. The problem arises when one player identifies with a particular archetype while the proverbial min/maxer takes another archetype and through a fault in the rules, or sure numerical genius proves that his character can do much more than the proper archetype.

The example I will give is from back in my AD&D 2nd edition days, not so long ago. I have a passion for thief type characters so I selected one of those. It then came as a great surprise to me then when the wizard turned out to be a better thief than me, walking around invisible and casting "knock" compared to my hide in shadows and lock picking skills respectively. This obviously plays to a bigger issue with the game, but one ought not to succeed better in one path over another when one is more strongly supported by its archetype.

I genuinely think that the problem with this part of min/maxing and gamism is that if a class based system is to satisfy pure gamist goals then there should not be archetypes associated with those classes.
As a general implication of this anecdote I offer that the one of the perceived problems with min/maxers is that they seem to toss away flavor in favor of effectiveness. This will frankly grate on at least someone's nerves.

If I were to "tweak" my character in a play group comprised solely of gamists then no one should have a problem, right? However, if that tweaking were to turn my character into something that looked just plain ridiculous then I might have a quarrel and even be called a min/maxer. And since "system does matter" then this ought to be blamed on the game system. Correct me if I am wrong.

I would certainly not endeavor to say that all min/maxing is a problem with system, but I will pose these two closing points:

1. At least some aspects of min/maxing are a problem with system—with gamist systems in particular. Furthermore, no game designer should dismiss the possible repercussions of min/maxing in their system with out some careful examination.

2. Min/maxing is a term and a topic that is a major part of role-playing. It should not be looked at as a singular thing, but rather a fuzzy category or some other anomaly of our undiscovered cognitive processes at large. At any rate, it certainly deserves some good examination and I am happy to see it being discussed here!

[Ben Lehman]

Hmm, I think I should clarify my position a bit. You probably still won't agree with it, but hey. To my mind things suchs as the GURPs Dx problem are broken, the system rewards the wrong strategies. That IS a problem. Similarly I consider currency and breakpoint issues to be a problem.

BL>  I imagine I won't :-)  I would like to point out that I don't think these are "wrong --" I just think it's important to look, quite clearly, at what is effective in a system.

To hold up a positive example:  Riddle of Steel is a highly minimaxable system.  However, the best possible minimax is to have an interesting dynamic character with powerful passions, drives, etc.  Thus, you get a game with a lot of interesting, dynamic characters, with powerful passions, drives, etc.

The minimax is a universal part of human behaviour -- we make minimax decisions every day.  The only reasons that one wouldn't minimax in an RPG context is:
1) If you thought you could get more power through a different method (see my post earlier in this thread.)
2) If the necessary exploration of system, and the system search and handling time, is not a goal in and of itself and is not worth the added ability to explore other areas (character, setting, etc.) or dilutes said exploration.
3) If the social credit for being "non-minimaxed" is worth a less effective character.

It is this last one which I have an issue with.  The social contract in gaming groups is strained enough without it relying on it to fix system holes.

However, suppose I have a skill-based system (I tend not to use stat+skill, just having skills) and I give the players 40 pts to spend on skills. I expect the players to create characters that are well rounded and make sense regarding their backgrounds. I certainly don't expect to see a character with: Sword 30, Health 10. But at the same time I'm not going to create a complex set of rules saying exactly what is and isn't allowed.

BL>  At first glance, I see that this is not a very well thought out system.  Of course, I don't know how anything else in the system works, so I can only make uninformed judgements.
 But it does but the players in a "Catch-22" situation -- Make the most effective character possible whilst appearing not to care about effectiveness.  Your unwillingness to put down your rules -- in writing -- strikes me as willing bad game design.
 What I read this system as is "Here are 40 points.  Spend them in an effective manner, but there are point caps, and I'm not telling you what they are, but if you cross them, I won't tell you not to, I'll just be passive-aggressively angry at you, as will the rest of our friends, and the game will be No Fun For Anyone."

But, of course...

 It really depends on what the game is and what the system is like.
 If this is a combat game and players are whacking each other with swords, this guy has made a feasible character, most likely the only feasible character given your system.  If we're on an open field stabbing each other, it doesn't really matter -- by which I mean effect play significantly -- that I used to be a champion gardner.  What matters is who puts the point end into the other man.
 If this is a combat game where the point is that people shouldn't be heavily combat trained, you should put a point cap on combat skills.
 If this is an Amber style game with a small number of highly important Attributes, the player has made a very weak, but perfectly servicable character.  You might want to talk to them about minimaxing MORE.
 If this is a social intrigue game, the problem is that the player hasn't minimaxed enough for the scenario in question.  In fact, they have made a highly ineffective character (what you might call a miniminned character).  You need to talk to them about minimaxing more.
 And, if you're complaining that the player didn't waste his precious credibility on things that will never come up in game, then I side strongly with the player on this one.  Player credibility is hard to come by in most RPGs, anyway.

I firmly believe creating characters should be an intelligent debate between the player and the GM, not just on stats/classes/whatever but also on background and personality.

BL>  So do I -- in fact, I have an impulse to do away with chargen systems entirely in my games (I usually only use it with new systems.)  This is wonderfully shocking to D&D players when I tell them:
"Just write down whatever stats, skills and feats you think are appropriate.  Be around 3-5th level.  I'll let you know if something is wrong."
 I'm all for making chargen an intellegent debate.
 But the second half of your sentence worries me.
 Say that, in D&D, I make a brutally effective fighter.  That doesn't mean that he doesn't have an interesting life, personality, friends, and background.  It just means that he's an effective fighter -- brutally so.  The "background and personality" stuff need not be systemed.  Some of the greatest characters I have ever seen played were minimaxed up the wazoo.

But say you have a system that expects that I spend some of my CP spent in areas where it is FAR less likely to come up than others (for instance, in a typical RPG, I'm going to get FAR more mileage out of "Sword" than "Flower Arranging."  In HeartQuest, it's the other way around...), there needs to be a counterbalance or the system is broken.  If I say to my players: "This game is all about flower arranging" I expect them all to make the best flower arrangers possible within the system.  If I want to see points elsewhere I will do something to encourage that, via point caps, a seperate pool of "non-flower arranging related points" or what have you.  If the game also involves detective work, I will relate to the players that detective skills also might be important.
 In your examples, you have essentially done this, but through social contract rather than rules.  This, I believe, puts undue stress on the social contract, and means that anyone who makes a slightly more effective character risks ostracization and ridicule.

I believe your attempts are good hearted -- you are trying to fix the point build system, which is not an easy thing to do -- but look at the ramifications.

yrs--
--Ben

[Jack Aidley]

I'm sorry to say, Ben, I'm struggling to understand your point of view. You seem to be viewing this all in a way very different from my experiences of roleplaying in general and GMing in particular.

There's no passive-aggressive anger at folks for transgressing hidden boundaries here. There's intelligent discussion. As I see it there are two primary models of character creation. Build first-concept later and concept first-build later. I nearly always use concept first - build later. The player comes up with a character concept. Then the player and I discuss and develop the rules-side choices that best represent the character they have described. There's no maximising effectiveness to be found, only best representing the character concept.

If a player wants to play a rock-hard fighter. Cool. We'll make a rock hard fighter, but we will also think about and model the aspects of the character that aren't rock-hard fighter. Maybe they were a farmer once, maybe the grew up by the seaside, maybe they know a little about animal care and handling. Probably they know some equipment care. And so on, and so forth. Obviously this means there has to be enough points to make a rock-hard fighter and still have some spare for flavour.

I disagree that I should have a fixed skill cap. In my experience skill caps simply mean that everyone builds characters who have skills up to the cap. This isn't what I want. It also removes the possibility of having the good turnip-grower, the very good turnip-grower and the exceptional turnip-grower. I also don't want to allocate points to different areas. I don't want to decide up-front that 40% (say) of their points must be spent on flavour. What if the character concept calls for more? Or less? Better, I feel, to solve this through discussion and debate based on the individual character concept.

To address your D&D example; I'm not saying that this is the only viable method only that it is a perfectly coherant and valid method. D&D is a system with a lot currency and breakpoint issues; its also a system that tends towards big emphasis on solving problems with stats and spells. My games simply don't work like that.

I'm not sure what you mean by Player Credibility; could you expand on that, please?