I’ve engaged in a couple of discussions recently about designing games to eliminate powergaming or character optimization.

My personal opinion is that you can’t: Any system featuring 2+ choices in order to create or advance a character in which those choices have a mechanical impact is going to have people using those choices to optimize their characters. (Note: Equipment choices count.)

This becomes even more true if the content of the campaign can be varied: As a I discussed in “Fetishizing Balance” many moons ago, a character loaded up with seafaring skills is going to be non-optimized for a campaign based around Robin Hood, but is absolutely going to rock it in a campaign based around Treasure Island.

When game designers try to solve the “problem” of character optimization they generally end up falling into the fallacy I describe in “Ivory Tower Design“: The belief that you can allow for meaningful choice in any kind of complex system without having some choices be inferior to other choices. And the result is that they limit either (a) the flexibility of character creation, (b) the scope of the game, or (c) both.

(Which isn’t to say that game designers should just completely ignore issues of balance: Correct obvious imbalances between options aiming at the same effect. Remove anything that radically alters the fundamental experience or nature of the game in unintended ways. But trying to make Blackbeard just as effective in Sherwood Forest as he is on the high seas? Or making fighters specialized in a bow just as effective in melee as the fighter who specialized in a longsword? You’re wasting your time and your efforts will probably cause a lot of collateral damage.)

GMs vs. OPTIMIZATION

So we’ve concluded that all RPGs are going to feature CharOp and we’ve also concluded that there’s very little or nothing that game designers can do about it. But what if you’re a GM faced with optimization-obsessed players and you don’t like the effect it’s having on the game?

First, recognize that there’s nothing about optimizing your character to be good at X that’s incompatible with enjoying a good story. In fact, a lot of character optimization originates as a completely logical, in-character motivation: I need to be good at X to succeed/stay alive, so how can I be really good at X?

Where this becomes problematic is when it turns into an optimization arms race: The PCs keep getting better at X, the GM cranks up the difficulty of X to “keep things challenging”, and the players respond by making their characters even better at X. Rinse. Wash. Repeat.

When there’s a clash of expectations or problematic behavior at a gaming table, I generally recommend solving those problems by sitting down and just talking out the issues involved. But the optimization arms race is a slightly different kettle of fish because it originates in-character. A conversation about expectations may still be useful (particularly if the behavior has become reflex or habit), but you have to solve the in-character motivation first.

(1) Diversify your game. If X is the only thing that a character ever has to do in order to succeed, then the character is highly motivated to optimize X. If players have to do a bunch of different stuff in your game, then they won’t be able to hyper-optimize one facet of their characters.

(2) Worry a lot less about “challenging” the PCs at whatever the PCs are badasses at. For example, if one of your players creates a badass hacker, you shouldn’t necessarily respond by cranking up the difficulty of every computer system they want to hack. If you do that, they’ll respond by trying to figure out how to crank up their hacking skills a few more notches.

In other words, just opt out of the arms race. And this remains true even if the results seem extreme: Have they created a character that can hack the NSA on a whim? That’s OK. Cyrano de Bergerac was the greatest swordsman who ever lived. Edmond Rostand didn’t negate that or ignore it, but he also didn’t respond to it by creating an elite brigade of soldiers who could match Cyrano in a swordfight. (Think about it.)

Sometimes, of course, you’ll want to throw a hundred swordsmen at Cyrano. Or have the NSA contact the PC because they’ve seen what she can do and they need some help cracking the encryption on an alien artifact that the Martian Rover just found.

What I’m saying, though, is that you should spend less time trying to one-up your players and more time thinking about the interesting stories that result from the consequences of success rather than the risk of defeat.

Last year I posted a thought of the day concerning “Disarming Magical Traps”. Although this particular thought of the day was most specifically and immediately prompted by Numenera, it’s also something of a sequel to that earlier thought and equally applicable to magic items in a typical fantasy campaign.

As I mentioned in that earlier thought, I think it’s important that a GM not allow any interaction at the table to become purely mechanical. Partly this is just an aesthetic preference on my part (it keeps things interesting), partly it’s ideological (rules are associated for a reason), and partly it’s because specificity and detail usually leads to creative gameplay.

In Numenera, as the title suggests, a great emphasis is put upon the leftovers of the older (and almost incomprehensible) civilizations that predated the Ninth World: “The devices, the vast machine complexes, the altered landscapes, the changes wrought upon living creatures by ancient energies, the invisible nano-spirits hovering in then air in clouds called the Iron Wind, the information transmitted into the so-called datasphere, and the remnants of visitors from other dimensions and alien planets—they call these things the numenera.”

Particular significance is given to the cyphers: Single use items that function according to forgotten sciences, proliferate throughout the Ninth World, and (from a purely mechanical perspective) serve to constantly vary and refresh the abilities available to PCs.

Before cyphers can be used, however, they have to be identified. Mechanically speaking this is straightforward: The character attempting to identify the cypher attempts an Intellect task with a low difficulty. If they succeed, they now know what the device does and how it can be used.

It can be difficult, however, to visualize or describe how this process of identification takes place: Consider the example of a pill you can swallow which will then allow you to teleport to any location you can clearly visualize with your mind. There’s no self-evident way to “experiment” with the pill short of swallowing it; and if you do that, it would be consumed and gone. So what does the mechanical resolution of the identification task look like in the game world?

Ultimately, it’s a combination of lore and/or limited experimentation. In the case of the teleporter pill, for example, options might include:

• There’s a sigil on the side of the pill. You recognize that as a sign associated with teleportation in other artifacts that you’ve used.
• Or you cross-reference the sigil using a lore book and discover that the armies of Salla Izirul once discovered a cache containing millions of these pills and used them to teleport entire legions behind defensive lines before his supply ran out.
• You scrape a little bit off the side of the pill and feed it to a mouse. The mouse vanishes and reappears next to a piece of cheese on the far side of the room. Guess it’s a teleporter.
• Yes, it’s a pill. But there’s a small metallic nodule attached to one end of it. And, yup, that’s definitely a transdimensional regulator. The only reason you’d be swallowing one of those is if this thing was going to teleport you.
• You’re a nano and you’re using a low-powered Scan to determine the energy signatures locked inside the physical matrix of the “pill”.
• The pill is actually encoded with a psychic memetic mesh. If you can just manipulate the articulated junctures of the pill correctly, it will basically download an instruction manual into your brain.

And so forth. My point is that there’s not a single or precise method that the character can use to identify the object, so you should feel free to get creative.

FAILURE

 Conversely, a failure on the roll might indicate that:

• There is no way to identify this featureless pill: All they can do is swallow it and see what happens.
• They’ve misidentified what the pill does.
• They’ve misidentified how powerful it is.

 PARTIAL SUCCESS

A partial success (perhaps succeeding at a difficulty one-half the required difficulty to fully identify the object) might yield some useful information:

• How to activate the device, but not necessarily what the device will do. (This would obviously be more applicable to cyphers that aren’t self-evidently pills. Although maybe this “pill” only works if it’s a suppository or stuck up your nose or surgically placed under the skin or ground up into the user’s eye.)
• A general sense of what the item does, but not its specific function. (“It has something to do with non-Euclidean travel” instead of specifically indicating that “it will teleport you X distance”.

My point with all this is that you shouldn’t be afraid to discover (or define) features of the world as the world is being explored.

 ARTIFACTS

More powerful technologies of the old world in Numenera are referred to as artifacts. These devices can be used multiple times (although there’s a risk that any given usage will be the object’s last) and it’s far more likely that the PCs are just using one of the many possible utilities the original device had. (The other functions may be inexplicable, irrelevant in this dimensional space, broken, only intermittently available, more likely to deplete the device, or dangerous to the user.)

On page 299 of Numenera you’ll find a really fantastic random chart for determining random quirks for an artifact: I recommend taking it to heart.

Roger the GS over at Roles, Rules, & Rolls posted some interesting thoughts regarding the use of the techniques I described in Xandering the Dungeon in small, one-shot scenarios. This, in turn, prompted me to ruminate on the application of xandering techniques on small scales.

Xandering isn’t a cure-all. But, in my experience, it does scale to almost any size and it’s almost always useful to at least consider xandering as a potential tool even if you ultimately decide against it. (I might even go so far as to say that you should default to it unless you have a really good reason not to. In no small part because, as I mentioned in the original essay, this is actually the way the real world works 99 times out of 100.)

To demonstrate what I mean about using xandering techniques at any scale, let me give you an example at an extremely small scale to emphasize the point: A two-room “dungeon” that I just got done designing for an Eclipse Phase scenario.

The “dungeon” in this case is actually a warehouse: The first room is a small security office. The second room is the big warehouse floor itself. Since it’s only two rooms, there’s really no way that we could apply xandering techniques, right?

(Spoilers: That’s a rhetorical question.)

Let’s take a look at a few xandering techniques:

First, multiple entrances: Skylight(s) on the roof of the warehouse. The loading dock. A door leading into the security office. (From a tactical standpoint, this is infinitely more interesting than just having a single door leading into the building.)

Second, multiple paths: Rather than just having one connector between the security office and the warehouse, what if we include several? There’s the door. A ladder leading to a trapdoor in the roof that gives you access to the skylights. Let’s toss in a trapdoor leading to a crawlspace that’s used for electrical wiring; it’ll let you pop up right in the middle of the warehouse (or maybe in multiple places). (If that crawlspace is actually a tunnel that leads over to the exterior generator we could also add that as yet another entrance to the complex.)

That crawlspace would also qualify as a secret or unusual path (another of our xandering techniques).

This obviously isn’t the only way to design a warehouse. (It might even be overkill.) But it does demonstrate how you can use xandering techniques even on the tiniest scales can organically create interesting tactical and strategic choices.

Back to Xandering the Dungeon

Seeing the development of the whole “friend-zone” concept is, in fact, enlightening about the pervasive misogyny that’s still culturally foundational in America despite decades of progress.

It started as an observation that once someone had placed you in the “friend zone” of their mind, it was difficult for them to consider a romantic relationship with you.

It then picked up negative connotations when it was applied to women who flirtatiously imply the potential of a future relationship in order to have men perform favors for them that they would not do for normal friends. This sort of thing probably wouldn’t exist if it wasn’t for the pervasive cultural assumption that it’s the man’s role in society to earn money and, therefore, the way to woo a female mate is to throw money at her in the form of gifts and so forth. But up to this point the term was at least describing an actual thing that actually happens.

But then the wheels come off the bus, because in the lightning-fast memetic chamber of the internet the term continued to expand: Now it was any woman who politely said “no” when you asked her out on a date. But, of course, the negative ethical connotations stuck to the term — so now the entire concept of “friend-zoning” implies that any woman who says “no” to a man’s sexual advances is doing something ethically wrong.

This also simultaneously expands the other side of the term: It now applies to any man who is friends with a woman. But here, too, the negative connotations stuck to the term. As a result, it implies that “just” being friends with a woman is somehow a punishment or a failure.

This rapid progression from useful concept to misogynist ideology is all built around the lingering cultural scaffolding in which women are objects of desire which are pursued like treasure. Although this scaffolding is slowly being demolished, it’s both interesting and depressing to note (from the sufficiently safe distance of being a white male) that, like any construction site, this transitional period can actually be more vile and misogynistic in some ways than what came before: Leave intact the “pursuit of the virgin” but strip away the idea of “no sex before marriage” and you replace Lord Wessex from Shakespeare in Love with pick-up artists who treat women like Super Mario Bros. power-ups and their sexual resumes like a Call of Duty leaderboard. Leave intact the idea of “no sex before marriage” as a moral imperative, on the other hand, and you end up with all women being “whores”. The jagged edges of these half-forgotten cultural memes can be dangerous. (Which doesn’t mean, of course, that we shouldn’t be getting rid of them. That would be like arguing that the slaves shouldn’t have been emancipated because they were more vulnerable to lynchings without the protection of their owners. It just means that you have to anticipate that it will be hard work and a tough slog before the light at the end of the tunnel completely banishes the darkness behind.)

A fellow named Caleb asked me in an e-mail recently why I’m not a fan of the way GUMSHOE handles clues. In writing a reply to him, I think I’ve found a better way of expressing my personal distaste for GUMSHOE’s approach than I have in the past.

Start by considering a scenario with locations A, B, and C.

First, let’s assume that each of these locations contains a clue which points to the next location. GUMSHOE says, “Oh no! What if they don’t find a clue? Then the adventure can’t continue!” And in order to solve this problem, GUMSHOE says, “It’s OK. We’ll just remove the resolution mechanic and we’ll simply assume that the PCs succeed.”

Investigative scenarios have been done wrong since the early days of roleplaying games. As a consequence, they’re hard to run and prone to grind to a halt. (…) You have to search for the clue that takes you to the next scene. If you roll well, you get the clue. If not, you don’t — and the story grinds to a halt. (…) GUMSHOE, therefore, makes the finding of clues all but automatic, as long as you get to the right place in the story and have the right ability. (Esoterrorists, pg. 26-27)

In other words, we’ll remove the chokepoint of failure by simply removing the possibility of failure.

So what’s the problem?

Well, now let’s assume that each of these locations contains a monster which you have to fight before you move to the next location.

Presented with this problem, we would expect GUMSHOE to say something like, “Oh no! What if they don’t defeat the monster? Then the adventure can’t continue!”

And in order to solve this problem, GUMSHOE would then say, “Well, that’s OK. We’ll just remove the combat system and we’ll simply assume that the PCs always defeat their foes.”

To be fair, GUMSHOE is right: If you make it so that the PCs automatically win, then they will never lose. It’s tautological and everything. And is there anything wrong with that?

Not necessarily: If the game wasn’t actually about fighting people, there might be little harm in skipping past the fights. But if the game was about combat, then you might have a problem.

And, in my opinion, the actual act of investigation is, in fact, a relatively major component of what a mystery story is about. GUMSHOE says it isn’t because you never see a fictional detective miss a clue. (But if they did miss a clue completely and entirely, how would the reader or viewer ever know? And, in point of fact, there are many mystery stories in which the detective does miss a clue and later goes back to find it or realizes that they missed it only after the crime has already been solved.)

In addition to this, as I’ve discussed in the past, GUMSHOE’s “solution” doesn’t actually solve the problem it claims to be solving: Failing to find a clue is only ONE of the ways in which the clue can fail. Since the problem hasn’t actually been solved, you still need to implement the ACTUAL solution to the problem (which is to not design your adventure around chokepoints in the first place). And once you’ve implemented the actual solution, you’ll discover that characters failing to find any particular clue is no longer any sort of problem… which means that the GUMSHOE “solution” isn’t required at all.