My first attempt at game-shepherding was actually a PC. I’ve alluded to it in my riff on my history as a traitor PC, but it did serve as the purest, simplest form of shepherding—my mission, as the player of a blackmailed character and the only one aside from the GM who’d played in the system before, was to get the group to an exact location on an exact day, and ‘make a scene’. It wasn’t hard. (Granted, what ended up happening was the other PCs making a scene, and mine having a crisis of conscience and making trouble. This is why I like free will.)

My best game shepherds were probably the string I used when I was running my primary Exalted game. They made something of a sequence of their own, all of them older and wiser (okay, one could question that last with Kes) than the group and thus serving as sort of mentor figures as well as guides. First was Kiara, a knowledge-spec leading a bunch of combat-primaries, who served as a local guide, Voice of What The Heck Is Going on Here, and general metaphysical adviser until she and the group were rather abruptly and unexpectedly separated. Kes filled the role at that point; originally she’d been Kiara’s more physically-oriented backup, which meant that while she could fight as well as the group, she was in mostly the same boat with them when it came to actually coming up with plans. In other words, while she could provide common knowledge, she couldn’t think for the group the way Kiara had (and, in fact, would respond to attempts to get her to with a mournful “I’m a demon hunter, not a….”). Kes, in turn, was joined and later replaced by Amaya, who filled a similar role mechanics-wise but instead of being a voice of exposition, that role having been taken over by Luath when he wasn’t botching hilariously, was more a voice of practicality and paranoia. After the first arc—it got a tad messier; all of the above were easily available for advice, but the most regular on-site NPC was genki girl, walking transportation system and one-woman army Ruby, whose main shepherding role seems to be “make sure it’s feasible they can get where they’re going, and remind people about the magic I planned for them to use if they get stuck”. All of the above tend to fade to the background when the group is active and stick out when their compatriots are stuck—or sometimes to run off and do awesome things somewhere else, depending on the situation.

After the first game-shepherd PC incident, I took on the mantle as something of an “I can’t think of a better character concept, is there anything you need?” approach. This led to interesting results, including my discovery that maybe volunteering for that position when the rest of the PCs had been coerced into their roles was a bad idea. This occasionally even worked for long-term characters, as when I played Tuyet; in that case, I was a sort of shepherd/muse/representative of the other players, suggesting or occasionally playing ways to ensure that the scenes he considered absolutely vital came to pass in some form or other, and pointing out when something might look like a good idea on paper but really screw with a character arc if not implemented carefully. (The GM in question remains the only one to whom I have ever given point-by-point explanations both of how to remove my character from a fight and how best to mess with her mind. He had a bit more luck implementing the first than the second.) What it took here was a subtle touch, a good understanding of the other characters and the other players, and a willingness to be an apparently neutral sounding board, occasionally even anticipating plot twists, talking them over with the other players, and warning the GM about their ramifications.

Come to think of it, it’s a small wonder I can still play normal PCs.

…okay. Breathe. It’s not that bad. (You probably couldn’t have told my younger self that, though.)

First off, remember that whether you’re an expert or not, if you read something into a story, you’re probably not going to be the only one. Heck, you might run into things that experienced writers (or at least, experienced participants in writing workshops) tend to bounce off of, quietly rationalize and thus don’t really notice. Which is to say: your opinion means something. If something bugs you, interests you, or in any way that you think might be useful to the writer affects you, say something.

Be complimentary to something. Some workshop formats mandate positivity at the beginning, I think I’ve seen some that want comments to stay positive in general, but either way—even if you’ve got something that you’ve used so much red ink on that it looks like the paper itself is bleeding, find something that you liked. The object of the game is to improve the author’s work, not make him/her feel like a complete failure.

If you don’t feel confident correcting the technical stuff, don’t correct the technical stuff. Same goes for meta. If the workshop’s larger than two or three people, someone else is going to be able to handle the in-depth analysis and the nitpicking over typos. Your job is to find the things that scream out to you, both good and ill.

Note as many of the things you interact with as possible. Bear in mind, one of the things a writer is supposed to be doing is getting emotional reactions out of people, so if something pleases you, annoys you, makes you chuckle, throws out your suspension of disbelief, leaves you in that state where it’s like you’ve just spent the last [amount of time you spent reading] being crashed about between waves and rocks and all you want is a moment to dust yourself off before you dive back in—the person you’re critiquing will probably want to know. If there’s one little detail that just screams out at you “too fitting to be false”, by all means mention it. The best part? You don’t even need to think about holding back on mentioning the kinds of negative emotions that follow with the story; we’re trying to elicit those, too (and some of us are much better at drawing forth fear, sadness and rage than joy and optimism).

When in doubt, leave a question. One of the biggest problems that people have is that they know everything about their worlds, and thus they miss the things that in their internal logic makes perfect sense but in our logic are skipping a few vital steps. This even happens to professional authors, and some of them even admit it—there are a number of examples, but the one that springs most readily to mind is Zoe Marriott’s blogged admission (you want the third question down) that she was powered by inspiration alone when ending The Swan Kingdom and forgot to make sure that the logic she used was clear to everyone else, so what to her was the surprising-but-inevitable ending that had been in the works from the beginning seemed thrown-together to many of her readers. If you don’t get what’s going on, it’s probably not just you. If it’s probably not just you, they need to know so it can be a little clearer to people who aren’t the author. The best thing you can do in a situation like that is ask why.

In short, the best way to get from one end of a workshop to the other if you’re feeling overwhelmed is to take three basic steps: read the story, react to the story, and keep close track of your reactions. The rest will take care of itself.

One of them is figuring out just how much control of the game’s direction to give the PC in question. Give them too much, and resentment builds, both in character and out of character. (Thing learned from personal experience: never, EVER give the newly added player who can’t think of a concept the role of the Person Who Holds the Leash if your players are for the moment supernaturally controlled, even if it would save you some trouble. Even if they don’t choose to use it beyond the minimum… they die quickly.) It’s hard to give them too little, if such a thing exists, though you do want to make sure you can get someone who can to some degree influence the game’s direction.

Another issue is how much wiggle room the game shepherd gets. I find the optimal situation is “ensure that this thing happens, but other than that, do what you want and respond how you want”; any stricter parameters and you lose your game shepherd’s player’s sense of agency. This is a useful trick if you’re dealing with a player who likes the idea of collaboration and reacts badly to being full-on railroaded; you can get the nifty scenes you like, and they’re less likely to resent it because there’s still a part of it that is, well, theirs. Besides, then you get the fun of the player creatively interpreting the instructions, or completely turning things around once the objective is met—something you wouldn’t necessarily get if you were dragging the story around yourself. As often as not, these end up being traitor PCs, though they really don’t have to.

One problem with the game shepherd as PC, though, is the potential appearance of favoritism. There are several tricks to avoiding this, though. One is to make the plots for which the shepherd is going to likely be an important directional element center on someone else’s character, so the star of the plotline isn’t the person who is managing your end of the steering. Another is to alternate between people deputized as shepherds, if you think you can manage multiples. If you’re likely to only end up with one (maybe it’s easier for her to get in contact with you outside of session than it is for the others), you might see if she’d be willing to make something of a bridge to the other players, giving you a sense of when your precious plot twist might not be so hot an idea for concept-deep reasons.

Another issue is how much the shepherd knows. If you’re prone to surprising people, it can be very difficult to keep secrets from the person who’s helping increase the odds that your ducks all end up in a row rather than scattered all over the place. In that case, you might want to get someone who really doesn’t care about the surprises—or use the shepherding sparingly and on events that aren’t necessarily going to directly hint at the conclusion you’re hoping for.

They’re tricky, but PCs can cover the shepherding part of being a game shepherd.

Without actually playing the games, you’d think the former of the two would go over better with the audience. Fi’s on Link’s side from the moment she first manifests; Midna’s using her Link from the beginning. Fi manages to keep her opinions (assuming she has any, which itself is somewhat questionable) to herself; Midna throws out the verbal abuse from her first dialogue screen. I can’t stand Fi. (I am by no means whatsoever the only one.) I love Midna to pieces. Since I’m pretty sure it can’t all be explained by my apparent love for characters who are willing to insult mine, what goes?

The biggest issue is the matter of when to exposit and when not to. Most of Fi’s negative reputation comes from the fact that if Link is anywhere but Skyloft and something plot-relevant happens, she will manifest and talk (this webcomic is not an exaggeration. No, not even the part with the bow). Midna isn’t near as likely to pop into a conversation—yes, a good portion of that is not wanting to be seen, but the decreased frequency helps. Most of the complaints I have about Fi boil down to “If I want help, I’ll ask for it.”

It doesn’t help that Fi’s voice isn’t all that interesting. She sounds, put simply, as if she’s dictating a lab report, dispassionate, sesquipedalian, and prone to spouting stats. Yes, she’s an artificial intelligence (no, that’s not a spoiler), but there are a lot of examples of AIs with strong voices in a wide variety of fields (GlaDOS. The tachikomas. Mike from The Moon is a Harsh Mistress. Shall I go on?), so as far as I’m concerned, being a construct is an excuse, not a justification. Midna’s voice was a lot more distinctive and, at least in my opinion, much more interesting; I rather looked forward to her scripted appearances to see what she was going to say next. Her personality shows through; she has fun for fun’s own sake in the process of saving the world, insults whomever she feels like, shows sympathy, and feels, well, alive.

Then there’s the character dynamic—or, in Fi’s case, lack thereof. While I have not finished Skyward Sword, I am more than halfway through and have yet to see any change in the relationship between Link and Fi; she’s in his sword, she’s on his team, she has no life outside of these purposes (again, logical given she’s a construct made to help The Hero, but dead boring). Midna—well, it starts as a partnership of convenience, warms up as each demonstrates competence, and—well, by the time one moves between the first and second chain of go-here-and-find-this, she’s earned the respect scripted into the cutscenes.

Though those are the important parts, sometimes the cosmetic and mechanical details make a difference as well, and this case is no exception. Due to being functionally parts of the hero’s panoply rather than separate characters, neither has to worry particularly about being the Load or ridiculously More Awesome Than You; they’re working with Link, not alongside him. It’s… not hard to cut a little more slack for someone who provides access to certain locations, teleportation, shapeshifting and a useful multitarget attack than for someone who provides simply advice on gear, a bit of navigation help, and a limited-target directional locator. It’s also pretty easy to draw dubious comparisons between a walking example of the Uncanny Valley with next to no facial expression and someone who avoids the Valley entirely by being Most Definitely Not Human (and has nifty eyes. I’m a sucker for eye detail on animated characters). Qualities like this aren’t guaranteed to apply quite as much to one of our games, mind, but they will probably make a difference.

Long story short, I know who I’d rather have as an NPC party member, put-downs and all.

The first problem with the game shepherd is the careful balance she has to maintain between the extremes of being the load and looking like a DMPC. Most people, once the concept of DMPC has been explained to them, tend to err on the side of less powerful, recognizing that the point of these sorts of characters is not to overshadow the PCs themselves, and certainly not to be a walking ego trip to the GM. On the other hand, it’s easy to overcompensate and make them too weak—and while there are ways to make a group like a character who endangers them through her incompetence, it makes for a titanic handicap if you want the character to be liked.

The second is knowing when to exposit and when to shut up, particularly when the situation requires thinking through a problem. Granted, an overly silent NPC can easily come across as somewhat stupid—which would only be more aggravating if the players themselves are completely stuck. On the other hand, answering the question before it’s asked, expositing without call to do so, and expounding on the obvious just don’t go over well. Noting the obvious is particularly obnoxious; if the character isn’t aware that the people she’s talking to already figured out what’s going on, she looks stupid and oblivious, and if she is, she looks condescending. Neither is a good trait for someone the group’s supposed to want to work with.

What about autonomy? Since the game shepherd is going to be guiding the PCs right the way through, she’s going to need to have good reason to stand by them—but on the other hand, if you’re not careful, she might look like she has no existence whatsoever aside from explaining things to the PCs and carefully nudging them in the right direction when necessary.

Then there’s personality. While you’ll want to make sure it’s not the kind of personality that will cause the PCs to kick the game shepherd out of the group through sheer annoyance, there should probably be one—a good game shepherd is one the characters would want to associate with anyway, and who wants to associate with someone who has no personality? Having personality also means you can get distinct, evolving social dynamics between PCs and NPC, which makes the NPC more interesting and as a result more worth keeping around on her own merits.

Game shepherds aren’t impossible to run; if I can pull it off on my first real GMing attempt, it can’t be that hard, right? But knowing what the pitfalls are makes it a lot easier to know how to avoid them.

One: I’ve been doing this regularly for two years and change. At one short essay worth of evil advice per week, that is a LOT of purely academic evil. Add to this the fact that the Management generally avoids things with villains that make fools of themselves and thus are left wide open to serve as inspiration if not outright illustrations, and it gets hard to keep churning these things out. Ideas don’t just pop fully-formed out of minions’ heads, after all! (For one thing, well, minions. For another thing, the few times they do, the medical bills for reconstructing the skull are astronomical.)

Two: Strange though it may sound given my masters and the altruism associated with the job, I am qualified as an advice columnist. I don’t consider you people my rivals; I’m mostly retired and attempting to further the cause of the Dark Powers as filtered by their need to uphold Dramatic Necessity. Short version is, it is in my interest to help you people, not just because it helps the Cause but because it gives me something to write about (see Point One) rather than just throwing out generalities. I even do backlinks, if you want them—you’d be amazed by how intriguing a good plan that just needs a teensy bit of going over can be.

The point of all this is simple: for the love of the Dark Powers, ask me some blasted questions! It’s what I’m here for!

I started, at about the same time, with the conflict and the scope. Very old, very powerful enemy wants to break boundary between two worlds on behalf of his much larger and more incomprehensible patron (hey, my prior GM had done that, I figured if I followed his pacing approximately, we’d be fine). I’d originally expected a prior meeting, a whole lot of running around finding ancient wonders and things to help out with the big fight, and then—behold, our resolving event—a big fight somewhere that wasn’t quite the land of the living and wasn’t quite the land of the dead.

All right, so far so good. Add PCs, stir. It took me five sessions to figure out his name. Two more to figure out what he was doing. Another dozen or so to design and introduce his offspring.

Right through the plotline, we had to deal with the regular question, “Why aren’t we there yet?” Needless to say, the whole time, “He’s too powerful as it stands” was a reason, but never the only one. What I came up with was something along these lines.

Sessions 1-5: We know someone’s out there, but we don’t know who.

Sessions 6-10: We know who’s out there, but we don’t know where he’ll strike next.

Sessions 11-15: We don’t know where he is, and we’re a little busy right now (fighting for our new friends, mitigating some near-fatal damage we’ve done to our reputations, fighting demons…)

Sessions 16-44 or so: We know where he is, but we’re not there yet, and we keep finding things to deal with en route.

Sessions 45-46: We’re there (sort of) but we haven’t gotten in the fortress yet.

Sessions 46ish-48ish: We’re there, our main baddie’s not home, one of our teammates might have switched sides and the main baddie’s subordinates are being reasonable.

Session 49: Forget fighting, we’re outta here! (This one was…. mostly the players’ fault.)

Session 50: DIE DIE DIE DIE DIE! (GM: whew. Finally.)

It kept us busy.

When concocting an answer to this question, I find it very important to make sure that it’s an answer that the characters can take some sort of action to deal with. I’ll admit, it isn’t always possible–“The lunar eclipse required for the ritual isn’t for another three weeks” isn’t exactly a circumstance under anyone’s control, for instance—but in general, and particularly when dealing with PCs, you’re going to want to make sure that the main characters have something to do that feels like it’s making progress.

There is, however, one exception to this rule, and that’s “Nobody’s figured out what the major problem is yet.” This sort of complication to the major conflict makes for an interesting balance of factors; on the one hand, you might have impatient people trying to figure out where the story is, but on the other hand, if the characters and the situations they’re getting into and out of are engaging enough—or if it wasn’t clear that there was supposed to be a metaplot, maybe a little foreshadowing but nothing more—people are likely to have a lot more patience for events that in a more focused plot would seem like digressions. (I may come back to this later.)

For most RPGs and a somewhat lesser number of stories, “We’re not powerful enough yet” is one of the primary answers. The enemy’s higher level, built on more XP, a greater Rank, pick your poison, and either way any confrontation involving the main characters and their primary antagonists would stand next to no chance of ending in a victory for the forces of protagonism, narrative immunity be hanged. It’s a valid reason, but for the plot-oriented creator/GM for character-oriented players, I would strongly recommend not making it the only reason—unless it makes in-character sense that the characters consider their best option for dealing with the difference between their power and their opponent’s to be running around getting into fights, not having a more actionable reason that just happens to involve racking up the requisite XP in the process can throw off the suspension of disbelief. For writers, the equivalent is probably “The story hasn’t gone on long enough yet”–which I suppose is a valid excuse if you’re in the middle of NaNoWriMo, but does demand a little extra justification if you expect the story to go over with an audience.

Why can’t the conflict be resolved immediately? Think about the answer carefully; it’s going to matter!

It’s important to know what general form the major conflict will take. Note that this does not mean that you need to know every detail about the major antagonist, his plans, what the main characters can do against him, what his contingencies are, so on and so forth, at the very beginning. In fact, unless you’re working on an overarching plot where mystery is an important enough element that you need to seed clues in from the very beginning—and possibly not even then—I would suggest leaving a certain amount of wiggle room as you work on the rest of your skeleton and start in on creating the story or game itself. (I for one have never had an overarching plot in which I did not have to determine at least one detail—including at one point the antagonist’s name—either at the last minute or in retrospect of the event it referred to but before it needed to be finalized.) You can have something as vague as “The main characters, outsiders in a society which is not yet accustomed to them, use their external view to help them hunt down a traitor to said society”, “A gaggle of prison escapees in a large city dodges feuding nobles as they solve a mystery that could turn the entire city upside down”, or “New transfers to a highly regarded school discover their teachers have a dangerous secret.” Neither does it mean that there needs to be only one major conflict (some of my favorite books have woven together three or more), nor that the major conflict has to be clear from the very beginning of the narrative (more on this later).

What danger does your primary conflict pose? Without the stakes, it can be hard to properly motivate PCs or keep a reader engaged—or, for that matter, make sure a character’s motivation is consistent with the situation. Threats to life, limb and the existence of the world may be common, but bear in mind that there’s plenty else that can be thrown on the table. Love, fortune, dignity—no concept is safe.

Consider also what sort of event might serve as the climax. While this can be (and in fantasy, often is) a battle, that’s not the only option—there are plenty of examples of conflicts resolving with something else. The webcomic Digger climaxes in two well-done verbal confrontations and a knacky bit of crowbar use, Lois McMaster Bujold’s A Civil Campaign climaxes with a political body voting—one can have appearances in court, long runs, cunning tricks. Knowing what you want, though, allows you to start setting your tone to fit your desired result—if, for instance, you want a nonviolent resolution to a situation, you may wish to stress the importance of the rule of law in the setting and/or the potential for the antagonistic characters to see reason, and you will probably want to avoid instilling bloodlust against the antagonists in the protagonists.

Not only will knowing your major conflict help focus the direction of your narrative and figure out what you need out of your main characters, but it will likely help to motivate you to keep going; I, for one, have found it a lot harder to invest in a storyline where I don’t really know yet what’s being threatened or how. Plan what you can, then run with it!

Scope is, as plot variables go, somewhat messy; there are a lot of things that it can encompass, all of which we need to take into some degree of account.

Particularly in a roleplaying game, the first thing we need to consider is scope of power level. With the system causing the characters to constantly improve, it’s necessary—we have to figure out where they’re starting, who we need them capable of going up against by the end, and what we’re going to have to do with the pace to get them from the first to the second without accidentally killing them en route or making the whole thing into one gigantic cakewalk. My emphasis on the necessity for a roleplaying game doesn’t necessarily mean a writer shouldn’t take it into account, either; you don’t have to look far to find examples of series in any medium that haven’t suffered from some degree of insufficiently thought out or excessively open-ended scope.

On the other hand, there’s the kind of scope that determines just how much of a given setting is affected. Some stories stay entirely within one region, or even one city; some encompass entire worlds, and even go beyond into alternate planes. This type, geographic scope, is an important consideration for a worldbuilder, since it gives her a sense of how large a map to draw, how many different places to detail out, and how important those places are going to be.

Then there’s how the plot affects other characters, or population scope. Sometimes this twines with geographic scope; theoretically, you’re going to be touching the lives of more people if you’re on a quest to save the world than if you just plan on solving mysteries in a large city. On the other hand, sometimes it’s inversely proportional to geographic scope; while they’re in the same general location as thousands of people over the course of their travels, the protagonists never stick around long enough or get thoroughly enough involved with the people in the places they’re visiting to actually leave much of an impact. (Heck, think about all those video games in which you’re saving the world and the shopkeeper can’t be bothered to give you even a five percent discount!)

If you know what sort of scope you expect your plot to take, you’ll find it easier to know what to do about the way your plot expands itself—when to speed up, and when to stomp on the brakes. Don’t neglect it!