DM Style: Feeder or Arbiter?

[jdripley]

I've been rattling this around in my head over the past month or so and I think the ideas have crystallized enough to write them down and ask the group about them.

A DM can be many things depending on the game, the players, and his/her own personality and strengths. But I think there are two basic archetypes: Feeder and Arbiter.

Feeder: This is the style I'm most used to. In fact I'd say that 90% or more of the DMing I've played under has been this style. The DM comes up with an adventure, makes up a fun plot and a bunch of details along the way, and then he sits the players down and works them through it. By carefully feeding information to the party he guides them from point A to B and so on until the party triumphantly wins the day and marries the princess, or whatever "winning" looks like. The DM has to do a LOT of prep work ahead of time in order to have his ducks in a row and not keep the players waiting as he mulls over every detail, and to be sure he has a well-flowing plot.

Arbiter: This is honestly an unknown style to me. I've seen snippets of this style mixed in with Feeder style, and I think it could work out, but I'm just not sure. Anyways, as the Arbiter, the DM plays referee for the players, and the players come up with the story as they go. So instead of the DM saying "you're going to rescue a princess from a dragon's castle," the players say "hey, wouldn't it be awesome to go an an epic journey to a castle where a dragon keeps a princess prisoner?" And you as DM say "ok, what are your first steps?" And then the players come up with the whole thing. The DM sits back, enjoys the story, throws a few monkey wrenches in there, roleplays (some? most?) of the NPC's, but isn't even averse to letting the PC's roleplay some of the NPCs themselves, plays the role of their adversaries both socially and in combat, and referees the players so that things don't get out of hand (i.e. "I see a dirty peasant. I kill him. Holy smokes he's got a vorpal longsword hidden in his jacket!"). The DM must have a pretty thorough understanding of the campaign world, and honestly so would the players, but the onus is on the players to come up with the story, motivations, etc., and the DM is there to referee, add details, throw monkeywrenches, and play the opposition.


Has anybody successfully used the Arbiter style? I love DMing but I hate all of the prep work. And I'm always bummed for my buddy when he DMs and players aren't too into his story. So I think that maybe just having the players drive the story from the get-go may be a better approach for me. But it's a mysterious unknown to me!

[WarpedWolf]

I try to run Mutants & Masterminds in more of an Arbiter style. I know what the supervillains have done, what they are doing now, and what they plan to do. At that point, I start reacting to player actions. Investigation, checking up on contacts, that kind of thing. And if they come up with something I never even dreamed of, I've been known to toss my entire prepared plot out the window. (I've been told my improv game is better than a lot of GM's planned game.) Unfortunately, I've usually got a lot of players who are used to riding the plot railroad, and I have to poke them more than I'd like. Getting players used to the idea that they can come up with what happens next is a challenge.

I'm hungry to try FATE or something similar.

[baronvonchaos]

Ah, so I am about 40/60 respectively. 40 - feeder, 60 - Arbiter. I'll give them ideas of what's going to happen, and they can say what they're going to do.

Them: So they're going to go into a forrest to look for the plot hook. (if they just went in and found it, the story would be lame and over, I need them to level up first).

Me: Well they're going to get attacked by some Woldwardens (spec'd above their player level). If they start failing, (which is kind of my point) they'll get saved by a forward band of Exemplars. For the trouble, the Seneschal or Gravus (whoever I write in) will ask them to come on grounds of suspicious activity or to heal their wounded depending on if they resist.
The story progresses and how they interact with the Scrutator or Vice Scrutator there will dictate their favor with the Menites locally and how the menites will help them if they're in trouble or hunt them down as heretics and enemies of the church.

The same will go for every faction in the game. How they play the diplomatic parts out will dictate how their conflicts will be set up in the future. I have a point a and point b which is beginning and end. Whatever they do in between is their go... If they want to procure better weapons or better warjacks/beasts, etc. they can. I'll let them figure it out.

It also matters if my whole group spec's combat heavy. The idea that Simon Berman put out in the 5/16 insider about Companies helps me a little bit. The PC group composition will dictate how and who I make them encounter. If 1 is studying to be a warlock or warcaster, I'm going to acquaint them with each of the factions and whichever one they ally with they'll get intermittant help (and possibly jacks/beasts) from.

If one's a troll warlock, you'll leave your dire troll in the woods (because it would freak out the locals) and there'll be random encounters along the lines of, "You realize that nagging feeling in the back of your head is really Fuork the DT Mauler who's super hungry and he's approaching a farm looking at a cow. Trouble! and with me its, "...and react!"
The thing I'll track is favor with factions and otherwise it's, point a = you are here. Point b = End of story.

I've played games with a GM who Arbitrated too much and it stopped the storyline and everyone got confused and helter skelter. We were supposed to stop this drug trafficing ring and 1 of the folks in the group bought a house at an undisclosed location and started growing and distributing it. "unbeknownst" to the rest of the group. Well we asked him to tell us where his house was. We went to the authorities, and they wouldn't tell us. We went to our black ops headquarters and couldn't get any information. We were stumped. It ground the campaign to a halt and nothing could progress and the GM finally fiated that his house was burned down and moved us to a distant city. Essentially a new campaign but same characters. It was horrible. Frustrating. Several good weeks wasted with 2 PC's holding everyone else hostage.

I've also played games where it just moved from meaningless encounter to meaningless encounter. Good story is amazingly hard to find.

[baronvonchaos]

[...]And if they come up with something I never even dreamed of, I've been known to toss my entire prepared plot out the window. (I've been told my improv game is better than a lot of GM's planned game.) Unfortunately, I've usually got a lot of players who are used to riding the plot railroad, and I have to poke them more than I'd like. Getting players used to the idea that they can come up with what happens next is a challenge.

I'm hungry to try FATE or something similar.

This is what I run into. My meta is used to being spoon fed (in a non-condescending tone) a story. The choose your own adventure is what I like to go with as I can deviate from the intended plot course easily. You need good RP'ers to come up with their own story. M&MM may be a little easier if people play evil chars so their plot is domination... etc. In the IK, or DnD, anything can happen so I'm not entirely sure how to completely arbitrate without story hooks.

Fate is cool btw...

[Blackraine]

My friend tried the arbiter thing, and it's pretty hard. The hardest thing about it is reigning the players in to a reasonable pace... we tried it with a pathfinder game and ended up ascending to godhood around lvl 15, when it should've been more like lvl 20 or 25. Part of that was his fault, because he wasn't throwing enough roadblocks to our plan in the way, but part of it was our fault, as players, for not self-moderating and realizing that that's more of an epic campaign sort of thing, and should probably wait until then. The Arbiter style definitely has its pros, however... it's much easier to tell a political-based story from this perspective, as the GM can just, more or less, play his own antagonistic character and roleplay his political approach alongside the players.

I'm somewhere in-between the two. I sit down with a story in mind, but don't work overly hard on how to get from point A to point B. Instead, I've found it easy to just sketch out major plotpoints, like a storyboard sort of thing, and then make up the details as they come along, while letting the players figure out how to get there. Meticulous planning just gives the players more ways to throw a wrench into your plan and leave you up late at night for the next week while you meticulously plan a workaround. However, simply presenting the next point on your storyboard as a mission objective or a plot twist will allow the players to do all the heavy lifting for you. Beyond that, I just need a few ready-made encounters to cover the players' most likely plans of attack, and go from there.

As an example, I ran one game where I literally just came up with a few interesting NPC's and picked some interesting monsters out of the monsternomicon. I just went through, week after week, and found new interesting stuff to throw in. The first 2 or 3 sessions were just random encounters and travelling around as mercenaries (I had a mercenary contact in Five Fingers who was feeding them missions to drive what little plot there was at that point), but then they ended up going to Five Fingers to meet their mercenary contact and work on "Something big", having proven their value through the execution of these somewhat menial tasks they'd been doing. They ended up infiltrating the gangs of one of the captains and working their way to the top of his outfit, and he began dispatching them to investigate some Cyriss cultists who had recently moved into his turf. Through that, they started uncovering the dark secrets of the Cyrissists until, unexpectedly, their Mercenary contact (Deadeye Darcy, my GM character at that point) betrayed them. It turns out she was working all along for some Iosan triplets who were convinced that the Cyriss cultists had stumbled onto something important, and were going to use the power from the leylines the Cyrissists had discovered, along with their own selfish acts (including the sacrifice of hundreds to get this far) to ascend to godhood along the path of Thamar and take the place of their fallen pantheon to save the elven race from extinction. And for that entire campaign (which lasted about a year and a half), I had 6 plot points: Get to Five Fingers, Introduce Deadeye Darcy, Infiltrate the Pirates, Come up with an excuse to check out this cool Cyriss base under Five Fingers, Betray the Party, Have a big battle with three Iosan Malefactors of Thamar. Everything else just came from browsing the book and going "Oh, that's cool" and letting the players figure out how to get from point A to point B.

[AJ the Ronin]

In the last few years I have tried to be an arbitrater but my players are so used to feeders GM (mine and others gm fault) that if I don't feed them plot some games may come to a halt.

[J. Beatnik]

In my gaming group, we have 3 GMs. Matt, who is a "Story" GM, which is a Feeder GM. He writes an epic story lasting the campaign and we basically play through his written story. He tends to railroad us a ton, but over time he's gotten better at knowing us as players and being able to write us cool parts that we enjoy. There's a little bit of leeway in the story, but as he doesn't think on his feet quickly, it tends to fumble up what we're doing at the time. His campaigns tend to be full of these great spotlight moments where the characters are doing cinematic things to further the plot. (dice may or may not be involved)

Then we have myself, who is an extreme "World Building" GM, which is an Arbiter GM. I build the world in intricate detail and have all kinds of encounters written for the location they are in. There's a ton of things going on behind the scenes that the characters may or may not run across in the campaign. I don't usually have a specific goal in mind, but I usually throw some plot hooks into the game and see what catches some PCs. If any of them gets passionate about something, then I know where to lead the campaign for the time. The downfalls of my campaigns are that there really isn't any direct main storyline, but rather wandering around doing whatever the PCs feel like until they get bored and we start a different campaign. With Matt's Campaigns, the climax and epilogue tend to be the highlights, where in my campaigns the crazy stunts the PCs pull and the fully fleshed out NPCs are the highlights. I may throw in a big, bad at some point, but generally the PCs end up adventuring for a time, then buying a cottage, farm, or keep and settling into a life of peace/retirement for a time.

Then we have Andrew, who is huge into Game of Thrones. His campaigns tend to be a little bit of both, with great city building and deep, meaningful PCs. He runs a very political game where social skills are as powerful as combat skills, and the players can change the direction of the campaign at a whim. Players keep secrets from each other and sometimes characters make wrong decisions and die abruptly. (Matt and I tend to be more cinematic and not kill characters without good reason) If it makes for a better story, Andrew will kill your character even at the beginning of a session. My last character ended up being the whistleblower of the organization we worked for that was turning viciously evil, and Andrew and I talked about my character becoming the martyr for the new revolution cause that would spring up out of it. It was all super awesome and had elements of both DM styles.

I don't think that either of the styles is better, it just depends on your players and what they enjoy. With the Feeder style, you have to know what your players enjoy. With the Arbiter style, you have to make goals to reach for, challenge for excitement, and an end game in mind.

[susan]

Interesting reads . I've actually had very similar conversations with my gaming groups, although instead of "Feeder/Arbiter" we use "Writer/Improver" to describe basically the same thing.

(As an aside, I highly encourage all GMs/potential GMs to take Improv classes. For Seattle-area folks, the Jet City Improv 101 Class was one of the more amazing and fun things I've done in my life, and most major cities should have local groups that offer classes.)

I believe all successful GMs are some combination of Writers and Improvers. It's possible for GMs to be 100% one or the other, but at that point you're probably playing Hero Quest (100% Writer) or Fiasco (100% Improv), and while both of those are extremely fun, they're probably not the Tabletop RPGs you were looking for when you sat down at the table. In my experience, most GMs begin leaning heavily towards the Writing and Planning side of things, coming to games with copious notes and pre-planned combats and maps, while as GMs and Gaming Groups evolve and grow over the years they drift further and further towards the Improv side of things. My personal play style is heavily Improv, though for the first intro session to any new game or group I will sit down and write up a traditional adventure just to give players a chance to learn their characters and the system (ie, my First IKRPG Adventure Thread posted here a couple weeks ago). After a set amount of familiarity is established, I'll tend to just come to a gaming session with a few notes, a couple bad guys that they may or may not fight, and an introductory paragraph describing where the players are at at the start and reminding them of why they came there. This is not to say I don't Feed the occasional plot hook; far from that, I love to dangle hooks for players to grab (play the "Yes, and..." game to its fullest, leave wide openings for action). But trying to pre-plan for a Players' reaction is foolish and distracting and almost guaranteed to be wrong.

Two quick anecdotes to illustrate my points, first starting with the dangers of planning too far ahead in Feeding/Writing style and the need to integrate Arbitration/Improv: In one of my Brother's Mutants & Masterminds games, he began an adventure with my character (a paranoid Wizard) receiving a mysterious package at his doorstep. His plan was for my character to open the package and find an iPad Tablet, through which an NPC would call me and remotely give me an assignment. As a player, I destroyed this plan by seeing the mystery package and assuming it was a bomb (in my defense, he's done that to me before), leading me to blast the package down a hallway and dive out my penthouse window to "safety". The plot hook literally destroyed, my Brother was forced to throw out his plans for my character and find an alternate way to get me connected with the action much later in the game after the Villains had put several plans into motion. It worked out in the end, and was fun for all.

A second example, this time highlighting the difficulties of a nearly pure Arbitration/Improv style and how to plan for it: I was running a Shadowrun game for some players. The first intro adventure, the PCs located a secret medical experimentation lab, freed the subjects, killed the evil scientists, and burned the place to the ground in heroic fashion. The second adventure, I wanted to throw a twist at them, so they were contacted by a new Mr Johnson who wanted them to dig up information about "a group of terrorists who destroyed one of his clients bio-medical research labs and murdered several doctors and nurses at the facility". I hired the PCs to hunt down themselves. Now, I knew that the PCs were going to have a very dramatic reaction to this proposal, but I had absolutely no idea what that reaction would be, so I literally had no notes or adventure planned after two paragraphs of intro conversation, one of which described the coffee bar they were meeting in. After the Mr Johnson left, I allowed the PCs to talk among themselves as they were hashing out their plans, and as they were talking I was taking notes about what they were saying, quickly referencing their random tangents to enemy stats and locations, and creating a framework on the fly for the remainder of the adventure. I did have to bring them back on track several times, build some intensity to the scene and throw down a few perceived roadblocks when they started to stray too far, but by and large the PCs were allowed to create a wholly original plan and implement it. I won't go into details, but I will say that the plan at one point involved several PCs sniping security forces while dressed as Girl Scouts. Much fun was had by all.

[baronvonchaos]

QUOTE=susan;1441092][...]As a player, I destroyed this plan by seeing the mystery package and assuming it was a bomb (in my defense, he's done that to me before), leading me to blast the package down a hallway and dive out my penthouse window to "safety". The plot hook literally destroyed, my Brother was forced to throw out his plans for my character and find an alternate way to get me connected with the action much later in the game after the Villains had put several plans into motion. It worked out in the end, and was fun for all.[/QUOTE]

Ditto for an interesting read!

And HA HA HA HA HA. To "safety." The people I play with are wacked out like that, but none have been as diabolical as to leave psychological scarring that would prevent a story hook from happening. Everyone I play with is the, "OOh! a mystery box!" Rather than, "Hide! It could be a bomb!"

http://www.youtube.com/watch?v=5CuK18CZOG4 || Akin to this...

[cryptomancer]

Flexible Feeder is my typical style as DM. That said, I've played games where we, the players, completely come up with a plot twist on our own, plan an adventure on our own initiative, and then execute the plan with the DM playing, as you say, referee.

The trick is balance between these two styles. The problem with playing Pure Arbiter is that (and I've played games like this before) the game world becomes ill-defined. The rules of the world, not the game but the world, are not well established or known.

Ultimately, the DM's duty is to provide the medium in which we players swim. How do I know my player wants to lead a slave revolt? Well, the parameters of the world (slavery, resentment, a foreign occupying force, huge un-tapped slave population) have been established by the DM. How do we know the shape of the town? Or the country? Or the continent? How do we know who is where and what their motivations are? The DM provides.

Granted, the Iron Kingdoms RPG is a good place to try out the Arbiter style, since the parameters, rules and details of the world are so well established. Say you want to run a Broken Coast game. Well, give the players a small tramp freighter (do the words Millenium Falcon leap to mind?), a map, and a general motivating force ("you're three days away from running out of food, four from running out of coal") and see what they do.

IF the players feel a sense of sandbox freedom, IF they've got the capacity to get around the map on their own, and IF there's still a reason for the party to be a party (instead of just a bunch of folks in the midst of mutual back-stabbing, which sounds fun on paper but doesn't work well in play), I believe the Arbiter style can work, somewhat, sometimes.

-crypto

[Lachlan the Mad]

I've played a lot of one-off Paranoia games with an Arbiter-ish kind of GM; Arbiter-ish in the sense that he rolls up new random characters every session and mostly wings it. It's probably about a 20:80 Feeder:Arbiter ratio. His basic technique is to send us to a briefing room, where we are given a mundane Troubleshoot; "Go to the Box Packing and Unpacking Factory and shut down unpacking operations", for example. Then he just lets us argue our way out of the dormitory and off to the factory. His main role is dropping obstacles in our path, creating mazes of different-coloured doors, and encouraging inter-party conflict. Half of the party is inevitably on at least their second clone by the time we're out of the dorm.

Here's an example of some of the arbitrary stuff this arbiter comes up with:

GM: You approach the factory. The front wall has a black door, a red door, and an orange door.

PC: Okay, so the black door is for workers, the orange door is for management... what's the red door for?

ME: ZOMBIES!*

GM: ...A zombie comes out of the red door!

ME: I give it a thorough hygiene inspection! <rolls dice; critical success>

GM: Okay. You just scared a zombie away with a hygiene inspection.

ME: I WIN THIS GAME FOREVER!

*This was a spur-of the moment yell; there are not normally any zombies in Paranoia. The party are good enough at killing each other that they don't need monsters to do it for them

[jdripley]

Thanks everybody, I've found these responses to be very helpful I'm going to go start another thread about defining an end to a campaign and I'd love to hear your input on that too!

[StJason]

My pal Phasmaphobic has written systems that focus on the 'Arbiter' system. In Cannibal Cognition, you use a faceoff with cards, and the highest card tells how the story goes at that point.

I think that it's more of a continuum then a binary opposition. Most people I know run games with a plot in mind, but with considerable leeway, and readily adopt improvisations of the players. I honestly think that a 100% either way would be unplayable. A 100% feeder would remove the character from the player's control, and you'd end up listening to a story. A 100% Arbiter would quickly bog down as there is nothing much for the characters to do. Sooner or later, the hand of fate needs to step in, and this is what a RPG is all about.

[Sosthenes]

"Arbiter"? "Feeder"? Where's "Antagonist", ye milksops?

[susan]

"Arbiter"? "Feeder"? Where's "Antagonist", ye milksops?

[Kriegtanzer]

Our main GM is very good at stringing together various adventures. In our Warhammer campaign he has run adventures based on WFRP adventures, AD&D adventures, D&D 3.5 adventures and make it all feel connected. He is big on using published adventures as a framework and modifying them. Part of his modifications is to add in the little details that make them feel connected.

When I have run games and I am not using published adventures I usually start with the story idea. I will prepare a number of possible villians. I will lay out the story and then react to the direction the party takes.

[Sosthenes]

...snip...

Eternal vigilance is the price of game mastering!

[The Happy Anarchist]

One way that fits somewhat in the Arbiter method, but with less requirements on the players is for the GM not to come up with plots, but to come up with NPCs - and then play them to the hilt.

Blackjack from the old dumpshock archives did a great article on it.
http://web.archive.org/web/200610310...gamemaster.htm

In fact, read the whole series. Lots of good, unorthodox stuff there, some of which is applicable to different games. And if you don't like that method of GM? The CLUE files from there are always worth laughing over too.

[Iron_Peanut]

Re: Feeder or Arbiter? Yes to both. When I first started GMing, I was heavy into the Feeder style, but I quickly realized that it was a lot of work and I am very lazy, which was a bad combination. I also learned my first important lesson:

• No plan survives contact with the enemy (PCs).

I learned this pretty much right away. Assuming you've got a decent group, the players are people you think of as peers. They are at least as clever as you are, maybe even a little bit more clever than you are. There are a bunch of them and only one of you. When you think that the most likely ways a scene will play out are Result 1 and Result 2, the players will frequently shoot for Result 7f-subsection B. You cannot possibly anticipate how the players will respond to a situation, or even the most likely way they will respond. Keep the plan loose and don't worry about the specifics too much. Be ready to improvise - it's the most important skill to develop.

[Iron_Peanut]

Further thoughts on improvising:

• Always punish PCs for everything.

I learned this playing Mouse Guard. In Mouse Guard, the example offered is this: Say the you're at a river. A flash-flood has knocked out the bridge. In the water, clinging to a brach, is your brother. Caught on another branch is a vital shipment of medicine needed to cure an epidemic in a nearby village. You know that another swell is coming soon that will sweep everything away, and you only have time to save one. Which?
Does he save his brother? Punish him! The medicine is lost and now the town will die! Does he save the medicine? Punish him! His brother is lost and now he shunned by his family and friends as cold and unfeeling. Does he waffle, and do neither? Punish him! He is known far and wide as a coward! Does he attempt to save both? Punish him! He is swept away by the storm, both his brother and medicine gone, and is lost on a distant shore! Does he, in a way you absolutely had not planned for at all, concoct a genius scheme that neatly resolves the situation with minimal cost, and then intelligently use his available character resources and skills to successfully implement said scheme? PUNISH HIM! Players do this with aggravating frequency! Let him have his win, and be sure to up the ante next time!
Player success is different from character success. If the characters encounter endless problems and have to give everything they've got to make any headway at all, great! Things are exciting! Assuming you're not being a total jerk about it, the players should be having a good time! If not, fine tune it a little bit. Dial the jerkness back a notch.

• First rule of improvisational theater: "Yes, And..."

I learned this playing Exalted. In Exalted, the PCs are golden demigods who are the divine rulers of all that they survey. There's pretty much nothing that's out of their power - even core gameplay and setting concepts can be ignored or worked around if they've got the right tools. So saying "no" to players is pretty useless. And we had a lot of fun playing Exalted. And I port that mentality over to whatever else we play.
Saying "No" brings play to an abrupt halt, which is not fun.
The "Yes" keeps play going, which is fun.
The "And..." part is where you throw the consequences of the "Yes" part at them.
Every insane plan, world-altering scheme, and canonically impossibly notion your players put forth is an adventure hook waiting to happen. Do they want to build a time machine, travelling back to to the Orgoth invasion and stopping it before it starts? Sure! But how are you going to build a time machine? Maybe you need to infiltrate Ios and decode their secretive magical formulae. How are you getting into Ios? Where are you getting someone who knows the ciphers of House Shyeel? Can anyone even read Iosan? Maybe you need to rob a temple of the Cult of Cyriss for their fantastic mechanika. Maybe you need to broker a deal with the Infernals to gain the more ephemeral parts of what you need. The Infernals are jerks - what if they want someone you're not willing to give up? The PCs should be theoretically able to get whatever they want, but they've got to be willing to put in the work (which might be more than they can accomplish - for now) and pay the prices needed (which might be more than they can afford - for now) to get it.

This mentality extends to mechanical stuff to. If the players need to do a thing to advance the story, like interrogate a guard or pick a lock or climb a wall, what happens when they fail at that check? Do they stop completely and grind the story to a halt? That's dumb. Do they just keep re-rolling until they don't fail? That's also dumb. Keep things moving, and spin their inability to solve one problem into an exciting NEW problem that they need to solve.

[rucdoc]

I don't think I could defined as either a feeder or arbitrator completely. I do design stories and encounters but I do so understanding that how my palyers encounter these may or may not be the way another DM would. when I start a campaign I do a bit of world building I make the big bads, some little bads, I figure out where the group will start and the area around that, and then I turn them lose. I'll throw them a hook or two if they bite good if they go off I will improvise. I do however have one strict rule at my table, if you say it at the table when it comes to the other players you better be ready to roll for it. so if someone jokingly says I think I'm gonna steal that cool item joe just got, then I expect him to roll for it. while at the begining the group i was running thought it was bad, after the second session they understood why I did it.

The second session I ran two of the party members were argueing at the table, they disagreed on what the party should do, one of them muttered under his breath I should deck you, as the other member was being rather pig headed. I heard him quiet clearly though and told him to roll. this started a rather large fight in a tavern in the game that ended with one dead party member three dead npc patrons and the understanding that if they take the game to seriously they will all end up killing each other. after that everybody had alot of fun.

[cryptomancer]

Peanut? You're the man. I must do this now!

-crypto

[Munindk]

Most people I know run games with a plot in mind, but with considerable leeway, and readily adopt improvisations of the players. I honestly think that a 100% either way would be unplayable. A 100% feeder would remove the character from the player's control, and you'd end up listening to a story. A 100% Arbiter would quickly bog down as there is nothing much for the characters to do. Sooner or later, the hand of fate needs to step in, and this is what a RPG is all about.

Pretty much what I was thinking, my own campaigns always have a roughly outlined plot and then I let the players actions dictate the small stuff. I think the best sessions are when I set the scene and then spend a few hours listening to the players describe their actions, only butting in to describe how their surroundings respond to the actions of the players, very much in the style of Iron Peanuts "Yes, and..." concept. Then I set the next scene based on the outcome of the previous and what I think needs to happen to advance the plot.

This only works with players who are active and interested, some sessions I end up more or less being a feeder, due to lack on player participation. Since this does nothing for advancing the plot I always try to switch the players over to a sup plot, preferably something that at least on player is heavily invested it so I'm not the only one trying to get the group active, this occasionally fails and I have to resort to throwing some random combat encounter in, nothing wakes up PC's like mortal danger (and a seemingly random combat can often lead to an interesting sub plot, or even be worked into the main plot).

[Iron_Peanut]

This only works with players who are active and interested, some sessions I end up more or less being a feeder, due to lack on player participation.

This is especially a problem if your players come from a DnD background, especially if they leaned heavily on pregen modules. Even more so if you have first-timers who are used to videogame RPGs - they're accustomed to being led by the hand. You need to break them of this bad behavior. Through play and OOC talk, players need to know that indecision merits punishment more than bad decisions do.
More involved players pick up more of the work in telling the story and running the game, which eases the burden on the GM and makes things fun for them. The GM deserves to have fun too.

As a player, I'm a deeply selfish person and I want the story to be about my character as much as I can manage. The best way I've learned to do this is to give the GM plenty of rope to hang my character with. I'm invested in my character, and I want to see him succeed, but I know the GM wants to make his life miserable. If I make that easier for the GM to do, he will spend more time tormenting my character. I'm guaranteed that my character gets more time in the spotlight.
When I'm putting together a character background now, I spend more time on is coming up with good, solid pieces of rope for my character. These are the people that this he cares about, and these are the ways those people could be threatened. These are the mistakes he's made in the past, and how they could come back to haunt him. The people he's made angry with him. The debts he owes. The mistakes he's likely going to make, and how someone else could capitalize on his shortcomings.

[Munindk]

I think its more a problem with the modules than with DnD, its hard for an adventure module to flexible and still provide a cohesive story, no matter the system. I picked up the new Marvel rpg recently and apart from some not very intuitive dice mechanics, it looks as though its all about the arbiter style of gaming, even character creation is very much up to the player.

[Defenstrator]

but I will say that the plan at one point involved several PCs sniping security forces while dressed as Girl Scouts. Much fun was had by all.

This is the essence of true role-playing.

I'm a mix. Some adventures I plan out, others I just have a framework for. I try to be more free form, but I find it can be a lot more hit or miss than working it out ahead of time. I honestly think that any campaign is going to be a mix. Even the most free form group is going to need some basic hooks to give them some direction, while no matter how much you plan every GM has to make stuff up on the fly when the players do something you couldn't forsee.

[WarpedWolf]

I won't go into details, but I will say that the plan at one point involved several PCs sniping security forces while dressed as Girl Scouts. Much fun was had by all.

That mental image will be inscribed in my brain forever. Right next to the dump truck full of bowling balls.