Quote from: Black Vulmea;967552Throw away your character backstory and write a list of three goals, with a couple of bullet points for each on how you plan for your character to achieve them.

Quote from: Gronan of Simmerya;967582Get a copy of Tony Bath's "Ancient Wargaming."
http://www.lulu.com/us/en/shop/society-of-ancients-and-tony-bath-and-john-curry/tony-baths-ancient-wargaming/paperback/product-15463540.html

Read "Setting Up a Wargames Campaign," and do that with your world.  Then, run your world at the kingdom level using his rules for five years (game time, this should take about an evening.)

His rules include things like wars starting, assassinations, natural disasters, etc.  After five years all sorts of shit will be happening in your world.

Now turn your players loose.

What these two said.
There's a reason why I purchased Tony Bath's book for use with RPGs;).

Quote from: Larsdangly;967612I've never seen a 'structured' adventure that didn't make me what to drive an electric drill into my head to stop the voices from driving me mad.

Seriously, I think anything remotely 'railroady' in a table top rpg adventure just ruins the whole thing. It's like siting down to play chess only to find you are only allowed to re-enact bobby fischer's greatest game, complete with the voices, and if you screw up someone will roll their eyes and put the piece where it was supposed to go.

That's funny because it's so true:D!

Some nice advice, I forwarded a link to Chris to read so maybe he'll sign up to ask a few specific questions if he has them

Quote from: Skarg;967560But prepare things not so much in terms of a tree of expected events, but in terms of what and who that's relevant is where the PCs are going, what they are doing, so that the GM can determine during play what happens, as a natural result of what the PCs (and other agents present doing things nearby) do.

Yes, that is very good advice.

Quote from: saskganesh;967588For a new campaign with unknown players I usually start with a railroaded first adventure (*just goto the fucking local dungeon eh*) and then gradually open the world up. An NPC patron can also act as a sandbox midwife offering quests and tasks and hooks etc.

Once players are somewhat immersed, events and adventures usually start taking a life of their own. A good campaign has momentum, so it's easy to run, especially if you develop some decent improv skills and robust prep habits

I think this is a good compromise approach that works well. Eg in my Nentir Vale game the Vale has enough detail thanks to Threats To the Nentir Vale to be a viable sandbox, but the PCs have a patron, Lord Markelhay of Fallcrest, and a default activity of embarking on missions to protect Fallcrest, serve their lord, and get loot and xp. One thing I do is try to offer a variety of rumours including potential hooks to adventure sites, for starting out and whenever the PCs are less proactive.

But it's a good idea to frequently sit back and let the players run with their own ideas, don't let your mission-hooks get in the way of letting players follow up on their own ideas and interests.

Quote from: Voros;967616The trick is to make the NPCs and sidequests actually engaging enough that they function as more than Skyrim automatons handing out fetchquests.

Lost Mine of Phandelver failed really hard there. Frankly some of Skyrim's NPCs are a lot better. The town NPCs are just quest hubs, one reason I don't really recommend it unless the GM is going to put a lot of effort into fleshing out Phandalin properly.

Quote from: -E.;9676444) Provide a map -- of some kind. Give the PC's a bunch of stuff to engage with. Give them a star-chart and say, "You're here, there are 6 other planets in the system, and you understand these two are full of dangerous ruins and treasure. That one is run by an evil cyberlitch who people say has designs on conquering everything."

If you don't have a world with frontiers to explorer and places to go, it's hard to engage.

5) Provide useful NPCs -- Ideally a short list of them. These should include Patrons, antagonists, people who need help, power-brokers. Maybe an underworld contact or two. This is really another kind of map. If you're not sure what to do, go talk to Fred The Merchant who always needs caravan guards, or talk to the Scheming Executive who always needs mercenary operatives, or whatever.

More really good advice. Lots of sandbox GMs fail hard at providing adequate information to the players, especially out-of-world stuff like maps & NPC lists that parrallels in-world info the PCs should know.

The DM should have a discussion with the players before putting any work into the campaign. There is no sense in preparing a great sandbox environment then staring across the table at a herd of cows milling around waiting to be milked.

Make sure the players are into seeking opportunity and adventure. If the group cannot see its way to to do anything without being attacked or spoon fed a quest indicating the world is about to end then shitcan them and find a better group or abandon the sandbox. This will save a load of work and heartache.

Assuming you do find suitable players, start with a small piece of a world and expand as you go. A starting town & enough surrounding area with adventure opportunities is enough to start. Having an idea about what is beyond this is fine but there isn't really a need to finely detail all of it unless you really feel like it. If there is enough interesting stuff happing locally then you won't need anything else for a while anyway.

With a starting area fleshed out in some detail, there will be places to explore and people/monsters to meet. All of these monsters & people want something. Many of these desires result in action being taken and thus we have plots. Plots are not antithetical to a sandbox environment if used properly. While the players plot to achieve their own goals, the NPCs in the area will be doing the same. It is where these plots come into contact with one another that make for interesting game play. Some will become conflicts,and others may align together. As long as these plots originate from an entity within the game world they do not affect the structure of play.

Once the area is detailed and the inhabitants are brought to life and their plots set in motion it is time to introduce the PCs to mix and let them explore. Information, especially information that everyone in the region knows, should be be freely shared. There should also be opportunities to learn no so common information if the effort is made to do so. Besides a lack of proactive players, not being generous enough with initial information is probably high on the list of causes for sandbox campaign failure. Once the players have sense of what is in the area and somewhat of an idea about some things that are going on, they can decide what to do and where to go to best further their own goals.

From there it is all about following the PCs activities, and seeing what effect they have on the current status quo, and how the other inhabitants of the area react and modify their activities in response to everything.

Amazing advice here! I'll complement citing some RPGs that come ready with campaigns/advice/techniques/built-in systems from the get-go for sandbox style play:

- Beyond the Wall (with Further Afield)
- Mutant Year Zero
- Apocalypse World
- Blades in the Dark

They are worth a read if you're interested in the style, even if you don't intend to play them.

Quote from: CRKrueger;967640That's nice, but it's easy to say, and they've probably been playing a different way, so what can the GM do?

• Come up with some major situations that are happening in the area, perhaps related to hooks the players may have in the world through their history/backstory.
• When the players are presented with these situations, encourage them to react, through in-game support.  NPCs are looking to hire them, rewards are given, etc.  Make some things seem like normal Plot Hooks.  The key is make enough of them that the players have to choose what they are going to focus on.
• When they go "off track", let them.  Let things stall for a session, let them be responsible for the flow of what happens.  If they think outside the box and catch you unawares, let them do the same to the NPC, reward non-linear thinking.

You do these three things, you'll start moving them towards advancing their character goals as characters as opposed to dealing with "campaign plots" as players.

I would add to that list:

Make information easier to get than you might think.

People making nefarious plans are realistically going to keep them secret, but information is the currency of motion in a sandbox game. With no information, nothing is likely to happen -- the PCs have nothing to react to and no clear way to pursue their own goals.

At the same time, you don't want to overly-privilege one set of direction over another. If the only thing you know is that the Duke of Schemes is potting to shoot the Archbishop, it's less a sandbox than a linear plot: you could ignore it, but what then?

I think giving the PC's an unusually privileged source of info is a good idea and explains how they know things (even at their relatively low level of experience) that other (far more important and embedded) people don't.

Something like
- Rumors overheard in a bar are the traditional go-to, but 1) that's a bit cliche 2) rumors about ancient ruins are fine, but rumors about plots and conspiracies raise questions that if the PCs know about it, doesn't everyone know?
- A privately commissioned map showing multiple ruins and other places of interest
- A collection of intelligence files about "who's doing what"
- A network of informants who work for someone else but are personal friends with the PCs and leak to them the same thing could be access to police records or well-connected, streetwise PCs who hear what's going on before others do
- A black-market cyber-clearinghouse for wetwork operatives. Craig's List for freelance spies could at least give the PC's some idea of who's hiring and if it's secure / elite enough that sort of/kind-of answers the "how does everyone not know about this?!" issue

I would avoid making the source of info a patron -- if their source of information is a powerful friend who expects them to act on what's been told, then it's less of a sandbox and more of an agent/mission type game, which is fine but not the same freedom you're going for.

Note that sandbox games can easily and productively evolve into an agent-type game: that's fine. But if you start out that way you're kind of stacking the deck.

One other note on information: After giving it some thought, I've decided that in at least some cases NPCs are worse liars, easier to break, and way more lose with key info than they might be in real life. This is a personal judgment call but
1) I don't want to force the PCs to face "do we torture this guy or get nothing" a lot of the time. In many cases, forceful questioning is enough to scare someone into talking
2) I want dealing with NPCs -- in general -- be productive and interesting. Having everyone always lawyer up is dull and just forces escalation, possibly past the boundaries of good sense (see below)
3) Characters who are experienced interrogators would know how to get people talking. I've had formal military training in interrogation and have role-played profession training scenarios where I'm being interrogated and had to force the trainee to find my weak-point. Most of the people in my games do not have that experience. If I force them to roleplay it... we're never getting anywhere.
4) As I said above, information is the currency of motion. To the extent that I withhold information, the game stalls. Getting info can be fun and exciting, but I find the game is always better when the PC's know what's going on (to a reasonable extent) and move from collecting intelligence into action

Cheers,
-E.

Quote from: -E.;967644If you don't want characters scattering to the 4 winds, ensure they have reasons to stay together -- ideally pre-existing relationships & shared goals

No. Just fucking no.

Stop "giving" the players anything beyond information about the game-world. It's up to them to decide why to stay together, to negotiate with one another to make that happen, and for fuck's sake stop relying on character 'backgrounds' to tie them together.

Just fucking play, and sort it out as you go.

Quote from: -E.;967644My thoughts on running sandboxes:

1) If you don't want characters scattering to the 4 winds, ensure they have reasons to stay together -- ideally pre-existing relationships & shared goals

Yep I prefer to have something that ties the PCs together, esp in a sandbox. Personally I like to have them roll on a party bonds table. I think this is more important than a detailed "backstory" in fact. I like to see a one or two sentence history, plus party bonds to the other PCs, and off they go. Then play to find out more details about their history - via insta/improv history - as the game unfolds.

Quote from: Black Vulmea;967720No. Just fucking no.

Stop "giving" the players anything beyond information about the game-world. It's up to them to decide why to stay together, to negotiate with one another to make that happen, and for fuck's sake stop relying on character 'backgrounds' to tie them together.

Just fucking play, and sort it out as you go.

Oh, please.
For someone who's group meets six times a year, you have awfully fucking strong ideas about how those of us who play every week ought to do things. You'd discover your impractical idealism is ivory-tower crap, and that people who actually game have figured out how to make it work.

If you come down off your high horse, you might find that

1) If the people in the game want to scatter, fine -- I explicitly addressed in the post you quoted -- but if you don't want that, then work with everyone to ensure that there's alignment. This is basic stuff.
2) The GM -- the traditional RPG model -- is the elected group leader. Show some fucking leadership.

Cheers!
-E.

Quote from: Black Vulmea;967720Stop "giving" the players anything beyond information about the game-world. It's up to them to decide why to stay together, to negotiate with one another to make that happen, and for fuck's sake stop relying on character 'backgrounds' to tie them together.
.

I do it both ways depending on my campaign, but I definitely don't see anything wrong with having players get on the same page with background stuff or motivation so the campaign functions smoothly. Both have their upsides and downsides. A campaign can definitely get difficult to manage if players are both not in congress enough with one another and not enjoying having people going after different things. I think the latter is really important to determine though. I have some groups who love that kind of freedom and are fine with the group not being on the same page, but I also have groups that get annoyed with that. There is also the GM factor, the less on the same page the party is, the more work you might potentially have in terms of prep between sessions. I think is a very individual thing based on the players present.

Even in a sandbox game you've got to have the players stick together even if the reason is fairly obtuse. It's fine for them to pursue their own goals, but it'd be better with the assistance of the other players (in turn). Well, more interesting for the whole gaming group. Using a spotlight from time to time is great of course and gives the characters their own time to further their agendas.

Having a commonality works very well in a game like 'Beyond the Wall' and the mechanics reflect this in interesting ways that really help the GM to build on the game from the ground up.

Personally, I'd prefer a more WFRP approach. Such as greed/money keeping the group together. Or fighting some common enemy out of revenge. I'm not all that enamored at playing 'good guy' heroes per se.

Being bonded to the players through their backgrounds can be a double-edged sword because it can be restrictive if not done correctly. If you can lump them together initially (which is easy enough), then a bond will form naturally and you should be good to go from there. Unless you players are being deliberate dicks or somthing.

I'm withE here, a starting bond and common goals between players tend to help. Beyond the Wall and the other games I've cited does this to great effect.

But if the group prefer otherwise, just go with it!