I wanna get back on the railroad

[S'mon]

Which will lead to people telling you that 3.5 is poor for sandboxing and instead you should be playing [insert favorite past edition or retro-clone here].

I think it can be done - I ran Lost City of Barakus in 3.5 D&D over ca 35 sessions; it's basically a sandbox mini-campaign. Vault of Larin Karr is another example by the same author. Improvisation requires use of the Monster Manual and the pregen DMG NPC stats, plus plenty of Warrior NPCs - elite orcs, human guards and such (for some annoying reason the NPC classes don't get pregens). The 3.5 DMG includes wilderness encounter tables, but they are full of annoying stuff like unstatted 5th level gnoll ranger NPCs that the DM is expected to detail.

I'm planning to try some sandboxing soon for my tabletop game; I've thought about tweaking 3.5 but I'll probably go with C&C.

 

[resistor]

You should read Ben Robbins' West Marches series of articles for advice on how to run a kick-ass sandbox.

Seriously, highly recommended.

 

[Stoat]

Note the second of the two rules Robbins had for the West Marches:

2) The players have to tell the GM where they plan on going well in advance, so he (meaning me) has at least a chance to prepare anything that’s missing. As the campaign goes on this becomes less and less of a problem, because so many areas are so fleshed out the PCs can go just about anywhere on the map and hit adventure. The GM can also veto a plan that sounds completely boring and not worth a game session.

 

[Noumenon]

I didn't actually want to prepare a sandbox, mostly because I never had the idea of a sandbox populated with Goodman Games modules. I just wanted to give the players more freedom and more illusion of freedom, but it ended up just ambiguity and they didn't know where to go. I definitely will force them to tell me what they're doing now -- the next session's in three more days, hope they can figure it out.

 

[Stoat]

Don't be shy about giving the options or suggestions, particularly at the beginning of a new campaign.

I like to start a new campaign by giving my players two unambiguous hooks. Things that are real obvious that they can do if they want. In my experience, it prompts the players to do something, rather than just flounder around.

 

[Skallgrim]

Also, depending on the "speed of play" of your group, preparing a sandbox can simply be more like a "shrub" adventure than a "tree" adventure. Rather than having prepared a long and dependent series of encounters, interaction, and descriptions, you might be able to prepare short chains of linked encounters. Once the PCs have committed to one of these branches, most of your prep can focus on the choices on that individual branch.

If a flightly and indecisive group is giving you problems by constantly shifting focus, look at ways to keep them from backtracking. In one way, this IS railroading, but in another, they still have a wide variety of choices, but none of those choices "resets" them to the first group of options.

I know that sounds "gamey", but in the real world, making some choices means not being able to make others. It's possible to "direct" PCs through having realistic consequences and resource constraints, rather than artificial plot constraints.

Also, I wouldn't be afraid of simply using narration to handle a lot of PC exploration in "undeveloped" regions. You can be really cheesy with this "You look for days and find nothing. What now?" On the other hand, if one guy starts up a bar fight when you have nothing prepped, I don't see anything wrong with describing colorfully how the fight is resolved. The PCs are usually Big Damn Heroes and are probably going to win a bar fight anyway.

Again, to some DMs, this smacks of railroading, but on the other hand, I've actually played in groups which INSISTED, despite no one enjoying it, in roleplaying out an "equipping" session, rather than just deciding what they wanted to buy and spending the gold. There's a balance between DM narration and party narration, and you can often slide it one way or the other during the game.

 

[Remathilis]

I didn't actually want to prepare a sandbox, mostly because I never had the idea of a sandbox populated with Goodman Games modules. I just wanted to give the players more freedom and more illusion of freedom, but it ended up just ambiguity and they didn't know where to go. I definitely will force them to tell me what they're doing now -- the next session's in three more days, hope they can figure it out.

In my opinion, 4e got this part right.

Ask each player, at char-gen (or in your case, NOW) the following three questions...

1.) What made them want to be an adventurer? (This can be a detailed background or simply "it seemed like fun")
2.) What goals does your character seek? (fame, glory, honor, justice, love, power, revenge, etc?)
3.) Where do you see this character at 20th level? (This is the capstone question; does the PC seek an empire, ancient forgotten lore, personally serving his god, founding a thieves guild, settling down with a family, or conquering the lands around him to rule under an Iron Fist).

The first question is where the PCs have been. The second is where they are now, the third is where they want to be. Some player's won't give you anything worthwhile; that's ok. Those that do, start sprinkling your game with clues, plots, mysteries, and events that further these elements. An old rival returns to settle the score, a rumor of the PC's target of vengeance in town, a chance meeting with a lovely elf-damsel, or a chance to become commander of a lonely keep on the borderlands.

Now, weave these plots in between your Goodman Modules (oh, and grab some Necromancer & Paizo ones while you're at it, shameless plug). I typically go for every full module, I try to run the next few sessions revolving around a PC (or more, if some share similar goals or interests.) I also use these "home" games to further the plot and pad levels so they are ready for the next module I wish to run.

And sometimes players will ignore the Palace of Unspeakable Ickyness to explore a tangent plot (the six-fingered man IS here, lets search the town!) have a few stock encounters (or shamelessly rip them from unused modules) and some RP opportunities, then glide them back on track (the six-fingered man's goons said he went to the Palace of Unspeakable Ickyness to find the sword of doom! We must stop him!)

The trick is to communicate this before you play; you want some cool adventures (modules) and they want to do their own things, so you compromise and everyone wins.

 

[Raven Crowking]

I just wanted to give the players more freedom and more illusion of freedom, but it ended up just ambiguity and they didn't know where to go.

If that's the case, all you need to do is make the choices clearer.


RC

 

[Janx]

Well, it's no secret that I hate the term sandboxing. And I think railroading as a term is used too loosely.

One the question of: Are your players having fun?

Consider that the GM is also a player. A better question is "Is Everyone having fun?"

On the point of you are not allowed to make plot hooks, I say hogwash. The GM determines what NPCs are doing. And if one of them has a problem, he's free to make it the PCs problem, just as the PCs are free to make problems for the NPCs. Running an environment where ONLY the PCs initiate things and calling it a sandbox is not, it's a wishfulfillment station.

In the same vein, many players are LOUSY at initiating anything. They work best when presented with a problem. You've got to present them with problems and opportunities, and that may get them involved to taking action. Many players don't just decide to go somewhere and explore someplace. If it ain't broke, don't fix it. Why go to a dungeon, if there's no problem with it?

An important element to any "plot hook" is to plan "what happens if the PCs fail or do nothing". A consequence. It's often the threat of consequence that drives players to action. I may not be keen on yet another rescue a baby from the evil cultists, but I am keen on not having their dead god resurrected and eating the people who own my favorite restaurant. If the players see a problem, and don't help, either somebody else will (might be consider deus ex machina) or the problem gets worse. You've got to be constant about that. If you have somebody else fix it, you're undermining the PCs as protagonists, as well as teaching the players that somebody else will clean up the messes.

In the same line, if you create too many plot hooks (problems), the players can realistically only solve 1 at a time. This means the others will fail. This will drastically change your campaign world, especially as the problems get bigger per party level. Because of this, to run a realistic sandbox, if too many problems exist (GM or PC initiated), many of them should get worse, in a way that threatens the campaign stability. The solution is to NOT create too many problems at once. It won't help you run a good game.

The solution to fix that is to introduce only a few simultaneous problems, and make them interesting and relevant to the players. This can be taken to mean plot hooks the GM creates, or projects the players decide to undertake. If the players are working on taking over an old castle, don't introduce other problems, outside of that. Because you force them to choose between the castle project failing, or the new problem.

In that light, my GMing style is such that for a 1st game with new PCs in a new campaign world, I write an adventure that I think would appeal to the PCs as I understand them. I expect the PCs to bite the plot hook. They're first level. They are NOT going to be taking major steps towards their "where I see myself in 20 levels" goals. At the end of the first game, the players will have met NPCs of the area, made some friends, and gotten a sense of the world. At the end of that session, is when I ask, "what do your PCs intend to do for the next session?" I then write the next adventure, based on that input. So if the party doesn't bite the hook at the 2nd session, they are violating what they told me they wanted to do.

The result of this style:
I only write material I need for that session
The players pursue goals they want to pursue
I'm not stuck winging it for the night, because the party keeps doing random things

 

[Beginning of the End]

I think the pacing is the biggest problem with no railroading. I like to have my balance of combat and non-combat: "The party will have a discussion over what to do with the Black Pearl. Then something will attack. Then some role-playing, then another attack. Then some investigation, then the module proper."

When you're not railroading, you're largely (but not entirely) passing the responsibility for pacing over to the players.

I consider this quite liberating, because it means that I don't have to work so hard to provide whatever gaming experience the players are in the mood for that evening: I don't have to try to read their minds. I just let them do whatever they want with the expectation that they'll do what they want to do.

There a couple of potential speed-bumps here:

(1) The players choose to do something that you, as the DM, don't enjoy doing. I haven't had this problem, but it sounds like this is your primary objection: You want to be running some combat and the players aren't obliging.

There's a couple of options here: First, load the sandbox with stuff you like and don't include stuff you don't like. Second, talk to your players about the issue.

(2) The players choose to do something that they don't actually enjoy doing. This seems outrageous, but it's also something that happens frequently with groups who are used to being railroaded. They've been "trained" to see a plot hook and bite it (because otherwise the plot hook will bite them), so when they see the plot hook rich environment of your sandbox they go biting at everything.

If this is a problem, the solution is pretty similar: Talk to them about expectations and the difference in play-style. Make it clear that you don't actually have a preference about what they choose to do -- if they choose to bypass a hook, that's OK.

"Do what you think will be fun shall be the whole of the law" is one of the fundamental commandments of sandboxing.

2) I tried to get them to stay in town by having the townsfolk urge them both to stay and go, but have the mayor offer the fighter sweet armor to stay and be the sheriff. Result, they manage to get on the beach in a boat leaving town before I am able to get the mayor in contact with them, making the railroading obvious instead of a subtle nudge.

Either you're railroading or you're not. Trying to "not railroad" except "when it's really important" is still railroading. Only it's likely to result in awkward and transparent railroading of the type you describe here.

If you're actually interested in giving sandbox gaming a try, then give sandbox gaming a try. Otherwise you're more likely to end up with all the worst elements of both sandboxing and railroading.

3) The party is splitting up over individual motivations -- not fighting, just not the kind of smooth, fun teamwork we get when we just assume motivations and go kick butt. I can't use "you're the sheriff" for adventure hooks because the other two have no interest in being hired on as deputies.

Is this a problem or an opportunity? Let it play out and see what happens. Try providing multiple hooks to the same scenario.

If it's really a problem, then it's time to have another of those chats about expectations. Which can be as simple as: "I don't care what decision you guys make, but you all have to stick together because I can't handle running a split party."

I recommend this essay: Don't Prep Plots

The author is my DM. He runs an awesome campaign.

On the point of you are not allowed to make plot hooks, I say hogwash. The GM determines what NPCs are doing. And if one of them has a problem, he's free to make it the PCs problem, just as the PCs are free to make problems for the NPCs. Running an environment where ONLY the PCs initiate things and calling it a sandbox is not, it's a wishfulfillment station.

This is utter truth. A sandbox is not devoid of plot hooks. To the contrary: A properly run sandbox is a plot hook rich environment.

In 3.5, every time you use a monster, include its stat block in a Word file. Print that Word file out and keep it at the table, with its (eventually) hundreds of stat blocks. Where monsters have special abilities that require special info, put that info into the stat block.

This.

One of the things 4th Edition got 100% right was putting every single piece of information you need to run a monster right in the stat block. But there's no reason why 3rd Edition stat blocks can't be designed the same way: I've been running them that way since 1999 (I was in a playtest group).

That's really annoying that you can't run a monster straight out of the book in 3.5.

Not really. It's no more difficult to run a roper in 3.5 than it is to run a roper in AD&D. If anything it's easier to run in 3.5.