D&D 5E Pros and cons of a sandbox game, and what to do about them?

[Xetheral]

There have been a lot of good high-level suggestions so far. Let me add a few more-specific ones:

1. Start the campaign in combat. This gives players an immediate chance to show off their new characters' abilities, and the aftermath will provide some context to help the party decide what to do next (e.g. pursue enemies, flee more enemies, report an incursion, investigate, etc). Make it a relatively easy, medium-long combat, and don't include any NPC allies--the first fight needs to be about the PCs.
2. When your players ask questions about the game world, frequently include an extraneous detail in your answer, and write that detail down. If your players follow up on the detail, great, you've got a new plot hook. Keep a list of the ones they don't follow up on and have about half be significant (but NOT detrimental) later. One of the keys to a making a sandbox seem alive is plot-weaving (i.e. running multiple concurrant plots), and whether you're a master at plot-weaving or a newbie, a list of already-established details for later weaving is a godsend.
3. When players visit a location they've been to before, make sure it has changed in some way, big or small. This makes the world seem dynamic. Occasionally have the change be a plausible consequence of the PCs' actions from the first time they were there. If it's a location you're fairly confident will be recurring, come up with the changes before the first visit, and foreshadow them. Adjust the planned changes to the PCs' actions, if any.
4. Have plot hooks the PCs learn about occasionally expire if not pursued promptly, rather than appear to just wait for the PCs' arrival. A sandbox gives you more opportunity than a linear adventure can to have the PCs outright fail at a particular plot thread, so use it: have villages be sacked, hostages killed, and treasures already looted. (Caveat: never do this punitively. The failue is a simply a logical part of the evolution of the world, not a punishment for the players' choices.)
5. Have some recurring NPCs who are neither allies nor enemies that move about the world according to their own motives--don't tie every unimportant NPC to a fixed location. In a sandbox it is particulary important that the players and the villains aren't the only ones who move about. If the PCs decide on their own to make enemies or allies of the recurring characters, that's perfect, but include enough such recurring characters that they can't co-opt/antagonize them all (except maybe for the campaign's finale).
6. In a sandbox, geography is more than merely background... know how long it takes to get from one place to another, and be consistent, so that players can make educated choices about the opportunity costs of travel. Relatedly, don't make everything interesting too far apart (or have too many random encounters), or the cost of travel will be too high. Ideally, have multiple competing routes to get from one location to another, each with different costs and benefits and risks.
7. If the players' ideas about the solutions to a mystery or plot would be better (i.e. more fun/awesome/badass yet still plausible) than what you had in mind, invisibly adopt their ideas instead. Make sure being (retroactively) right gives them some appropriate advantage though--changing your planned solution to screw the players is unlikely to be fun for anyone unless the idea you're co-opting is truly epic.
8. Pacing can be a problem in a sandbox. Have a list of upcoming events (some random, some plot-relevant) generic enough to use anywhere and insert them early if the game starts to bog down. Reprisals from enemies, violent fueds between two groups of NPCs, receiving messages from the PCs' contacts, and civil unrest are all great options that can be used almost anywhere. Such tools are even more useful when foreshadowed.

I hope those help!

 

[77IM]

Step 1: Treasure map!
Step 2: MORE treasure maps!

 

[happyhermit]

Honestly, sandboxes are why I came back to D&D and the hobby in general. I will still play or run different types of games and I have fun, but they have to compete with video games, board games, storytelling games, and other RPGs. The itch that is scratched for me when running or playing in a world where the PCs have the freedom to truly attempt whatever they want is something that I just can't get anywhere else. It is pretty extreme for me, honestly, I don't even want a "main plot" most of the time. Just give me an interesting world, interesting characters (PC and NPC) and some semi-realistic conflict resolution mechanics for things like combat (including someone to decide whether or not some obscure thing we just thought of actually works) and I am going to have a great time. I really don't play or run to "tell a story" I do it to have the characters interact with the world and each other.

There has been good stuff mentioned so far and this will probably be rather incoherent but anyways.

Some PCs are built for sandboxes, some are... more difficult.
Heroic young noble wanting to prove themselves by defeating brigands and monsters to make the lands safe, SWEET, easy to hook, can add some politics, can have info about relative safety of different roads or areas, etc.
Lazy or unmotivated person, can be hilarious or extremely annoying. No more than one at a time in a party unless they have a real good reason to keep up.
A curious person interested in learning about monsters and/or ancient ruins, JACKPOT, great way to seed info about difficulty of an encounter, easy to hook.

Consider starting in medias res, the party is all in one place, something has already happened or is in the middle of happening. Ie; caravan guards, refugees, shipwrecked, the inn just burned down. Starting with just the party together makes things easiest,

I like to start somewhere a bit remote, mostly personal preference but it definitely makes it easier for me as a GM to start in a little village, with maybe only one road through town so I have a good idea of where they are going but don't have to make that clear or push them too hard in any particular direction. It does also seem to work well with new players who often get overwhelmed when starting out in or near a large city.

I like to have some "dungeons" (towers, ruins, whatever) that might in and of themselves be rather "linear" by their inherent nature, but not in any sort of contrived sense. I have even on occasion gone so far as to have a party get trapped in such a place where there was only one reasonable (and maybe a few unreasonable) ways to proceed in the general sense. OMG, but he said he loves sandboxes so much!?! Yup, logical in-game situations don't bother me though, unless they became repetitive. They do make things easy on the GM, and can be a nice change of pace in a very open worlds.

I like come up with a starting concept, decide more or less what type of campaign it will be and communicate that with the potential players, then have them pitch character concepts to me. I let them know that the game is going to follow the "party" so if their character wants to just open up a bait shop on the docks and kick up their feet, that's perfectly fine, but they shouldn't expect the game to follow them if the rest of the PCs keep adventuring. They can roll up another PC if they want. They also shouldn't expect the party to stick with them if it wouldn't make sense for the characters.

 

[clearstream]

I plan to shamelessly take advantage of this community's wisdom and knowledge, so I have some basic questions with complicated answers. I will be running a game for new players, and after some time teaching them the rules I plan to kick it off in my own world as a sandbox game.

What are the pros and cons of a sandbox game?

How can I maximize the pros?

How can I minimize the cons?

Resting is a con. With more time spent travelling or on downtime between encounters you need rests to scale appropriately. I'm about to start using

Short Rest
A short rest is a period of downtime, at least 6 hours long. Along with the standard benefits of a short rest, taking a short rest allows characters to reduce exhaustion, or examine or attune an item, or change prepared spells or perform a downtime activity.

Long Rest
A long rest is a period of extended downtime, at least 2 days long. Along with the standard benefits of a long rest, each day of a long rest allows characters to reduce exhaustion. So long as a character is not exhausted, each day allows them to level up, or examine or attune an item, or change prepared spells or perform a downtime activity. After taking a long rest, a character can’t benefit from another for 2 days.

 

[clearstream]

I plan to shamelessly take advantage of this community's wisdom and knowledge, so I have some basic questions with complicated answers. I will be running a game for new players, and after some time teaching them the rules I plan to kick it off in my own world as a sandbox game.

What are the pros and cons of a sandbox game?

How can I maximize the pros?

How can I minimize the cons?

Pros - you can go in any direction! With that in mind, motives and means are more important to track than set pieces.

 

[hastur_nz]

I've played in a 'pure sandbox' that was pure DM homebrew, and it was a disaster - he put in heaps of time prepping heaps of stuff, and annoyed us so much with all his house-rules, restrictions, extra rules and so on, that after one session we all quit. We were all experienced players and/or DM's, not newbies. But the moral of the story was that our DM got into his head more and more ideas of what HE wanted the game to be and how it was going to be fun, while WE didn't agree, but he forged on regardless until it blew up in his face - so be mindful of your players and what they are interested in and will have fun with, unless you want to play the game by yourself.

One other piece of advice I've not seen - be mindful of the number of players. I forgot this one myself a couple of weeks ago, when my game went up to 6 players and I just happened to have a pretty open-ended kind of urban adventure that provided lots of the classic 'sandbox' elements... and the players wanted to try and do everything on offer every day, including spending an age discussion why, what, how, splitting up, etc etc. The more players you have, the more I think a decent set of "rails" is a real bonus. The most focused games I've played in, are where there's only three of us at the table, then we can easily decide what options to take and get on with it, and everyone still gets a decent amount of time in the spotlight. With lots of players, it's too easy for one or more people to need to run in their own direction, find their personal spotlight, etc, and if you present a heap of options chances are you'll get a group which wants to try and do them all, or takes forever deciding which one if any they might do. So with lots of players, limit their visible options and help them focus on doing one thing from a limited set of options, or even better give the illusion of choice where the choice is pretty obvious but they feel like it's their choice not yours via some NPC (e.g. show them, don't tell them).

 

[clearstream]

I find "sandbox" to be a term to which many have positive associations but a campaign style for which many have a strong dislike. One of the first questions I was asked when I kicked off my last campaign was how "sandboxy" it would be. I asked the player what he meant by that and he didn't even have an answer.

A sandbox campaign is one containing places, creatures, items and events systematically connected through paying regard to motives and means. The components of a sandbox are "in flight": they don't wait around for the PCs to trigger them (although the PCs can trigger a series of events, too). Many people describe that as a "living world" because the world is conceived to be alive. Events are playing out even when PCs aren't present to see them. For example, creature X desires item Y because they need it to have seize control of place Q, feeding their underlying love of power. Item Y is to be found in dungeon Z. Players are free agents in this web, typically disrupting it. Most sandboxes are experienced as an area with nodes that are loosely connected by narrative hooks producing various kinds of quests.

I've played in (and run) campaigns where the PCs basically stood around trying to amuse themselves because there was no obvious story or plot. I've played (and run) campaigns where once they realized that the group was in control of it's adventures and destiny, each of the players became intensely and personally devoted - to courses of action often diametrically opposed to one another.

I'm going to advance a "no true Scotsman" argument here. You've played in unsuccessful attempts at sandboxes. It sounds like they lacked the compelling regard to motives and means that drives one. Because a sandbox is in theory so vast, it takes good intuition to start in the right place and detail what is most needed, and will engage the players. For me Runequest nailed sandbox best, and Griffin Mountain was one of the best resources they published.

My personal experience suggests little value for "sandbox" style campaign, is the takeaway here.

Because this statement is so negative, I'll respond with a strong positive. DM'd convincingly, a strong sandbox is the most rewarding kind of campaign for players.

As a result, my compromise is to provide a living world as a background for the characters. I weave subplots and stories that wax and wane in importance but are always present, with recurring NPCs, to give the players a sense of a world that is large and involved.

And I run published adventures, which are often glorified (or not so glorified) dungeon crawls.

I read ahead and I have a general idea of what I'm going to run in what sequence, and that allows me to introduce important NPCs and organizations and cities well ahead of time, so when they become relevant it seems like a natural occurrence in a living world.

My players love it. I find it rewarding. I can tie into their backgrounds and interests and help develop their characters and push forward with their goals without ever having anybody feel like they have no control OR feel like they have no direction. Change a tribe in the adventure to match one the player invented or mentioned in their backgrounds. Change a magic item to have a history closely tied to a character's past. Change an adventure's location to match tales you've told and rumors you've set.

That's how I've been running games for over 20 years. It's a nice balance of basically everything my players and I enjoy about fantasy, stories, and RPGs. It's a learning curve for sure: 20 years ago my reach definitely exceeded my grasp. But you can make it work, and it's amazing when it does.

It sounds like you're just about running a sandbox, but because of your negative experiences with bad campaigns that were labelled "sandbox" you shy away from calling it that. Sandboxes are a stretch to run because as you identify you have to go beyond pre-scripted scenarios. Foreshadowing and recurrence are key tools. For me, running creatures according to their motives and means, rather than simply as cutouts that fight and die in one scene, is essential. For me it's a more honest way to DM and requires moral fortitude. For example, the evil Knight will take time to destroy an organ of a downed PC, when her compelling motives suggest that she needs to make sure that PC stays down. Or she might flee rather than die, when her compelling motive goes beyond simply appearing in room 15.

One thing I've liked about some of the published adventures is that they make a nod toward a sandbox, providing some of the tools needed. Unfortunately they are still learning to write them, or perhaps choose not to because they don't want them to be too difficult to DM. For example, in OOTA we see the drow families Mizzrym (6th house), Duskryn (9th house) and Vandree (16th) house represented. Appropriately there is enmity between a Duskryn and a Mizzrym priestess. The PCs have an intriguing, perhaps even key role that they could play in that. We also find a Duskryn in Blingdenstone, left behind due to a horrible transformation. Duskryn held the charter to loot Blingdenstone after the siege. However, we don't see any detail on those houses in the published adventure. A rich source of living stories is neglected.

 

[clearstream]

I've played in a 'pure sandbox' that was pure DM homebrew, and it was a disaster - he put in heaps of time prepping heaps of stuff, and annoyed us so much with all his house-rules, restrictions, extra rules and so on, that after one session we all quit. We were all experienced players and/or DM's, not newbies. But the moral of the story was that our DM got into his head more and more ideas of what HE wanted the game to be and how it was going to be fun, while WE didn't agree, but he forged on regardless until it blew up in his face - so be mindful of your players and what they are interested in and will have fun with, unless you want to play the game by yourself.

It's generally best to start light, because you don't know what direction things are going to unfold in. Often an area map, a sense of what is where, a few key personalities and a starting mcguffin will do. However, you should also be prepared over time to evolve a lot of house-rules, restrictions, extra rules and so on. But that can emerge naturally rather than being clamped in place at the start.

One other piece of advice I've not seen - be mindful of the number of players. I forgot this one myself a couple of weeks ago, when my game went up to 6 players and I just happened to have a pretty open-ended kind of urban adventure that provided lots of the classic 'sandbox' elements... and the players wanted to try and do everything on offer every day, including spending an age discussion why, what, how, splitting up, etc etc. The more players you have, the more I think a decent set of "rails" is a real bonus. The most focused games I've played in, are where there's only three of us at the table, then we can easily decide what options to take and get on with it, and everyone still gets a decent amount of time in the spotlight. With lots of players, it's too easy for one or more people to need to run in their own direction, find their personal spotlight, etc, and if you present a heap of options chances are you'll get a group which wants to try and do them all, or takes forever deciding which one if any they might do. So with lots of players, limit their visible options and help them focus on doing one thing from a limited set of options, or even better give the illusion of choice where the choice is pretty obvious but they feel like it's their choice not yours via some NPC (e.g. show them, don't tell them).

This is very true. Each player needs more screen time in a sandbox because the situation is typically more ambiguous. With more going on, the DM needs time to parse their choices and actions. For a "diceless" sandbox I found 2 players ideal. For a sandbox with rules e.g. D&D, 3-4 players is okay. 6 players is tough, unless they aren't all playing at the same time.