D&D General [rant]The conservatism of D&D fans is exhausting.

[Thomas Shey]

You may expect more of them, I don't.

Then we don't have much to say about it. "To those for whom much is given, much is expected." That applies to power and control as much as anything else.

 

[Thomas Shey]

I trust my GM to do the best they can. I don't expect them to be perfect or never make a mistake.

I don't either, but the more opportunities to do so, the worse the experience is going to be in some cases, and I don't see any reason to do so in areas where its not necessary.

 

[Thomas Shey]

at least theoretically, yes, how would you figure out how corruptible a guard is without trying to bribe them?

Ask around? its not like that's liable to be a big secret if its at all extensive; word gets around.

 

[pemerton]

I’ve gone back to Gygax’s Player’s Handbook (1978), specifically the section titled "Successful Adventures." While I do not claim that Gygax supported sandbox campaigning in the modern sense, his advice forms the historical foundation for some of the techniques I (and others) use to manage sandbox campaigns today.

Here are a few key ideas from that section:

• Players are expected to set objectives before play begins.
• Strategic planning and preparation are crucial, including mapping, spell selection, and party coordination.
• The world exists independently of the characters; some threats are avoidable, others are not.
• Retreat, reassessment, and information-gathering are key skills.
• Characters succeed based on in-play decision-making, not story beats or preordained outcomes.

This strongly parallels various aspects of my approach to sandbox campaigns:

• I present a setting that would make sense even without the PCs.
• The players are free to interact with it however they like.
• I adjudicate their actions based on what’s plausible within the context of the setting.
• I do not predetermine outcomes. Players succeed or fail based on their decisions and how they engage with the world.

I actually don't see many parallels in this list.

There is nothing in your Gygax dot points about the dungeon or the setting "making sense". And his dungeons are full of stuff that doesn't make sense, but is part of the shared experience of dungeon-crawling play built up at his table with his play group.

The players being free to interact with things "however they like" isn't something that appears in your Gygax dot points either. Gygax takes that for granted, because it is a basic feature of RPGing.

The Gygax dot points you've listed say nothing about how action declarations are to be adjudicated, and a fortiori say nothing about "plausibility" as the touchstone. Gygax's DMG does talk about action resolution, and outlines a variety of approaches. Off the top of my head, the one case I recall is where he suggests that the GM should say yes to a player suggestion about the local topography of a region, in the context of the player having their PC construct a stronghold, when the player's suggestion is not implausible. Appendix B also talks about plausibility as a constraint on random hex-map generation; but that probably doesn't count as action resolution. (Unless we think of the player's declared action as I move into the area that will be mapped as the next hex, to see what is there).

The only real parallel between the two lists is that outcomes of declared actions are not pre-determined.

Which would not be remarkable, except that over the last few hundred posts of this thread, multiple posters have asserted that the GM pre-determining that an attempt to persuade a NPC to do <this thing? will fail is consistent with sandboxing!

 

[The Firebird]

It has been repeatedly framed that any NPC with strong motivations or immovable positions is evidence of railroading unless the players can predict in advance what will happen.

By whom?

Not by me. Not by @hawkeyefan. I think not even by @EzekielRaiden.

Not of a mood to dig up the quotes, but this was how I understand the argument. Am I wrong about that? If so, what is a more charitable way to state it?

Which would not be remarkable, except that over the last few hundred posts of this thread, multiple posters have asserted that the GM pre-determining that an attempt to persuade a NPC to do <this thing? will fail is consistent with sandboxing!

Indeed, your last statement here seems to me in line with robertsconley's characterization. I'm interpreting "NPCs with immovable positions" in his statement is what you mean by "the GM pre-determining that persuading a NPC will fail".

What am I missing?

 

[pemerton]

Suppose the PCs have no other interest in making the trek to White Plume, but they need Blackrazor because it can be used as a focus for a powerful ritual that can break the siege. Is it now railroading?

It starts to look like it, yes. (Depending on things like how was it determined that Blackrazor is needed for the ritual and *how was it determined that Blackrazor is beneath White Plume Mountain. I read you as presuming that the GM has made these decisions as part of managing the setting/backstory.)

I think the key difference is you're looking for a sandbox which centers the PCs goals and desires.

The players'.

That is what makes a sandbox player-driven, and means that the players exercise control over what play is about.

If it centres the GM's goals and desires, then I don't see that it is a sandbox.

Imo that can hurt verisimilitude because the world does not revolve around the PCs. A sandbox will feel more realistic when there are competing factions and interests which are unrelated to what the PCs want.

The second sentence strikes me as unrelated to the question of how do we establish what play is about?

That is, the presence in the setting of factions and interests unrelated to what the PCs want does not, in itself, bear upon whether or not the players are exercising control over what play is about. How those elements of the setting are drawn into play (or not) of course does matter to this.

As for verisimilitude, what you posit has never been my experience, in multiple decades of running games that - prompted by the discussion in this thread - I would describe as sandboxes.

 

[robertsconley]

You've asserted these things. I don't know what you mean when you say you've established them.

But turning to some of them: Players are allowed to attempt any actions their characters are capable of doesn't tell me anything about sandbox RPGing. That's just a basic feature of all RPGgs, It's part of what characterises them as a game form.

NPC behaviour is rooted in setting logic and plausible motivation and outcomes are adjudicated based on plausibility and context doesn't tell me anything about whether or not the gameworld is verisimilitudinous - it won't be, if the GM doesn't allow for things like Joan of Arc and the knighting of Arthur - nor about whether or not the game is a railroad. Because your ostensible contrast between a fact of the setting and a railroaded plot point doesn't hold, in my view. One standard part of the railroader's repertoire is to establish facts of the setting.

And the comparison to 19th century wargaming is not all that compelling. The referees of 19th century wargames were (ostensibly, at least) experts in warfare teaching others how to make successful decisions in warfare. But the RPG GM is not an expert engaged in a teaching process - unless we mean to say that they are an expert about their setting and they are teaching the players the content of their setting. Which is hardly a description of player-driven RPGing!

When I say I’ve established certain points, I mean I have already presented, justified, and illustrated them in earlier replies. At this stage, we’re circling back to the ground, which has already been covered.

That said, a few brief comments are worth making:

Saying that NPC behavior is rooted in setting logic and that outcomes are adjudicated based on plausibility and context is what makes it a verisimilitudinous approach. That’s the core of maintaining coherence in a fictional world, when outcomes and reactions emerge from consistent internal logic rather than fiat or narrative convenience.

As for “allowing for things like Joan of Arc or the knighting of Arthur,” that is exactly the kind of extraordinary outcome that can emerge organically through play in a sandbox campaign., often through the use of dice. A key skill of the referee is knowing when uncertainty warrants a roll. I discussed this on my blog back in 2017.

From My Axioms of Sandbox Campaigns

Like with the example of the NPCs above, be aware of your bias. At first keep a running count of how you decide things and if you are bias to a particular type of outcome then make a chart to roll on to change things up. Most people can spot consistent patterns especially in social interactions.

Jon Peterson in Playing at the World 2e Vol II has this to say about the importance of understanding the past.

As a result Jon Peterson’s Playing at the World Vol. II offers a helpful framework for understanding the Three Pillars that made D&D and tabletop roleplaying novel. Since sandbox techniques, like all RPG techniques, are part of that history, Peterson’s research is useful for identifying what’s actually novel about sandbox campaigns.

 

[Thomas Shey]

I don't think anyone supports forcing things on others in the context of voluntary leisure activities. To the extent that I don't really see why you feel the need to say this.

I suspect its seeing rules around some actions as doing that while others aren't. This is not a new argument.

 

[Thomas Shey]

IMO you either play a game with mechanical restraints on the GM and are therefore forced to abide by those restraints, or you don't play that game. My preference is option 2.

I'm going to step in on this one. I'd suggest you absolutely do--any set of mechanics whatsoever applies restraints on the GM--its just that there are areas you're comfortable with it, and areas you aren't.

 

[robertsconley]

I actually don't see many parallels in this list.

There is nothing in your Gygax dot points about the dungeon or the setting "making sense". And his dungeons are full of stuff that doesn't make sense, but is part of the shared experience of dungeon-crawling play built up at his table with his play group.

The players being free to interact with things "however they like" isn't something that appears in your Gygax dot points either. Gygax takes that for granted, because it is a basic feature of RPGing.

The Gygax dot points you've listed say nothing about how action declarations are to be adjudicated, and a fortiori say nothing about "plausibility" as the touchstone. Gygax's DMG does talk about action resolution, and outlines a variety of approaches. Off the top of my head, the one case I recall is where he suggests that the GM should say yes to a player suggestion about the local topography of a region, in the context of the player having their PC construct a stronghold, when the player's suggestion is not implausible. Appendix B also talks about plausibility as a constraint on random hex-map generation; but that probably doesn't count as action resolution. (Unless we think of the player's declared action as I move into the area that will be mapped as the next hex, to see what is there).

The only real parallel between the two lists is that outcomes of declared actions are not pre-determined.

Which would not be remarkable, except that over the last few hundred posts of this thread, multiple posters have asserted that the GM pre-determining that an attempt to persuade a NPC to do <this thing? will fail is consistent with sandboxing!

When I summarized Successful Adventures, I was highlighting how Gygax’s guidance aligns with principles I use in running sandbox campaigns. You’re treating my bullet points as if they were direct quotations rather than a synthesis of what Gygax’s procedures imply in actual play. That’s incorrect.

You say Gygax doesn’t talk about plausibility or freedom to act, but his entire structure assumes both. Players are expected to plan, make decisions with incomplete information, act on their own goals, and have the dungeon respond. That only works if the setting behaves in a consistent and understandable way, in other words, if it makes sense. If it didn’t, there’d be no point to the level of preparation, mapping, spell coordination, or strategic decision-making Gygax emphasizes.

The style of play Gygax describes is not narrative-first nor it is a sandbox campaign. But its elements, players choose their objectives, interact with the setting, and face the results, form an important part of sandbox campaigns. The techniques he recommends helped lay the foundation for sandbox campaigns.