Quote from: Benoist;379956And knowledge of your subject matter, and knowledge of your setting, and common sense... OMG! You have to trust your GM knows what he's doing! Scary! :eek:

I ask these questions not as a player with GM-issues but as a GM who's instincts don't always lead him right.

I am a GM, have been for a long time, but I don't always know what I am doing, hence the questions.

Soylent Green, it should be noted that Benoist is the first person in this thread to talk about "sandbox", while you were originally interested in Situational GMing.

I think there's been some unclarity because the approaches, "sandbox" and "situational" can be viewed both in terms of the type of prep they require, and how they're used in the course of play.

While the prep part is fairly clear for "sandbox" and "situational", including how they differ from each other, the use in play isn't. That's what led to some confusion in the other thread and it's causing confusion here.

Basically the idea of prep in both cases is to establish certain facts and frameworks that will either be used directly, or at least to inform judgments made by the GM. The two approaches differ in terms of what gets prepared and what gets left out, with sandbox focusing much more on concrete detail ("thing X is at location Y", "person P has powers Q") and "situational" focusing more on relationships and motivations. A sandbox is almost impossible to conceive without some sort of map of physical space that situates all the setting or scenario elements; a "situation" might be able to do without.

[I should say that Clash may jump in at any point to correct me; I'm trying to explain his style based on little things I've picked up in forum threads and blog posts.]

Now, in play, with either method of prep, you're going to run into questions whose answers aren't spelled out beforehand. Then the GM has to decide how to decide, not only what criteria to use, but also whether there's room for doubt, risk, or whatever you'd use dice for.

With a well-detailed sandbox, the GM may not have to decide on the spur of the moment whether the hospital is close to the police station. But there are certainly some details, or some NPC reactions, or some dynamic situations, that are hard to model deterministically from first principles. So, for example, if the question is "Did Vincenzo send his soldiers to search the warehouse?", the GM might not know for sure. For a "situational" approach, that question might apply but so might the location of the hospital. And unlike the sandbox, maybe the GM would have a better idea of Vincenzo's motivation, whether he'd be more likely to negotiate with the PCs or fight them, etc.

My feeling is that, when there's substantial doubt, it's a good idea to go to the dice--much as Mythic suggests. And the reasons are what we've been talking about: dice help avoid bias and stereotyped patterns, they help generate risk and excitement (largely because they introduce an element of the unknown), and they help model complex situations stochastically.

Nope! You are bang on there, Elliot! As for maps, a Situational GM may have as much use for a relational map as for a physical map. Personally, I like physical maps, but I could easily see someone preferring to run with just a relational map.

-clash

Exactly Elliot, it is a question about deciding how to decide. I like what I am hearing and I think it could be particuarly apt for my group in which we are all experienced GMs, so we all know GM make it up as they go long half of the time and cannot possibly "know" exactly where every NPCs is every single minute and what they are doing.

Yes, I was talking about sandbox GMing, to be clear. For me, a sandbox may be a physical map, just like it might be an organigram of relationships between factions and NPCs. Depends.

Your "Situational GMing" thing is a bit fuzzy, to me.

Quote from: Benoist;380030Yes, I was talking about sandbox GMing, to be clear. For me, a sandbox may be a physical map, just like it might be an organigram of relationships between factions and NPCs. Depends.

Your "Situational GMing" thing is a bit fuzzy, to me.

I think it may be related to math, actually. Some people are geometric, and think best in relationships (Situational) while others are algebraic, and think best in concrete numbers (Sandbox). Of course, there is a spectrum between, as always. I am personally extremely geometric, and have no problem visualizing and mentally manipulating objects in time-space, highly curved n-space, and other weird geometries, but have trouble with simple numbers - they are just instantiations of particular relationships at a particular scale to me. I started working Situationally because it was easier for my mind to deal with.

-clash

Quote from: flyingmice;380024Nope! You are bang on there, Elliot! As for maps, a Situational GM may have as much use for a relational map as for a physical map. Personally, I like physical maps, but I could easily see someone preferring to run with just a relational map.

My general preference is to have a few notes on relationships, a few notes on authority structures, and a list of "contacts, encounters, and confusion".

These don't usually form a relationship map, but the purpose is similar-ish.

A physical map is possibly my favorite game accessory, even when it's not necessary.

Quote from: Soylent Green;380029Exactly Elliot, it is a question about deciding how to decide. I like what I am hearing and I think it could be particuarly apt for my group in which we are all experienced GMs, so we all know GM make it up as they go long half of the time and cannot possibly "know" exactly where every NPCs is every single minute and what they are doing.

Yes, that brings up an even better example. Whether you use a more location-based, concrete prep (what I'm calling "sandbox", Benoist might disagree on terms), or a relational, situational setup, either way it might be obvious that Vincenzo will send his soldiers to the warehouse--but how does the GM know whether they'll be there at the same time as the PCs, and who got there first?

This is the kind of thing that's prime railroading material, depending on what the GM thinks is most fun. He can deny or feed the players clues this way, or he can force a fight, etc. By using the dice you reinforce verisimilitude and increase player agency.

Also, even the most detailed sandbox could be missing some details, not because the GM or setting designer intentionally thought "there is no widget at point X", but simply because they couldn't think of everything and note it down. Therefore, is there a baseball bat behind the counter at the bar? Could be--decide a probability and roll dice. (Again, I like how Mythic does this with a plain-English scale of likelihoods.)

Quote from: Levi Kornelsen;380045My general preference is to have a few notes on relationships, a few notes on authority structures, and a list of "contacts, encounters, and confusion".

These don't usually form a relationship map, but the purpose is similar-ish.

A physical map is possibly my favorite game accessory, even when it's not necessary.

That's much how I do it, right down the the (at times) unnecessary physical map.

-clash

Quote from: Soylent Green;379806Thanks, that makes sense. I can see how it works. I imagine a minority of players might find the rolling of dice to determine the state of reality a bit anti-imersive, but I'd be fine with it.

As long as the ultimate state of the game reality is consistent and has verisimilitude, how it's arrived at is largely irrelevant.  Where the dice can improve reality for me is that they'll often lead the game reality down paths that a GM wouldn't pick and that often feels more realistic to me than the reasoned choices of a person, especially if they are always picking the most likely or obvious choice or are making choices for metagame reasons that have nothing to do with the setting such as story, balance, what they think is fun, and other personal biases.

Quote from: Levi Kornelsen;380045My general preference is to have a few notes on relationships, a few notes on authority structures, and a list of "contacts, encounters, and confusion".

These don't usually form a relationship map, but the purpose is similar-ish.

A physical map is possibly my favorite game accessory, even when it's not necessary.

I do that too, actually. It's weird. I'm talking about Sandbox, but according to your descriptions there, guys, it sounds to me that you'd say I'm conflating sandbox game play and situational GMing into my own way of running things. It's interesting.

Quote from: flyingmice;379762I judge how probable it was that there were witnesses, and if any, were they taken to this particular hospital. Then I roll the dice.

I often use a much more basic roll than that.  I roll whatever dice are handy and high rolls mean that things go in the player's favor while low rolls mean that things go against the player's favor.  The highest roll possible might mean that there was a witness who saw everything at the first hospital that they check while a mediocre roll might suggest one or more witnesses that can only provide a few weak clues and a bad roll might suggest that nobody saw anything useful.  If I need more detail, I'll roll again.  For example, if I wanted detail about the witnesses, a high result on the first roll tells me that someone knows a lot, a low second roll tells me that it's only one person, and a mediocre roll on the third roll tells me that they need to check two or three hospitals to find them.  I don't do any detailed odds.  I just roll on a good to bad scale and make a subjective assessment based on the result.  I often do those rolls closed but sometimes do them open (if, for example, I'm answering a specific question from a player instead of generating behind-the-scenes details).  I can run a whole game like that.  ADDED: And, of course, the situation and what the PCs are doing factor into my assessment of the roll.  If the PCs are in a good position and have covered their bases, what a bad roll means will likely be better than what it might mean if they are in a bad spot with no plan and a good roll will likely be better, too.

Quote from: Elliot Wilen;380049Yes, that brings up an even better example. Whether you use a more location-based, concrete prep (what I'm calling "sandbox", Benoist might disagree on terms), or a relational, situational setup, either way it might be obvious that Vincenzo will send his soldiers to the warehouse--but how does the GM know whether they'll be there at the same time as the PCs, and who got there first?

On this specifically, yeah, I sort of see where I wasn't getting it. For me, a sandbox can be expressed as a physical, concrete representation of areas and locales, or take the shape of more relational material between groups, individuals, who thinks what about whom and so on.

So clearly, to me, you guys are making a distinction between the two terms, sandbox and situational GMing, I wasn't instinctively making, because to me they are in effect the same thing. As I said somewhere else, in my mind, I switch an X on a map for a potential situation or event in the game, switch a dungeon for an investigation, a room for an NPC, a trap for a red herring, and I'm there.

I think I see what you guys are talking about, now.

Quote from: Benoist;380227On this specifically, yeah, I sort of see where I wasn't getting it. For me, a sandbox can be expressed as a physical, concrete representation of areas and locales, or take the shape of more relational material between groups, individuals, who thinks what about whom and so on.

So clearly, to me, you guys are making a distinction between the two terms, sandbox and situational GMing, I wasn't instinctively making, because to me they are in effect the same thing. As I said somewhere else, in my mind, I switch an X on a map for a potential situation or event in the game, switch a dungeon for an investigation, a room for an NPC, a trap for a red herring, and I'm there.

I think I see what you guys are talking about, now.

Like I said before, there is a spectrum between Sandbox and Situation where a lot of GMs find a place they like. I'm pretty far out on the Situational wing, but I should not be taken as representative or typical. Sandbox and Situation are both sets of techniques to avoid railroading players by giving them control of where they go and what they do. They are not in any way conflicting - all Sandbox GMs use at least some Situational techniques to bridge unavoidable holes in their prep, and all Situational GMs use at least some Sandbox techniques to set a - perhaps sketchy - baseline for the setting. Some people use these techniques fairly even-handedly, which sounds like is your case. It's really a difference in emphasis, not kind.

-clash

Quote from: flyingmice;380235Like I said before, there is a spectrum between Sandbox and Situation where a lot of GMs find a place they like. I'm pretty far out on the Situational wing, but I should not be taken as representative or typical. Sandbox and Situation are both sets of techniques to avoid railroading players by giving them control of where they go and what they do. They are not in any way conflicting - all Sandbox GMs use at least some Situational techniques to bridge unavoidable holes in their prep, and all Situational GMs use at least some Sandbox techniques to set a - perhaps sketchy - baseline for the setting. Some people use these techniques fairly even-handedly, which sounds like is your case. It's really a difference in emphasis, not kind.

-clash

Yep. I see. Glad I'm getting it now. The nuance was escaping me.