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

[SableWyvern]

For my part, I am trying to get clarity.

For instance, are you assuming some version of a D&D rule for when to roll?

No. I'm assuming I call for a roll when I think it's necessary. I grew up on Rolemaster, and the nature of open ended rolls resulted in a playstyle where rolling was generally kept to minimum, used when the outcome is genuinely in doubt or the consequences for failure are significant and and the roll actually matters.

Are you saying that a rule which only called for a roll when both success and failure are interesting would be an implausible rule?

I don't know if I would describe such a rule as implausible, but it is a rule leaves me with no means to determine success and failure when the result is in doubt, but only one of success or failure is particularly interesting.

Note above, while significant consequences are one guildeline, the outcome being genuinely in doubt is another, and I see no need to wait until both success and failure come with plausible*, interesting outcomes.

*Plausible in this case needs to be considered not just in the case of "in this instance" but also, "in general, if this situation were to occur time after time, would it make sense for there to always be an interesting outcome?"

 

[Lanefan]

How is there any other interpretation of the statement, "I say the default should be to play them out unless the players say not to", other than we play out all the scenes unless the players say not to?

And oftentimes the players' default is to say not to. The option to say yes, however, is always open to them; I-as-DM won't shut it down should they choose to exercise said option.

 

[Lanefan]

Is there a fail-forward on that roll too? Does it matter that the whole thing's retroactive based on the scenario you described?

Most of the time I prefer they declare they'll be herb-hunting before they start the trip; if nothing else, herb-hunting slows travel speeds considerably.

 

[Lanefan]

Time’s mutable in an RPG. Things like that are an easy fix. The kind I imagine most of us have done in a more casual way.

Usually it’d be something like “Ah crap… I meant to buy more arrows when we were in town… do you mind if I spend the money and add some arrows now?”

I mean, I expect that kind of thing happens in most peoples’ games at some point. And it works just fine without reality collapsing or anything. There’s no reason you can’t simply add it to the tools at your disposal as GM.

Things like that happen; but also sometimes people do legitimately forget things. To reflect that, in relatively trivial cases like this what I usually do is get them to roll under wisdom. Make it, and you remembered to pick up your arrows; blow it, and you forgot to.

 

[Lanefan]

I don't understand what you mean.

The rule I stated - roll the dice only when either outcome of a roll is interesting - isn't "unrealistic". There are RPGs that adopt it, or something like it, as their principal rule for when to roll the dice.

Just because some RPGs do it that way doesn't make it any more sensible.

Rolling the dice when one outcome is interesting while the other outcome is the status quo is more than good enough. You're rollng to see whether that interesting outcome occurs or not (and oftentimes without fully knowing what that "interesting" outcome might consist of).

Seartch for a secret door. Fail = status quo. Succeed = any one of a wide variance of outcomes of a wide variance of interest levels. You might find an empty compartment similar to a safe, long since cleaned out. You might find treasure beyond your wildest dreams. You might release poison gas that auto-kills anyone nearby. You might find a dusty passage leading into darkness. You might find you're back at somewhere you've already been, arriving from a different direction. Etc., etc.

 

[EzekielRaiden]

And oftentimes the players' default is to say not to. The option to say yes, however, is always open to them; I-as-DM won't shut it down should they choose to exercise said option.

But, again, those two things are not the same!

"We always do X, unless someone says no" is ENTIRELY different from "We can always do X, if someone wants to".

Like...let me give this a physical implementation. "We always do X, unless someone says no" vs "We can always do X, if someone wants to" in the context of...let's say an amusement park. Disneyland, if you prefer. Let X be "ride a ride when we pass it".

The former is, "We always ride a ride when we pass it, unless someone says no" means you WILL ride, definitely each and every time, unless someone speaks up and declines. The park-goers who follow this rule must ride the ride, unless someone says not to. This dose not, in any way, imply that there's any negative connotation to say no. But it does mean that someone actually has to nix it, otherwise it WILL happen, regardless of what people feel about it.

The latter is, "We can always ride a ride when we pass it, if someone expresses interest." That creates no obligation, nor does it require anyone to "speak up" to make it not happen. The option is left open, if anyone feels like pursuing it. There is no specific expectation that the group will do so, but it requires interest on someone's part in order for it to happen. If nobody feels particularly interested, the group defaults to not taking the ride.

That's the fundamental thing I was arguing against here. You explicitly said the former thing originally. There was no "saying yes". Yes was automatic, unless vetoed by an active "no". The other is a presumption that, if no one voices interest, nobody is interested. Interest has to be expressed. It is not presumed.

If your standard is, "There is always the option if someone expresses interest, but I won't force it to happen", then that is not and cannot be "we will always take that option unless someone says they don't want to." I was arguing against the former specifically because it is forcing things to happen regardless of whether the players are interested. That is what was originally said. It will happen, unless someone says no.

If you meant "It CAN happen, as long as anyone says 'yes'," then you communicated that exceptionally poorly before. I accept that that's what you're saying now, but it's pretty annoying to have people act like I'm somehow being weird or unreasonable for reading "I say the default should be to play them out unless the players say not to" as meaning...y'know, that the players have to say not to!

 

[SableWyvern]

I have been giving some thought to how Blades would have worked for me if we had only called for a roll if we felt there were plausible and interesting outcomes for all possible levels of success or failure.

And it would absolutely have broken the game for me. I was OK with the results not always being as plausible as I might normally expect; that's what we'd bought into and we had a lot of fun with it. One of my players found it extremely fun just to see what kind of complications I would throw at them -- he'd love making a roll when he couldn't imagine how there was any way it could go horribly wrong, and then having me pull something out of nowhere. And it would be plausible enough, but only in the context of the over-the-top, all-action-all-the-time, keep digging until you're out the other side, game we were running.

If we'd put a stop to the any roll occurring until the normal threshold of plausibility was met for all possible check results, the game would literally have just come to halt.

So, yeah. There is an extreme edge case where you might be able to satisfy my normal plausibility threshold, but not while I still consider the game playable.

Edit to add: On top of that, when the dice told me I had to come up with a complication, it was often challenging. I found Blades easier than any game I've ever run to literally run on zero prep. Show up, no clue what would happen -- not the faintest -- and it all just goes off perfectly. But actually running the session was more mentally exhausting than any game I'd ever run, because I was always being forced to improvise. I had no choice. The dice said "Complication" and I was honouring that, so complication there must be. Had I been in a position to have to decide before the dice, "Are there meaningful complications to be had here for every possible outcome," that's a lot more mental overhead, as I have to assess every single, not just one. If someone had me at gunpoint, forcing me to stick to, "no roll unless there are interesting outcomes for every possibility," I would almost certainly have just vetoed most of the situations where players were looking to make rolls. "Nope, not enough interesting outcomes, no roll."

 

[Hussar]

Are they? You've looked at every sandbox ever made, including the homebrew ones, and determined this is the case all the time?

It's funny. You: sandboxes have to be done this way. Me: No they don't. You: THAT'S NOT HOW SANDBOXES ARE SUPPOSED TO GO!!!

Maybe you should stop One True Scotsmanning here and just accept that your particular view on this is not universal.

LOL.

No. I haven't looked at every single sandbox ever made. That's true.

Just every single one that's been brought up as an example in this thread.

Show me a sandbox for D&D where the level of the characters is not built into the sandbox.

 

[Hussar]

You don't get to tell us how we get to run our games, man. We are telling you that you can run into CR 20 stuff. Remember that running and/or non-combat options are available in games. If the only thing that could happen would be combat, then sure, we would have to telegraph powerful things.

And wandering monsters aren't telegraphed encounters in any case. They are monsters that wander into you(often while you are camped), and can be waaaaay outside your level range.

And, most often, they aren'T outside your level range. And, in cases where they are, there is almost a guaranteed exit possiblity. Either the massively overpowering encounter will "just talk" in which case, level isn't important, or, the baddie won't chase if you run away. On and on and on.

But, sure, you routinely have your low level characters run into CR 20 stuff. As in once a session? Once a level? Once in a campaign? Just how often does this happen?

 

[clearstream]

...if you wanna defend crappy design, be my guest.

If I read this correctly it relates to the debate on fail-forward where the "crappy design" is the simple fail mechanic not handling a 'single point of failure' edge case. It seems to assume an intended line of play that cannot be pursued, somewhat in contradiction to the general premises of sandbox but I suppose to be arising in some modes of play that could include narrativism were a line of play meaningful to a player blocked.

The debate seems somewhat mired to my reading, however the reference here to "design" suggests to me looking at the problem through the lense of kinds of design innovation. One kind is that which does a job I want to do without changing my purpose (I still want to do exactly that job.) An example might be the evolution from THACO to d20 roll over AC. Another kind changes the job I want to do in a way that I value (I now want to do a different job, whether that be mainly similar or greatly dissimilar.)

To my reading, some participants have a job they want to do and see fail forward as changing that job to do in ways they do not value. Non-adoption of fail forward for them is not tolerating crappy design, it's non-adoption of an innovation that doesn't fit their purposes (and there seems ample testimony that it goes against them.) Through a 'satisfying-jobs-to-do' lense, simple fail is "crappy design" only if fail-forward improves on the jobs simple fail satisfies without changing them. (Including that the point-of-failure edge case would need to matter to those with those jobs to do.)

The reason I assay this line of reasoning is that I repeatedly observe that game mechanics can be improved. They may be improved severally (refinements on individual mechanics), in the ways they are structured and work together collectively (the way they are assembled into procedures, are invoked and interact), and in contexts created for them by principles (how they are interpreted, given weight, subject to discretion, extrapolated from, assigned, and so on.) So it matters to me whether simple fail has been straightforwardly improved upon by the strictly-better fail-forward, or whether fail-forward changes the job to do.