Interesting.
If that’s the frame of mind, what is the relationship between “arbitrariness/unintended consequences” and acceptance of die rolls as valid content-generators?
There’s an extent to which players have some effect on a die roll and maybe do what they can to improve their odds. Could that be seen as less arbitrary? Also, if the stakes of any pivot point are known to the players (who, let’s presume agree to proceed in acceptance of all potential outcomes), does the fact of their knowledge and acceptance of weirdo possibilities make those outcomes more legit?
per one of the examples, if you’re playing as Son of Rorke and you’ve got these expectations of making a name for yourself and you still choose (knowingly) to put Son of Rorke in a situation whereby he’s merked by Klarg, is that a failure of any kind? Did the system fail? Did the player? Is there tension between expectation of outcome and actual outcome? Is it all arbitrary? And can it actually be arbitrary if you know ahead of time that there’s like a 10% chance of an undesirable outcome so you go on ahead and do it.
Are “that sucks, but that’s how it goes” moments game system failures? Agenda failures? Or part of the game that we should reasonably expect to have to roll with?
I'm going to answer your questions by talking about specific games and game situations (its difficult to answer your questions in the abstract).
1) Dogs in the Vineyard is my favorite TTRPG of all time. But that doesn't mean I'm unaware of how a particular design choice affects play. Similar to 5e, there is a "bounded accuracy" component of play that creates an extreme disincentive for the Dogs (these are gun-toting paladins meting out justice and upholding the Faith in a Wild West That Never Was) to escalate to guns against superior numbers (I'm not even talking profound superior numbers).
This can create a problem for play in my opinion. I want my Dogs to be bold, aggressive advocates of the doctrine. Unfortunately, the orthodox dice pool mechanics can disincentivize (say) a gunfight with cattle rustlers and sin-brokers when the Dogs are outnumbered. Now there are already disincentives to get into gunfights in the game and those disincentives will crystalize "who your character is...what they're willing to fight and die for." It doesn't need any more.
However, the game is designed so beautifully that there is a trivial answer for this. Lower the GM's dicepool when facing superior numbers. You can still make fights insurmountable after a very reasonable point. So I've done that. When GMing Dogs, this is the way I run it.
2) I've run probably just south of 50 games of Dungeon World. The most recent one I've taken on features
@prabe and his wife. In the second session of play, things almost spiraled as his wife's PC jumped into this roiling cauldron of an abyss at the bottom of gorge. Poltergeist-like, it seemed to be spewing spirits into the material world. Due to a snowballing effect of action resolution gone south, the game almost abruptly ended. His wife made her Last Breath move and got a 10+ and this fundamentally changed the gamestate, the landscape of play, and the trajectory of the story.
Now like I've said above...I have run TONS of DW (and AW...and a few other PBtA games). No one can convince me that DW in particular wouldn't be a better game with FitD Position and Effect integrated into it and either a functional iteration of FitD's Stress/Complication Resist System or simply "Hold 1 at the beginning of ever session...spend Hold to turn a 6- result into a 7-9 result or turn a 7-9 result into a 10+ result."
My guess is in
@prabe 's limited play he may very well agree with that latter amendment to DW's system (even if he's agnostic on the former because he hasn't played a FitD game)!
3) In one of my more recent Blades games, the Crew was a series of Con Men...the game was kind of like The Sting meets Rounders. Now Blades has all kinds of robust conflict resolution mechanics. However, that doesn't translate to being even functional, let alone good for even
Genre Texas Hold 'Em. Forget about emulating the action tactical and strategic decision-points of Texas Hold Em...just the tropes and inhabiting the general feel of a Hold 'Em table.
So I had to iterate twice in order to make Genre Hold 'Em work out of Blades action and conflict resolution system and resource economy. The first time was just north of
meh. The second time was better but well south of
alriiiiiiiiiiiiiight. I'm not pleased with the present iteration and I'm not convinced at this point (though, I could be convinced by myself or someone else who conceives of and executes something brilliantly) that Blades can produce the sort of Genre Hold 'Em play that I personally would find acceptable (acceptable is a high bar for me). Meanwhile (circling back to 1), Dogs conflict system is fundamentally poker! It is built to be Genre Hold 'Em! But I can't just bolt that onto Blades. It won't work. At least not satisfyingly.
So to drive this all home. If a system doesn't reliably (reliably here means pretty nearly without fail) and coherently:
* produce the genre conflicts + pressure points + fallout that it intends...
and/or
* if the action resolution mechanics/resource economy + the tactical/strategic/thematic overhead doesn't facilitate the kind of decision-points and volition for the participants which are vital to coherent story outputs...
...then yeah, system is where you should be looking...