I think because you aren’t refuting the idea of only rolling when something interesting will happen.
I expect that’s not a rule you play by. You likely call for rolls with some other criteria… like “only roll when the outcome is in doubt”, maybe.
If that’s the rule about rolling dice, then sure, it may be implausible that every outcome would be interesting.
But what people are saying is that if the rule is “don’t roll unless the outcome will be interesting” then it’s not implausible for rolls to produce interesting results… the rule says they must!
Now, perhaps such a rule wouldn’t work for you… you may feel that having interesting things occur as frequently as such a rule would dictate to stretch believability. That’s my reading of what you’ve said… but it’s not entirely clear. I don’t think it stretches believability… it’s probably that rolls will be less frequent, but individually more weighty.
I have played in a fashion where rolls tend to be extremely rare since the 80s or 90s. If there is no combat, it would not be unusual for there to be no more than two or three player-facing rolls in an 8 hour session. When I'm running an OSR game, the only rolls the players are likely make outside of combat are saves.
The "problem" is not that I am used to to running games where there are lots of rolls, and my opinion would change if only I understood that I can run a game that involves fewer rolls at more critical times.
The problem is the philosophy behind the mechanics does does not suit my style of play, in part because I feel it stretches plausability. Even if only truly critical events (however you choose to establish what's critical) use the roll, the process still stretches credulity
for me.
If I really try, I can
almost see how I could accept the die results as plausible, if I really, truly, only called for a roll when there is a clear and obvious case of logical, plausible, interesting outcomes regardless or pass or fail. Note that this is not about
frequency of rolls, it's about the entire
reason for rolling. But at this point, I would see the process of when to call for a die roll being at odds with the role I want for dice in the game. So, yay, you can consider yourself a winner, because
maybe I would see it as plausible, it's just that you've now turned the game into something where it no longer matters if it's plausible, because I've already lost all interest anyway.
I do not need to be saved. I do not need to be show how awesome fail forward is, if only I truly get it. I do not need to alter the rate at which I call for rolls. And you'll be just fine, even if I never share your views on fail forward. But, if it really matters that much to you that I agree that fail forward, when used in the way all right-thinking gamers should, is always going to be plausible, then feel free to take my comments above as proof of your rightness in this matter.
But really, this just seems like another instance of judging one approach through the lens of another. Like you’re assuming the frequency of rolls to be similar and so you’re assuming implausibility based on that assumption.
I mean, you're definitely making a lot of assumptions about which games I've run, how I've run them, and how I would feel if I ran them differently. And you're definitely making erroneous assumptions about how often I call for rolls.