Events that potential or certain negative consequences for the character that they zero control over. In my experience it doesn't end will over the long haul if that handled through fiat. Players are far more accepting of the results if it occurred because of random generation.
<snip>
Players are more aware than one would think that the referee just happened to create a forest in front of them to adventure in. It can be gotten away with is done sparely but done over and over it become a noticeable pattern.
<snip
it takes away from the challenge knowing the referee is creating something out of whole cloth right then and there.
<snip>
Keep in mind player can and do make a bad plans.
<snip>
Because the bias is minimized as a result. So the result is perceived as more fair. Provided of course the random table itself is perceived as fair.
I'm not sure what you mean when you say "players are more aware than one would think that the referee just happened to create a forest in front of them". What do you think I think? Do you think I'm lying to my players?
My players knew that I narrated a storm and the landing of their ships on the Dalmation coast. They knew that we collectively agreed they were travelling overland to Constantinople. They knew that I narrated that they came to a forest. They knew that we were working from a generic map of Europe that gives one a general sense of what is between the Dalmation coast and Istanbul/Constantinople.
I don't think the concepts of
fairness or
challenge have much applicability in my RPGing, at least as you use them. I am not challenging the players in their ability to plan an overland trek. Or to avoid meeting the Bone Laird. The "challenge", such as it is, is to decide how to respond to an ancient ghost who lingers on for some reason.
If "sandboxing" means
play that prioritises geographical matters, then I am not running sandbox games.
Though that would still leave it a bit unclear to me why it is OK for the GM to establish, say, a challenging bit of topography (without random rolling) but not a challenging bit of weather (without random rolling).
So I think the timing of scenario design is not all that important. If the GM is coming up with the forest now or 10 months ago does not matter to me. What matters to me is the thought process behind it. What are they prioritizing? Are they trying to create a challenge? Are they framing something that provokes action? Are they trying to create something that should be interesting to explore? Are they guided by what they think will make the best story? For some GMs timing can matter because they feel more temptation to skew things away from what they really want to prioritize if they make that decision in play, but that experience is not universal.
In my case, I narrated a forest because that was what I had in front of me in the scenario I wanted to use. It was colour, easily incorporated - surely there were forests in Dacia/Romania in the 8th century CE? - but not the principle focus of play. The focus of play was the NPCs and their ongoing status as ghosts.
Celtic ghosts, as it turned out.
I would think that sometimes, even in a geography/architecture-focused sandbox, the GM must have to come up with details on the fly: colours of drapes, shapes of columns, manufacture of roofing materials, etc. Presumably most of the times these aren't very significant to what is at stake.