Everything Justin Alexander produces is good.
RPG
Discussion of table top roleplaying games.
Personally, I find even the three clue rule a little bit too restricting in practice. Like, what if the players just never investigate the locations where those clues were placed? I find that if you prep options A, B, C, and mayyybe D, players will inevitably take a hard right over to Q like it was the only logical option. (Plus, even if everything works out, it's triple the prep vs whatever actually becomes relevant in gameplay.)
The approach I've found success with is to prep secrets, not plots or clues. Write out a list of a dozen or so things that it would be helpful for the PCs to find out, and then actively look to shoehorn in clues about at least one of those secrets no matter what the players decide to end up doing.
Like, let's say one of the important secrets is that the captain of the guard is a drunk. If your players go to the tavern? They see him passed out at a table in the corner, and the tavernkeep remarks on how sad it is that he's been like this so often lately. If they go talk to other random guardsmen? One of them makes an off-colour joke alluding to the captain being a drunkard, then they all get really uncomfortable and tight-lipped if pressed in seriousness. If they break into the guy's office? They find a bunch of empty bottles and a letter of condemnation from his superiors. Etc etc, you get the idea. The big mental flip is, instead of going "okay here's the scene, what does it make sense to do here (and would any secrets come up as part of that)" --> to instead go "okay here's the secret I want to share, how does it make sense to share here"... there's still no suspension of disbelief needed, it's just a somewhat more narrative attitude to developing things without breaking an overall more simulationist framework.
This method does admittedly require a bit of thinking on your feet, but I find that having the secrets prepped and listed out in advance gives enough of a framework to think within that it's not actually as much mental load as you might think. A++ would recommend all round. I've shaved off tons of prep time since I started going with this approach, and have actually made my sessions more dynamic as well (since I'm more actively looking for chances to feed them info, instead of more passively guarding my hoard of secrets).
Personally, I find even the three clue rule a little bit too restricting in practice. Like, what if the players just never investigate the locations where those clues were placed?
The nodes thingy work by being basically a bunch of arrows. The redundancy of "three" makes it very hard for them to miss everything. If they did, the clues they had to begin with were lacking. Alex said in some blog post that he adds new clues on the fly when the players are lost and there is only one remaining for them to discover, but even this should be a rare situation.
(Plus, even if everything works out, it’s triple the prep vs whatever actually becomes relevant in gameplay.)
There's an article of Alex where he said it's not true, because the new clues could either give the players more assets (like new allies) or information about the issue at hand. Since there isn't supposed to be an order of events, this is good. There was this game I planned with three angles to approach the "High Mage": One by her discredited rival trying to desperately accuse her of crimes, other by listening to gossip about her as a great ally for the trans community and the third as the heretic who stole and raised a dragon as a human child. Each combination of those "clues" is a different way they'll think of her entirely.
The approach I’ve found success with is to prep secrets, not plots or clues. Write out a list of a dozen or so things that it would be helpful for the PCs to find out, and then actively look to shoehorn in clues about at least one of those secrets no matter what the players decide to end up doing.
The issue with this approach is that you seemingly shoehorn the clues at the start until the players finally bite, then move to the next one. This goes counter to the spirit of plotless prep. You aren't dictating how they will find out something, but you're still dictating the order in which they will. This is a structure for linear campaigns, not for dynamic ones.
I did the same thing some years back. It ran with success until one of my players complained about being railroaded (wherever I go in town, everything points out to the goblin cave!). I had an ugly discussion with him at the time, but now I realize he was right.
I’ve shaved off tons of prep time since I started going with this approach, and have actually made my sessions more dynamic as well (since I’m more actively looking for chances to feed them info, instead of more passively guarding my hoard of secrets).
What attracted me to plotless prep is that I was sick of trying to "hide my agenda" and I would rather have no agenda at all. I already was always actively looking how to keep them in the rail roads without looking like there are railroads, and this is just a way to do it with more leeway for error.
All being said, I also did a "secret list" structure that was just a list of topics with a rough order. But it wasn't for the plot, it was for the information that the Summoner's Angel Eidolon would gradually share by accident and I wanted "Ah hey I was a mortal once :) I was a mathematician called Jeff who lived over that hill!" (fluff) to come before "your god doesn't exist, although we celestials do, and we only pretend your faith is real because it makes it easier for us to do our bidding" (major world secret).