Like many since lockdown I have been caught up in the rise of tabletop role-playing games. Also like many, I have not always had many opportunities to get together with people to make a campaign a reality. What I have had is plenty of time to work on campaign ideas on my own, with occasional opportunities to try out my ideas with my friends. I have a number of take-aways that I feel are relevant to teaching:

1. Always give players something to react to, but never crystalize an idea on how they will react.
It is easy to get excited about a narrative idea and to imagine the possibilities. This isn’t a bad thing, but it’s important to not go too far down any particular rabbit hole. Your players will be improvising, and just because they pick up on your hooks doesn’t mean they will do so in a way you expect. I firmly believe this is where accusations of railroading emerge: the players choose a direction, the dungeon master realizes that it won’t work with their vision, and so rather than let go they say that it won’t work. (I do also think there are players who refuse to go along with the story, which is unreasonable. My favorite response is from The Dungeon Dudes of YouTube: “OK, here’s a fresh character sheet, roll me a new character who WILL go on the adventure.” Dungeon Dudes – YouTube) When we get lost in our own imagination we over commit ourselves to one possible reality that is not ours to make alone. Now, I think it is good to have something to aspire to, and you do need hooks and plotlines and characters that your players will have a strong reaction to. You also need to have some idea of what is happening: you can’t just improvise every detail of the world on the spot, it would take forever and kill any sense of momentum. Hence, my saying. Teaching is a little different. Our lessons are more structured, the learning outcomes set ahead of time. I would apply this rule in the classroom towards assessment and project based learning. If you are doing a project then know what you want students to learn and accomplish in a general sense, and do have clear entry points. If you just tell them to “do a thing” then they have nothing to react to. If they have their own ideas, however, or if they run into problems, do not hold rigidly by the specifics that you envisioned. The students can still accomplish your learning objectives.

2. DO over-prepare, but keep it to your speeches, descriptions of characters and environments, anything that is guaranteed to come up.
The worst DM advice that everyone gives (youtube.com)
Ginny Di, a YouTubers and DM I like, is one person to spell this out quite articulately. Put simply, the advice “don’t overprepare” is bad advice. You can’t just wing it on everything. You need to have work in place to describe the environments, the buildings, the people and the lore in your world. If a character is giving a speech then you should know what it will be. Would you just make up an important presentation on the spot? Probably not. So do over prepare, just keep it to the things that are guaranteed to come up, then improv the minutiae during the game.
This applies heavily to teaching. You really need preparation to be successful. Sure, students may have questions you didn’t think of or requests you didn’t foresee, but that does not mean that you can just go into class and make up everything on the spot.

3. Don’t treat people like a problem / No AITA’s or “that guy” stories!
Is there any question I hate more than “am I the asshole?” I don’t think there is, it is so problematic. For starters, we’ve already concluded that someone is a problem. Secondly, this has to be fought over, so we can determine who is the problem. Finally, and critically, what is the way forward here? How do you go from labelling someone an asshole to making things work constructively with them? And if there’s no way forward, why bother asking the question? Catharsis? And why, oh why, does it always become very clear that the person asking AITA has made no effort to communicate their displeasure to the person in question? “I tried being super passive aggressive but they just didn’t get it! Can you believe that?” The afore mentioned YouTuber Ginny Di has given good advice: “have you tried talking to each other? Like adults?” No, usually not.
This is pretty common in teaching. When we label a student as a problem, or a parent or admin or other teacher as a problem, we preclude any real solution. Why would we talk to them if they are the problem? How would there be a solution? We hamstring ourselves into bad relationships and classroom environments.

4. Players are extremely loss averse
A player’s character is all they have, and they won’t make reckless choices that could lead to a loss of that character. I played in a game with a DM who felt that no player deaths meant the game was too easy and lacked drama, so he went out of his way to get a player character killed. She revived later, but it was extremely stressful for her. I have similarly heard DMs online say things to the effect of “of course I’m going to try to kill your characters, I’d be stupid not to, that’s my goal.” This really ignores the imbalance at hand. A DM losing one of their villains or monsters isn’t the end of their game and isn’t the end of their involvement in the story. A more accurate parallel might be the DMs main villain being killed as soon as the players encountered them, leaving the DM to come up with a new plot.
Similarly, people had a tendency to take too long on their turns in combat, which he said and I agree. That said, when you only have one shot at your turn you don’t want to get it wrong. Pretty reasonable. Putting a timer on peoples’ turns is just kind of a stressor, and punishment is the wrong way to go I feel (not that we ever implemented a timer.) When you put people under stress of failure they will rarely get more engaged, they will either quit or be miserable. If you find a way to reward the players who are faster, however, then you encourage fast play.
Traps are another area where player actions can be slowed way down. Traps were originally added to introduce an element of danger and tension to the process of exploring a dungeon. In practice, it means the players will meticulously check every brick, wall, ceiling, chest and rock for trip wires, hidden monsters, pressure plates and anything else that might cost them. Jumping past spikes? Forget it! If it goes wrong they could lose their character, and that’s not happening. Maybe you as the DM know that the threat is not lethal but your players don’t, or they don’t know it for sure, so they won’t take the chance.
In teaching, students will not risk looking stupid. If you force them to try anyways then they will either quit or be miserable. You have to establish that the students are standing to gain something by engaging, and that they do not stand to suffer serious loss. Do not just say “oh you’re over reacting, what’s the worst that can happen?” Students have serious loss aversion, you have to allow for it.