On the other hand, when it comes to the adventures themselves, all bets are off. A certain amount of secrecy and mystery can keep the players guessing, and keep them engaged. While the story is going, the story elements stay secret.

But I generally reveal everything later on. After all is said and done, I usually show my cards, so to speak. Which NPCs were straight-forward (even if the players thought there was more there), which NPCs were treacherous (even though the players didn’t realize it, or didn’t realize the extent), etc.

I think baring it all at the end does two things:
1. It gives the players and GM closure, explanations, and something to chew over, maybe something to laugh over. (And hopefully, more appreciation for the work I put into my adventures. )
2. It also gives me an option to receive feedback. If they can poke holes in it all later, then that helps me improve myself for next time. (“Wait a minute. If Mrs. Poppins was actually a doppelgänger, then why did she…”)

If you know that they’ll be onboard with a certain type of bait and switch, or that they’ll enjoy a ‘layers of onion’ game, then go for it. If not, then either talk it over until you’re confident that they will, or don’t run that game.

IME, players are willing to do just about anything, as long as it’s fun. The catch is that everyone has a different definition of ‘fun’.

First was a one-off so I didn’t mind pulling a fast one. I sent out some invites to a few players and told them I was running a one-off. I didn’t explain anything beyond that. When they arrived I gave them some premades for Werewolf and we started. After about an hour I revealed I was npcing a character that was a GM and we were all sitting around a gaming convention. I then handed them some blank character sheets and told them to roll up their real characters.

A second time I was running Dragonstar, wherein the characters started as typical fantasy characters. We played a few sessions that way until the space empire showed up and assimilated the planet into their empire. They were then taken off world.

One thing I always do, even if I’m not pulling a secret campaign on the players, is to mention what skills may not be the best choices for the game. In the Dragonstar campaign I mentioned, “You will be moving around alot, so you will not want skills that are tied to one location; contacts and such are not advised. Wilderness skills will be seldom used.”

There are ways to give the players information, without actually telling them anything. You do not want a player spending character generation resources on something that will have no meaning in the “real” campaign.

As for springing such things on your players, you need to know your players. Will they like the campaign despite the secrecy? I know my players and have a fairly good idea of what I can get away with and what I can not.

My blog- http://bigballofnofun.blogspot.com/

Sure, talk about campaign ideas with the players before you start, get a sense for what kind of campaign people want – city based, wilderness, mostly combat, mostly intrique, etc. And definitely talk about flavor or theme – aberrant/cthulu, orc hordes, shadow darkness, tolkein-esque, etc. It might be as simple as “Let’s play a Forgotten Realms campaign”, or it might be more complex than that.

Also, be willing to adjust and include some directional changes from the players as well. The Forgotten Realms campaign I am running didn’t start out dealing with the influence of the gods at all, but has come to include a subplot dealing with the god Kelemvor because of player influence. The main plot of the whole campaign still works, but now it also includes a player generated subplot – no worries.

It is a balancing act, the DM shouldn’t have all their prep time and work negated by contrary players, but at the same time the players additions or changes shouldn’t be negated either. Balance the two if you can, and talk with the players about this balance to make sure everyone is still happy with the campaign.

6. Guide your players during character creation
If you’re planning a major twist or setting change, guide your players during creation to make sure it doesn’t gimp some of them moreso than others.

I say “moreso than others” because I’m reminded of an old RPG (Rise of the Phoenix, I believe) in which you created astronauts and then came back to earth to find civilization had crumbled into the stone age in your absence. In that kind of scenario, everyone is equally useless, and thus you don’t get the kinds of issues with balance and spotlight time you would with imbalanced gimpage.

So in that spirit, I try to involve my players in the game and give them anything reasonable they ask for.

Otoh, exactly what you’re saying is sometimes the case. Sometimes you know you have something cool and you have to be a little devious to pull it off correctly. In those cases, a bit of ginger handling can prevent exactly what RukiTanuki describes above.

I only recently noticed that secrets sometimes work well– I’d had enough bad experiences that I’m leery of a full bait and switch. If I was going to do so, I’d probably pick a system with very broad skills (like QAGS), so that characters can remain functional. I’d play up the cultural challenges for laughs and as increased difficulties, not making the characters equally useless.

I’d answered a call to play a d20 Modern game. The six players crafted their characters, with the DM’s requirement that we have backgrounds, goals, motivations, and so on. In the first session, he roleplayed each character, one at a time, through their day. Then, each character went to the college to attend the same presentation.

…And were kidnapped by aliens, taken to a far-off, hostile planet, implanted with control collars, and told to kill the aliens’ enemies. The game effectively imploded right then and there; two players showed up for the next session, and no one bothered after that.

I took it hard; the bait-and-switch left me playing a very different game than I wanted. Worse yet, it also left me playing a character whose strengths had just been negated, by being placed in an environment where Knowledge checks were useless, no one could be communicated with (let alone Diplomacied), and every weapon available required Exotic proficiency (which we were expected to take).

I understand the intentions here. When I hosted a one-off Zombiegeddon campaign in d20 Modern, and the players knew they’d be fighting zombies by the end of the day, suddenly every character had some quasi-justifiable reason to be carrying a loaded gun into the shopping mall: Mobster wife, Gulf War vet, embassy bodyguard, etc. I didn’t insist that they make a character who wasn’t carrying a gun (which, honestly, was the easy solution), and I couldn’t hide the fact that we were playing a zombie game, as that hook was what got everyone interested.

My best plan, at this point, is this: make it clear that there will be a twist of some sort; and work with the players to ensure both that they enjoy any unexpected genre change you attempt, and that their characters retain their utility.

I’ve run the same setting for over 25 years, so one of the negatives on that is not being able to totally switch things out.

And I don’t know why this post hit me the wrong way today, maybe due to a late night dealing with a crying baby and setting up 2 gaming sessions for this week (one live, one online), but part of me just looks at the following equation, GM 100work units > players 5work units, and says ‘He who does the work determines the plot’

Now, every group dynamic is different. Some have rotating GMs, some play lots of mini campaigns, some have a dearth of players and have to be careful about not losing them. So I can understand that.

I also don’t mean to minimize the importance of all the players haveing fun, as this is a friggin game, a pasttime, even a pro-social hobby, as it were. I read about a GM losing a few friends recently over theor gaming, and that made me truly sad.

But too many articles I have been reading about have gone a little overboard about player entitlement. A game is not worth losing friends over, but a GM has to stand behind his work and integrity. I am not disagreeing with anything Matt is saying here in the OP, as everyone of these points is worthwhile. But if you know you can create something special doing it a certain way in your game, at least think hard about your integrity as a creator. If you are the one putting in the work (and this assumes you are the type of GM who puts in the design time), have a little faith in yourself.