If you want to be the World's Best GM, you will have to answer this question: should I force my players into sticking to my perfectly designed adventure, or should I allow them to go wherever they wish at any given time?
Which one is the right way to go? I'll tell you: None!
Let's see why.
Sandbox = Aimless Roaming
We are asuming here that you are running a story-driven game, which is strictly essential to cause the kind of deep emotional impression in your players that you are looking for as the next World's best GM.
A story-driven sandbox game equals improvisation. Loads of it. Improvisation is a technique we borrow from theatre, but in an RPG you must take it with a grain of salt. Theatre has rehearsals. We don't have that in RPGs. If you are a theatre director, or an actor, you know that 90% of the time, improvisation sucks. You need to spend a lot of time improvising to find an inspired idea. And that's why actors have rehearsals. Since you cannot rehearse your gaming sessions (though I've made a few interesting experiments in that direction), you need to keep improvisation to a minimum. Otherwise, you are digging into a world of trouble: mental blocks, incoherence and inverosimilitude, to name a few. Of course, practice makes perfect, but a whole adventure based on improvisation will have some clear consequences.
I know what I'm talking about because for almost 8 years my games were exclusively improvised. It was role-playing at its core. Here's a fact: I was running a Heavy Gear campaign that started with 17 year-old PCs being recruited by the army. When I asked for the first dice roll, we suddenly realized we had been playing four months (3-4 nights a week) without rolling a single die. That campaign was one of my best experiences as a GM, but it also taught me a lot of things about story-driven sandbox gaming:
Lack of Direction
You are improvising all the time, adapting to your players by the minute. Your players are under the spotlight the whole time. Everyone feels great, at least for the first two or three months. But then you start to realize that the story is not going anywhere. Since your players have gotten used to endless role-playing sessions, they stop taking meaningful decissions. One day you realize you just spent a 8-hour session of role-playing where nothing has actually happened at all.
Lack of Purpose
Your PCs start to disolve when you don't push them to move forward. Eventually, players in a story-driven sandbox game turn into passive actors that prefer to discuss the world around them instead of trying to shape it. The frontier between PCs and NPCs start to disappear.
Lack of Action
Improvisation eventually leads to endless repetition. The classic three act structure is shattered, and PCs get stuck in a limbo.
Temporary Conclusion
After those four months without rolling a die, I came to a conclusion. I needed to take the initiative again. Players enjoyed my games, but it was hard to find a difference between one session and the next. I prepared my next adventure with a set of targets in mind:
* Structure in three acts.
* A clear goal, a clear motive and a clear enemy.
* A set time frame to start and end the adventure: 3 days.
I hadn't forgotten how to improvise or create lively NPCs, of course, and the adventure was a tremendous success. That was the first time my players clapped when the adventure was over.
I have been following those principles ever since.
Railroading For the Win
So the quick conclusion is that you need a structure, a good and strong one. Something that in cinema they call iron script. A clear plot with a short and shocking setup, a conflict full of turning points and a short and mind-boggling resolution. And you want your players to stick to it. Sure you do. The interesting part is that they also want to stick to it, even if they will never admit it. They want you to provide a solid ground to base their decissions on. Good players want to unveil the plot of the adventure, because they realize that will be more fun than wandering around without purpose.
But this is an RPG, right? So your players must believe, at every moment, they have free will and live in an open-ended story.
To make ends meet, I use a system of flowcharts. The fact you have devised a given amount of scenes doesn't mean they all need to follow a specific sequence. In order to guarantee a real sense of player freedom, I organize my adventures in pairs (or trios) in parallel. Which means that they can be resolved in any order.
For example, I'm currently running a saga of three sequential campaigns called The Unchained God. As you can imagine, it involves Tharizdun, but it is set in the Forgotten Realms. My players assign quests to themselves, and right now they have three in mind: one involves going to Westgate, another to Myth Drannor and the last to the Thunderpeaks, and they are currently discussing what to do next.
But of course, these three quests were planned by me ahead of time. They are in my flowchart as parallel adventures, that can be resolved in any order. And they can have any outcome, either positive or negative to the PCs. The combination of possible consequences will have a clear impact in the saga's future events. The end of the first campaign will determine the beginning of the second, and so forth. Eventually, the decissions of the respective PC parties of the three campaigns will determine if Tharizdun escapes and destroys Creation or not.
Real Conclusion
Here's the rule of thumb: Railroad the Present, Sandbox the Future. Create an iron script for your adventures, but a loose guideline for your campaigns. That way your game will be supported by a strong structure, but the campaign will remain flexible enough to adapt to every adventure's outcome.
