Robert McKee's Story seminar is considered the premiere writing guide, particularly when it comes to film and screenwriting. I happened to get the book from a friend. It's an excellent read whether you're writing for novels, plays, films, or just about any other medium because of its explanations of universal storytelling structures.
But applying Story to video games is tough. Not only is storytelling in video games a sensitive issue at the moment, video games have one element that completely separates them from every other medium: the player. Accounting for player involvement and interaction makes applying tried-and-true principles for other mediums (like Story) almost impossible.
Fortunately, I've already discussed the player and its archnemesis, the game. By taking that step and treating the game as an entity on par with the player, the balance of postive and negative forces essential to storytelling is restored. Which means we can start applying traditional storytelling principles to video games under this new context.
The strength of story structures lies in creating patterns that crescendo to an ultimate climax. Two of these patterns are: 1) the swinging pendulum of positive and negative outcomes, each pushing stronger and stronger until one ends up victorious, and 2) the protagonist's effort to achieve his or her conscious desire, and the gap that appears between expectations and reality that forces the protagonist to try a different, more risky action.
When these patterns are applied to action sequences in film, the result starts to look very much like a video game. I'll give you two examples.
While babysitting my little brother, I found myself watching Toy Story 2. You know that sequence with Buzz Lightyear at the beginning of the movie... that turns out to be a video game? You should probably watch it.
The action starts with Buzz Lightyear flying onto a rocky planet. It starts out simply enough, with Buzz diving into the atmosphere and weaving through a canyon. He then stops in a clearing, looks around, and finds that there's no intelligent life nearby. He looks down to make an entry in his mission log... and soon finds a thousand red laser sights trained on him from killer robots nearby. They prepare to fire and obliterate him, but he dives and fires a laser at a crystal nearby, which reflects and multiplies the laser to fire at every single point in the surrounding crowd. They are all hit, but they all explode in green fire and robot parts, sending Buzz flying.
He hits the ground running to avoid getting squashed by falling debris and flattens himself against a rock. But a mounted camera pops out of the rock and spots him. Buzz destroys the camera, but a hidden Z-shaped door opens up in the ground. He falls, the door closes behind him and leaves him in darkness, but his suit has a glow-in-the-dark component that lets him continue on. He enters a hallway, and the lights suddenly turn on. A spike wall appears behind him, and he starts running. A door threatens to close in front of him, but he jumps through.
Now he finds himself in front of a bridge spanning a bottomless gap. Ahead is the power source for the base. But the bridge is made up of small platforms suspended in midair. So he jumps forward, one platform at a time... and then the platforms begin to plummet into the pit below. Buzz manages to activate a button on his suit that, through some kind of hokey magnetic physics, allows him to fly back up to the platform. He approaches the power source and makes a grab for it... but it turns out to be a hologram. Zerg, his archnemesis, rises up behind him and attacks.
And then it turns out Rex is horribly proportioned for gaming, but that's beside the point.
The point is that there's a pattern in how the action progresses. It just keeps going back and forth: he's safe, he's in danger, he finds a way out, there's another trap, he steps on a platform, and the entire thing falls from beneath his feet. The situation swings between positive and negative, constantly.
The other example is an episode of Human Target - in fact, the only episode of Human Target I've bothered to watch. In episode 6, Lockdown, the main character Chance is required to rescue an unwilling engineer from the high-security headquarters of a corrupt weapons manufacturer. There's a Splinter Cell feel to it already...
The action opens with Chance skydiving onto the building's roof. All's well and good until they realize halfway that there's a camera on the roof with a two-foot wide blind spot. Whoops. Chance barely makes it, though his parachute nearly drags him off the side of the building due to the wind. He takes out the camera, but that alerts the security office inside. A guard notices him, but he takes the guard out and makes his way into the building. He disguises himself as a janitor and makes his way to the nerdy engineer's office. Undercover, he blows open a window, but that sets off a security breach alarm. Guards come in and shoot up his parachute, removing their initial escape plan. Now they have to escape a high rise that's been locked down.
After dealing with the guards, they escape to a floor still under construction. They figure out that they can use one of the experimental weapons to shut down the security system, but it's many floors below them. And with the security office monitoring the temperature of every room, they're going to get found. So they enter the vents. They avoid detection for a while, but then someone in the security room gets a bright idea and opens up the secondary shafts in the ventilation system. As Chance is crawling through the vents, a whole section opens up from under him, leading to an insanely long and narrow drop. He barely hangs on, but now he has to get his shy engineer buddy to make the jump. After some convincing, he tries, nearly falls, and Chance has to pull him back up.
They make their way into the "black room," barely hiding in the blind spot of a camera. They use a digital camera to fool it. They make their way across tables to avoid detectors on the floor. They reach the weapon, but it's not finished. And their heat signature has been picked up, so guards are on the way into the room while they are trying to fix it. At the last second, the engineer says, "Well, it could work or blow up in our fac-" and Chance flicks the switch. It works. With all power and security down, they make their way into an elevator shaft. But the head of security himself picks up an automatic weapon and finds them. An intense gunfight plays out in the elevator shaft, ultimately ending in the antagonist's demise.
They reach the entrance, and the FBI has shown up to take control of the situation. Chance gets thrown into a car... but one of his friends shows up in a suit and covertly steals him away.
Barring several of the obvious issues, this whole episode would actually be very engaging to play as a video game. It's full of dramatic twists and turns that directly affect the protagonist on a physical level. And when another obstacle appeared, Chance was forced to pick a riskier action: escaping to a different floor, moving into the vents, turning on a potentially unstable experimental weapon... There is a progression there that is deeply linked to storytelling principles.
These are examples of creating dramatic shifts in the physical plane - creating dramatic action. If that action were to be put into the hands of a player, it would become dramatic gameplay. With this principle of progression, you can now weave storytelling into the gameplay, strengthening the overall storytelling of the entire game.
In your single-player game, create dramatic shifts that directly affect the player. Arrange them so that each one is stronger than the last. And then give the player avenues to fight back with gradually increasing strength, creating the positive back-and-forth that can result in an ultimately satisfying climax. All other great stories - movies, plays, books - do the exact same thing, but only games make people feel like the ultimate climactic action is their own. That's the kind of power we want to wield.