Someone who’s seen a film treatment I’ve co-developed said that they very much like it, but were concerned about some predictable elements. The tone of their communication suggested that they felt this was a problem, and I can see where they’re coming from. After all, you don’t want to second guess a story…do you?

I saw How To Train Your Dragon earlier, and predicted much of what was to come. And you know what? It didn’t spoil my enjoyment of the film one iota. And it wasn’t predictable because of being a film primarily for children. Not at all. It’s just that, once you set up the characters and plot and world they inhabit, there are only so many ways it can plausibly go and hold your interest at the same time.

Sure, the dragons in the film could have torched the village and dropped the Vikings into a volcano. But that would have lent the story an unnecessarily apocalyptic tone for an audience of a tender age. And yeah, the hero — Hiccup — could have used his dragon-friendly ways to rule the Vikings by fear as their despotic leader. You know what though? Those would have been pretty shitty stories for a bunch of highly talented writers and animators to turn their attentions to.

So, How To Train Your Dragon is kind of predictable. Guess what though? The story is delivered with such relish, such panache, such joie de vivre, that it matters not a jot. What counts is how enjoyable the experience was, and let me tell you it was a hootenanny compared to the smartarse games played in Shutter Island. If Scorsese’s looking for a real challenge, I’d recommend he makes a film for a young audience, and find out whether his Hitchcock riffs and colour schemes pull the wool over young eyes as well as those routines lull his adult viewers.

Imagine a friend is telling you a story. Maybe you were at the event that the tale relates. Do you interrupt her as, laughing aloud, she sandpapers actuality and introduces other elements to tell a yarn that gets you feeling as good as she clearly does? Or do you pull her monologue apart for inconsistencies and contrivances that push it closer to a three act structure than what ‘really’ happened?

Being unpredictable for its own sake only goes so far if you’re interested in engaging an audience. It might fascinate David Lynch, but you’ll note his work rarely troubles the box office and that he helps fund himself through his online presence. That’s far from a criticism: it’s a good business model…at least for someone who has dabbled with something like the mainstream from time to time and profited from it.

Brains like patterns. Can’t get enough of them. So if anything, they’ll tend to find evidence of coherence even when there isn’t any — which explains the attraction of conspiracy theory. Work with that tendency — there’s a whole industry devoted to helping you to do so, with the likes of Syd Field and Robert McKee offering their versions of how film structure really works.

Thing being, it’s not the structure that people go and see films for. They go to be moved, to laugh and cry and empathise with people going through journeys analogous to their own, even if those journeys involve spaceships and spies and Eddie Murphy in a fatsuit.

If a friend asked for a fiver and then cleared off, never to be seen again, you’d be understandably annoyed. And confused. As with friends, so with films. A trailer helps create a contract in the viewer’s mind. One that tells them what sort of film they can reasonably expect for their money, and what kind of emotions will be involved in its realisation. You can honour that agreement, and ideally do so with some wit and style, throw some curveballs in to keep them on their toes. Or you can run off with their money. Only, do that, and they won’t be coming back to see anything of yours again — and they’ll make sure their friends know you ripped them off.