Why Game Endings Suck

An oldy but a goody. The reasons game endings suck (for the most part, naturally) has been documented before, but I wanted to think about it from a specifically narrative perspective.

So. Game endings suck because they're often the last thing developed, there's no budget left, everyone's exhausted and most players never see them anyway. It strikes me that there are two main things wrong with that picture: why do most players never see the end, and why is no one talking about story?

First off, I'm a big proponent of shorter games. Reviews spend a lot of time stressing how much value you're getting, and I can understand that, but by reducing the equation to "10 hours' playtime = £30" we're in danger of promoting bad design. It's been written about before, but frankly I would take 3 hours of Portal for £50 before I'd pay a tenner for 100+ hours of RPG grind; and I'd take 20 hours of Mass Effect 2 over 40 hours of Dragon Age.

Take a look at Uncharted 2 (I know, not a PC game?!). It's got some of the highest production values I've ever seen in a game, yet it still feels the need to pad out its playtime with cardboard cutout baddies with ever increasing hit points. The result is a denouement that lacks the spectacle, script and fun of the rest of the game - for the sake of an extra hour or two of guff. It's a vicious circle: players demand more mid-game content, that content gets diluted, fewer people bother getting to the end, and so more attention gets paid to the mid-game.

But apart from cutting all the filler, how do we succeed in delivering a satisfying conclusion in a medium where most of our audience never gets to the end?

- Don't have an ending. Some of the strongest narrative experiences are generated dynamically in games like SimCity or The Sims.
- Have lots of endings. Games like Weird Worlds or Atom Zombie Smasher encourage short, one hour game sessions, yet manage to maintain an unfolding sense of drama as multiple playthroughs are completed.
- Force the ending. Short of letting go of the controller, you pretty much cannot play Heavy Rain for more than seven or eight hours with seeing the conclusion.

Ultimately, though, I think there's a key problem with the way we're looking at the ending: we're looking at it from a gameplay perspective, not a narrative one.

What meaning does an 'ending' have within the concept of gameplay? It's not an absolute thing - most games encourage further play after they've officially ended in the form of new challenges, multiplayer, or plain pissing about. Perhaps it ought to be looked at as a question of scale, of increasing spectacle; but then games always increase in scale and spectacle as a matter of course - they seem to be able to up the ante throughout the experience, but so often fall flat when that spectacle is asked not to continue rising, but to reach a peak. It's that peak that late development often struggles to surmount.

But that peak's also responsible for the embarrassing-we-still-have-to-talk-about-it end of game boss. It genuinely numbs my brain that so many intelligent and well made games are still content to reduce the conclusions of their dramatic stories to pumping a big blue guy full of bullets. But when you look at a conclusion in a gameplay context, what choice do you have? You can't just throw more baddies at the player, but you can't throw no baddies at the player, so you need to deliver something different. Like, I don't know, a big orange guy with a glowing bald spot.

If we take another look at this from a narrative perspective, though, perhaps we need to ask "Why have a gameplay-related close at all?" I hark on about tying gameplay and story together, but if doing so in this case means taking all those fantastic, polished mechanics that worked throughout the game and throwing them out for big blue balls with his 1,000HP over there, what's the point?

An ending is a story concern. I think that if we as developers learnt to trust the gameplay and the narrative to do their respective jobs then delivering more of the same in the case of the former could be rendered satisfying by applying the time and budget to ensuring the latter imbues that gameplay with conclusive meaning. Portal's final boss - from a puzzle gameplay perspective - was rubbish. But the story lifted it to a satisfying conclusion. The endings to the first two Penumbra games were the elements I was most proud of, but they used the gameplay in only cursory ways.

One of my favourite tips when I was learning to write was "Start at the start, and keep writing until you're done." It sounds bloody stupid, but what it means is that you don't need to worry about artificially constructing lead ins and outs to your story. If you just worry about writing it then those things will take care of themselves.