The GMTK episode I threw away (Insider, May 2018) (Patreon)

So this month, I spent a fair amount of time on an episode that I just… threw away. Basically. I did the research, wrote the script, and was about to record it when I decided that it wasn’t right for the channel and so I tossed it.

The episode was about this concept of a “failure spectrum” (a term dreamed up by Gunpoint designer Tom Francis), which describes all the states between perfect success and complete failure. Some games (Celeste) have a teeny tiny spectrum whereas others (MGS V) have a huge one.

The idea was to explain the many virtues of having a wider, more generous spectrum. I would talk about how interesting stories crop up in the middle ground. How it supports and acknowledges failure, but doesn’t take you out of the experience. And how it can promote improvisational tactics.

I binned it for a few reasons.

• One: it just seemed a bit too short and obvious.
• Two: I was talking about the same games and genres that I always do (Far Cry 2 and MGS V).
• Three: I was quoting the same people I always do (Tom Francis and Clint Hocking).
• Four: It just seemed like it should be a small part of a wider discussion about failure. 

So you’ll probably see some of this stuff crop up in a future episode. But instead, I decided to make a video about a topic I’m less familiar with and a genre I don’t talk about as much.

It was way more fun to research and make the episode and I’m very glad I bit the bullet, threw away some hard work, and started again.

If you’re interested, here’s the script I tossed away:

In some games, the line between life and death is razor thin. 

But in many others, there’s a bit of leeway to make mistakes. Think about a shooter, where your health bar lets you take a few shots before death - while medkits can bring you back to full health. 

Or a stealth game, where getting spotted doesn’t instantly end the game, but instead transitions things into a fight. Unless you successfully lose your attackers and go into hiding, in which case you can go back to sneaking around. 

Tom Francis - creator of Gunpoint and Heat Signature - is the first person, I think, to give this idea a name. He calls the range of states between perfect success and total failure a failure spectrum.

“There’s a spectrum of possible outcomes,” he says, “and screw-ups can move you towards the failure end and recoveries can (sometimes) move you back up towards success”.

And so pretty much every game has some sort of failure spectrum, but some games - especially more modern and systemic titles - have a much larger and more generous failure spectrum, than others. Tom points to stealthy open worlder Metal Gear Solid V, which has a huge range of states between being a sneaky snake and a, uh, dead snake.

So, if a guard sees you, he won’t immediately start firing. He’ll just come in to investigate more closely. If you do get spotted, you enter into this slow-mo reflex mode, to give you a chance to headshot the guard in question. Screw that up, and the guard will need to manually call up his buddies for support, giving you a chance to stop him.

And even after all that, Snake can still escape. Or enter combat. Or even call in an helicopter and just take the whole “espionage” bit of the MGS slogan and dropkick it into the ocean. You’ll only die if you manage to screw all of that up.

So this is a huge failure spectrum, with all sorts of states between completing the mission without ever being seen, and bleeding out on the battlefield. And while mistakes will push you closer to the failure end, there’s still opportunities to turn things around and scramble back towards the other end of the spectrum.

So what are the benefits of a larger failure spectrum?

Well, one is that riding back and forth across the spectrum can provide some of the most emotional, interesting, and memorable stories. You know, like coming “this close” to dying, but turning it around and winning anyway. Or killing a panicked guard, seconds before they can raise the alarm. Maybe a switch to frantic action when the stealthy approach falls flat, or an exciting extraction when your plan starts to go hideously wrong. 

These stories get lost when the failure spectrum is so narrow, that a couple screw ups can take you to a game over screen, and straight back to the start of the encounter. When the spectrum is wider, there are far more opportunities for turnabouts and reversals of fortune. 

The other major benefit is that moving across the spectrum causes you to change your goals in an exciting, dynamic way.

In an action game, when you’re on the successful end of the spectrum you can focus on being aggressive and confrontational. But if you shift down towards the failure end you’ll need to change your focus towards getting into cover, finding health packs, or maybe even crafting a medkit.

Changing your priorities keeps you more engaged, gives you more to think about, and stops the game from becoming repetitive or single-minded.

Designer Clint Hocking made this sort of goal shifting a big part of his game, Far Cry 2. In a 2009 GDC talk, he explained that, originally, he wanted the game to be all about intentionality, which is achieved by having the game split between two phases.

There’s a planning stage, where you survey the scene, look for items of interest, watch guard patrol patterns, plan your escape route, choose your load-out, and even pick the time of day. And then an execution stage, where you actually carry out your plan.

But the creators decided that they didn’t want you to either have your plan totally work, or totally fail. Instead, they wanted you to suffer small setbacks - using tricky, messy, volatile systems like malaria, fire propagation, jammed guns, and car breakdowns - that would cause you to bounce out of the execution stage and back to the planning phase. 

Clint describes this type of gameplay as improvisational. It’s this idea of constantly moving between planning and execution - within a single, continuous playthrough. And this is only really possible when there’s a generous failure spectrum. There needs to be ample breathing room between life and death for the player to bounce around. To get knocked down, but come back swinging. 

And Far Cry 2 achieves this with a big health bar, loads of healing syringes, and the buddy system, where you can get one extra chance to keep playing after death.

Clint says “kicking the player out of the execution phase is easy, making sure they fall back into planning, not a loading screen, is something else”. Quite right. 

Because seeing a game over screen can take you straight out of the experience. Back to Tom, he says total failure “interrupts your immersion and ends your investment in this current run. It pulls you out of the game, and you find yourself in a menu, then at a checkpoint or a save game. Mentally acclimatising to how much of your story has been lost forces you to disengage from it, and you have to build up all that immersion again from scratch”.

“If failure isn’t game over, it’s still nail-biting when to come close to it. And when you do slip over the threshold, it’s just another development in the story you’re creating and living through.”

The worst type of game over screen, though, has to be when you fail for pretty much anything other than your character dying. This is often when you fail to follow the exact steps that the designer has laid out for you - and the game’s only response is to throw up a game over screen and ask you to do it again.

If you ask me, that’s not really a failure of the player. That’s a failure of the game designer. 

So it’s also worth thinking about a failure spectrum in this context as well. Look at Resident Evil 7. During the first boss fight against Jack in the garage, the designers want you to grab the car keys and get in the car - but the game doesn’t just end if you screw that up. Instead, it transitions into a different scenario, where Jack himself gets into the car and starts to drive towards you.

You’ll see it happen in LA Noire, also, where messing up a certain aspect of the investigation rarely leaves the case unsolvable. Cole just has to approach it from a different angle.

This does surface one of the downsides of a wide failure spectrum, though. If mistakes aren’t met with intense punishment, and if there are always alternative routes when you screw up, it almost feels like your errors are without meaningful consequence. Games become too easy and forgiving.

And that’s certainly true. But I think you could also argue that games with a wide spectrum simple redefine the concept of failure, from a binary win / lose state to something more analogue and nuanced. Look at MGS V, which tracks how often you get seen, or use reflex mode, or how you deal with enemies - and gives you an appropriate grade to match how well you did.

And in a game like Invisible Inc, there are loads of possible outcomes to a mission, from complete success with every objective and 100% stealth, to a botched extraction that just about keeps one team member alive. All are supported, but will have different repercussions down the line.

It’s just important to avoid spiralling feedback loops that will compound the impact of those poor runs, and potentially make the rest of the game impossible. I’ve got a whole episode on this topic, so check that out. 

Another downside to having a wide failure spectrum is that some players will simply decide that they will only accept perfect play and manually reset the game state the second they make a mistake. And, sure, that’s a valid way to play I guess - but it squashes the failure spectrum down to a razor thin line and removes all the juicy middle ground, where the emotional stories and shifting goals are found. 

For Clint Hocking, he reckons the answer is to have setbacks that are unpredictable and fixable. 

Those systems in Far Cry 2, like malaria and weapons jamming, occur randomly - so it’s pointless to quick load to an earlier save when they occur. And because they are easily survivable - just pop some pills or get into cover while your weapon unjams - it’s only a minor setback.

Unpredictable doesn’t really work for a stealth game, but letting the player quickly get back to hiding certainly does. Making running away and hiding both viable and speedy could help avoid that nasty “got spotted, quick loaded” type of play. The gold standard, of course, would be to have a game where the mechanics for running away are just as much fun as those for sneaking around. 

So ultimately, some games will always have a thin line between life and death. The player just has to do it right, or do it again. And they’ll likely enjoy dashing their face against a brick wall until they master it and break on through to the other side.

But in other types of games, it’s worth thinking about widening the failure spectrum, and being more leniant with the player before throwing up that Game Over screen. This provides space for emotional stories. It forces the player to change their goals and improvise new strategies. It gets rid of annoying game over screens for not following the script. And it redefines failure to something more nuanced and personal. I think we’ll be seeing a lot more of this in future games.