Going rogue – Reimagining the value of procedural games

Despite many years playing games with procedurally generated levels I only relatively recently started associating the term rogue-like with those games. Of late this genre of game seems to have become an indie darling and there are some fantastic examples including of course Spelunky, Don’t Starve, FTL and another couple I first discovered on the Indie Megabooth at Gamescom this year, Below and Desktop Dungeons.

Perhaps what separates these games from other procedurally generated experiences is their deadly and uncompromising attitude to mistakes. There is no going back and they all seem to exhibit some degree of permadeath.

For me Don’t Starve is the one which seems to endlessly endure. As the mad scientist (or one of the other unlockable characters) it’s my task to gather the materials necessary for survival and then from that point start to explore and build more and more sophisticated technology which is intended to make your survival easier, but in practice always seems to put me into deeper risks.

There is apparently more to it and an adventure mode which allows you to use uniquely discovered items to explore new worlds – but I’ve never gotten that far.

The constant lure of more surprises, of new set-piece elements in the game which I continue to discover after months of playing and the simple joy of gathering resources and seeming to gradually improve each time is brilliant. But the biggest illusion is that when I fail it’s my fault; not the game’s.

However, despite their apparent popularity, it still seems that the majority of these games remain paid up-front. They can sometimes even command a slight premium, for example FTL was launched at £6.99 as I recall.

This makes me wonder if there is something that might be intrinsically damaged if we made a game like this using in-app purchases (I don’t know if we are allowed to call this free-to-play anymore).

Let’s look first at the idea of permadeath. There is a strong risk that more ‘casual’ players will turn away if they lose all their progress. Especially in such games where there is an almost sadistic joy expressed by the way the game operates to exploit our human frailties. If we don’t pay enough attention to our health. If we forget to have a torch on us when it gets dark. If we picked all the flowers too early in the game and… “Hey Pal, you don’t look so well”.

However there is something, almost counterintuitive in the way that the more punishing the game, the more ‘hardcore’ players seem to keep coming back. That feels wrong for many freemium designers who used to believe that we shouldn’t let players fail and even those designers (including myself) who just think it’s generally wrong to punish players for not being ‘good enough’. Assuming this logic applies to the rogue-like mechanics is mistaken.

Failure is an essential aspect of our motivation to play; for without it we can have no success. Beyond that cliché however these games hide a secret truth of design. Failure is essential for the game itself to become compelling; but only as long as it reveals a promise of future success.

It’s not that we like to fail, but most of these rogue-like games allow us the opportunity to come back and to succeed in contrast to our failure. Each new session allows us to improve and that delivers a personal sense of competence as well as a sense of autonomy. We get to make our own mistakes and try again. This isn’t unique to the rogue-like games. Threes, 2048 and Flappy Birds all have this particular quality. However, it’s worth contrasting this with the FarmVille days where there were questions about whether it was a good idea to even have a failure condition at all.

Were the Zynga guys completely wrong? Is their subsequent decline a sign that they have nothing to teach us? Of course not. But we should re-examine what their data showed with what we see now with fresh eyes. Failure is important, but so is creating a positive motivation to replay.

If we leave players feeling that the game is too hard for them, then of course they are going to walk away. If we make them feel that the failure was a simple, avoidable misstep and get them back into the game fast, then they are likely to be more motivated to keep playing.

If we want to avoid a sense of punishment we have to look beyond the mechanics and instead look for continuity and progression in the Context Loop of the game. In other words how do these game make us feel between sessions.

It seems to me that the trick is to make the end of your characters ‘life’ the start of your next playing session; something few of these games try to do. However, the best of them give you new characters to play or unlock new equipment often not to make the game easier; but instead to add to the depth of play. For example each new character in Don’t Starve seems to mix up the dynamics rather than to make things easier: the same thing applies for the new spaceship variations you earn in FTL.

What I’ve learnt from thinking about games as a service is that it allows us to change our perspectives. We consider the players experience as a journey, and not just the core mechanic but the contesting which this game is played. Applying this to rogue-like games leads us to spend more time thinking about what happened to the player between sessions and how we can encourage them to repeat their play.

More than that there becomes an opportunity to build in-app purchases alongside the normal playing rewards which don’t just make the game easier, but add richer experiences too. These don’t just have to be shortcuts or resources; there is an amazing opportunity to personalise the story between each playing round to build deeper engagement still. Then if you have an ongoing release plan you can introduce new concepts over time that players would want to invest in.

We can’t be complacent however, we have always to be sure that expected reward from playing always exceeds the perceived effort, otherwise players will simply walk away. I suspect that long-term retention in games is related to a tension between the reward we expect to earn and a sense of unfinished business in the game. It’s a kind of itch we have to scratch, but that definitely needs more research.

In summary, far from being a genre that is ‘naturally’ paid up front, I think the rogue-like design model has huge potential to disrupt the current market of F2P copycats. But to be successful we have to deliver an intrinsically repeatable, delightful experience which reciprocates our players’ investment of time and money; giving them something to anticipate if they continue to play our game.