Evolution and Kickstarter

This article is about Kickstarter, but I’d like to open by talking about the role of luck in games – trust me, it’ll all make sense in the end. Suppose you’re in a game with two strategies, one will give 2 points, the other gives a 50/50 chance of getting 3 points or nothing. Which is the best option? Clearly the reliable strategy averages two points and the unreliable strategy averages one and a half, so the reliable strategy is the one to play. Nobody likes random numbers.



Now imagine that you’re playing several rounds of the game with a large set of other players. If half of the group takes risks and the other half does not then under some conditions you’d expect someone from the risky group to win. Not because their strategy was necessarily better, but because if enough people take the risk then it’ll pay off for one of them and that person will win.

Take this a step further, if you know that this is the case and only care about coming first (as opposed to caring about your overall place) then your best strategy is to take the risk. You know that the risky move is on average the worse move and that on average you’re likely to do less well in the game – but it still becomes the optimal move to take. The worst move becomes the best move because of the environment that it is in.


This property of multiplayer environments has some interesting emergent effects. For instance when you see worse players playing high risk moves you might conclude that they are playing unnecessary risky moves because they’re a worse player. However if you imagine the competition above in which your opponent gets 2.1 points for the reliable move and you get 2.0 then it’s impossible t win without taking risks – When a player knows that their opponent is better than them high risk moves make sense because they offer the only option to win. As a result someone who’s only slightly worse might appear dramatically worse despite taking the optimal moves for themself in that environment. Furthermore playing optimally in *this* game hampers their ability to ever learn to play the game well in a reliable way. In a multiplayer environment this could lead to all players picking suboptimal high risk strategies due to the effect described above.

As a game designer you need to think about what the decisions you make will do in terms of how they affect people who play your games.


So that’s the game design theory for today, what’s any of that got to do with Kickstarter? Well the platform can be considered as a multiplayer game that all of the KS creators play together. The rules aren’t entirely clear and in some ways it is a cooperative game (People bring their audiences and grow the platforms, best practices are shared for the good of everyone, creators promote other projects they like and generally make things nicer for each other) and in other ways it’s a competitive one (People only spend so much money and will tend to compare projects to similar projects when deciding what to back).

This second property sets an evolutionary path for KS projects which I don’t think is good for the platform, creators or consumers. Lots of people use the platform and some of them will take risks. The risks create unstable projects which may deliver above the odds of fail spectacularly. Being successful requires joining the group of risk taking creators as what an individual can reliably deliver is smaller than what an individual might deliver by taking a risk and backers gravitate towards the projects that offer them the most stuff.

That’s a bit abstract, so let’s get specific and talk about funding goals. Time and time again it’s been demonstrated that low funding goals are attractive to backers, it’s conventionally accepted wisdom that a lower funding goal leads to more funding overall – as people are more likely to back a project that they think will succeed and some people won’t back until the goal has been met. In principle any person with a successful project could have had an even more successful project if they’d set a lower goal. So everyone should set their goal to £1. If nothing else you’d be thousands of percent overfunded.


Except that’s obviously a stupid thing to do because if it costs you £20,000 to make your game and you happened to fund at £10,000 you either need to lose £10,000 (If you have it) or admit that you can’t deliver on your promises (If you don’t). Since nobody knows how well their campaign will do in advance you’d be taking a huge risk to build a campaign that requires a high goal but set a low goal on the platform.

Which you may remember as being similar to our hypothetical game setup above and I think it’s what’s happening. I see a lot of projects that clearly couldn’t deliver the game they’re offering for their goal and that have set it with the intention of over funding. It seems a logical consequence of backers preferring to back projects with lower goals and driving traffic away from a stable strategy of “make the goal match the amount required” and towards the risky strategy of “lowball the goal and hope to overfund.” This in turn leads to some creators failing with funded projects that aren’t sustainable – this failure may manifest in the project being cancelled at the last minute, the project exploding spectacularly and failing to deliver or simply the creator quietly taking a loss. More importantly, some creators succeed with the strategy, attracting enough backers to overfund and having the model works.

This seems dangerous to me because everyone involved feels that they did well. The backers feel like they picked a winner, because they backed a project and got what was promised on time, the creators feel that they chose wisely because ultimately their strategy worked. This makes the environment hostile to a responsible creator because backers now expect more than a sensible offer, old competitors successful enough to take a second swing believe in their risk strategy and new creators who base their plans off successful campaigns without studying failed ones take the same approach. As an environment it’s fair to suggest that while everyone makes sensible seeming decisions (do what successful creators do, do what worked last time, back projects with the best offering, feel good if an action succeeded) the big picture emergent property is that the platform becomes less stable, risk taking increases and cancelled or failed projects become more likely.


Except for one thing: This model is based on a competitive environment in which coming first is the only thing that matters.

As I mentioned earlier, in a lot of ways Kickstarter is more similar to a cooperative model in which creators learn from each other and benefit a mutual environment. It’s also the case that coming first isn’t the only thing that matters, it’s not like having the second most funded project on Kickstarter would mean that your campaign didn’t fund. We, creators, collectively generate our own destiny.

While the effect that I described could be applied to any risk taking behaviour: Offering things at rock bottom prices to try to make it up in quantity or gambling on stretch goals that weren’t planned in advance doing enough to save themselves – but I picked on the goal for a reason: If the goals are reasonable than that mitigates the risk from most other parts of the project. A gamble that you can produce ten thousand copies of a game doesn’t poison the well if your goal reflects that you need to get enough thousands of backers in at your base level to support it. If upon hitting the goal the project was already paying for itself then there’s a limit to how much damage you can do to yourself with all but the most outlandish of stretch goals. I feel that if creators, collectively, make an effort not to lowball their funding goals then the other parts of the environment won’t be pulled into patterns of unnecessarily risky projects undermining the platform by virtue of the higher rewards these risks offer being me attractive than a careful campaign.


I know that it’s not so simple as that, the existing environment is what it is and being present there requires a degree of compromise. I’m more pushing in the direction of trying to recognise that the collective result of individually sensible decisions is sometimes bizarre and trying to consider our impacts of our actions on this scale. This is a lesson that applies on every level, to game design as discussed at the start of the post, to KS creation as in the body and to many aspects of life and human civilisation in general.

Or to put it differently: Try to see the whole board.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>