The purpose of metrics in a game

(I originally wrote this for my blog at http://psychochild.org/?p=1407)

Years ago people stumbled across the idea of gathering data about players.  A few MMO developers realized they could gather information about what players were actually doing.  After all, all player activities had to be verified by the server, so recording that information and storing it for later was a direct process.

Then social network games sprang on the scene.  Driven by web development which had been collecting metrics for years, this idea became a no-brainer.  Measure everything and draw all sorts of conclusions about what the players really want, with an eye toward maximizing revenues.

So, what has happened to metrics in modern game design?  Let's take a look.

Craft vs. Metrics

Chris Bateman wrote a blog post (http://blog.ihobo.com/2016/07/the-craft-of-game-design-cannot-be-measured-by-any-metric.html) where he talks about the role of metrics in game design.  His thesis is that game design is a craft, and trying to define and confine it with metrics misses the point.  While he predicted that game designers should try to understand players, he says that he didn't quite see the rise of metrics.

Chris argues that modern analytic metrics are focus on "analysing where the leaks are in their cashflow, and acting as digital predators to suck spare change out of players’ digital wallets."  He laments that instead of trying to understand the player and deliver a game that meets the player's needs and desires, a focus on profit has displaced the craft of game design.  Some parts of the game industry has replaced a need to understand the players with an focus on understanding how to maximize income for a game.

His conclusion is that this focus on profit has turned what game designers do into less of a craft of creating joyful experiences and more like a soulless pursuit of profit.

Art vs. commerce

In a way, this argument can be seen as a rehash of the old "art vs. commerce" debate.  The extreme version of the argument goes that art is polluted by the pursuit of filthy lucre.  An artist should create art for the sake of art, and sellouts are those that profit from their work.  A more moderate take is that when you do creative work with a commercial focus, you have to compromise on creativity; it's incredibly difficult if not impossible to both be creatively unrestrained and serve the tastes of the public.

To be clear, I don't think Chris is really advocating abandoning capitalism and demanding game developers give away their labor for free.  Definitely go read his post, as I think he brings up a lot of nuanced arguments.  But, I think it is useful to look at his post through the lens of this old debate.

I dislike the absolutist nature of the argument, and prefer the more nuanced version.  As a creative person, I still like things like food, a roof, and perhaps air conditioning when the temperature and humidity get high outside.  But, I think it is important to realize that there is a decision to be made.  One can choose to pure creative energy to create experiences on one extreme, pandering to tastes and maximizing for profit on the other, and a lot of room between the two extremes.  And, as much as we might lionize the indie iconoclasts, the reality is that sometimes it takes a lot of work and understanding what people actually want to survive as an indie.

Fun vs. profit

This discussion might seem a bit familiar if you've read my blog for a while.  I wrote about social games and the focus on metrics before.  I'll draw particular attention to a comment I left later pointing to a talk given by Brenda Romero (née Brathwaite) and Laralyn McWilliams about the balance between metrics and the craft of game design.  (Note this is the same Laralyn that Chris references in his post, quoting how she grew disillusioned with "designing for 'friction'".)

The lesson here is that metrics are another tool for a game developer.  You don't have the binary condition of either metrics or the craft of game design, but they can both work to build better games.  Metrics tell you what is happening, but can't explain why.  It still takes the craft of game design to understand the reasons why players do what they do.  Otherwise you get quotes like I put in the blog post, where the people running games have a shallow understanding of game design and think, "if a player repeats something, it’s fun."

Of course, not having game designers isn't a sure recipe for failure.  People can stumble across good game design, and just as with the traditional game industry, cloning games can get you quite a bit of income from the game playing audience.  It's a game of probabilities: you're more likely to succeed in the long term if you have someone who knows how to retain players by giving them a joyful experience.  The single-minded pursuit of profit via metrics is more likely to turn people off, and can make it harder to follow up an initial success with another one.  It's telling that most social games that focused on metrics are floundering and few companies were able to have more than one notable success (outside of acquiring other companies).

But, again, a good game designer shouldn't discard metrics just because they can be used poorly.  Using metrics to help you understand what players are doing so you can start to understand why players do what they do in your game is too good to ignore.  Augmenting the craft of game design with data strengthens the power of good game design.

Metrics vs. the future

Ever so quietly, metrics went from a crazy idea to the foundation of some of modern games.  It took some outsiders to really show the power of metrics, but it will take experienced game designers to understand how best to use this tool.  Chris Bateman is right in saying that metrics have been abused for a while, but game designers need to realize that metrics are a good tool that provide useful information.  It can cut through the haze of what players say they want (or, at least, what a vocal minority say they want) compared to what players actually behave while playing.  Of course, you still need good community management because those vocal people are likely big supporters of your game that you want to keep happy, even if their demands happen to be in the minority.

About Brian Green:

I'm an experienced MMO developer, best known for my work on the classic MMORPG Meridian 59. I've done consulting and contract work for many companies. My professional blog is at http://psychochild.org/.  You can also follow me on Twitter @Psychochild or view my LinkedIn profile at: http://www.linkedin.com/in/psychochild.