A fast, efficient process for dividing game designers into groups

Talking about measurement is a highly effective way of dividing games designers into non-overlapping segments.   (An even quicker way to do this is to talk about experimentation.  That’s awesome.  But this one works really well, too. )

I had a fun time doing this last night at BAFTA’s games question time.    BAFTA lined up a great panel for the event, drawn from this year’s crop of BAFTA award game nominees:

• Dan Connors – CEO, director, founder, Telltale, Inc, The Walking Dead (@telltalegames)
• Mike Bithell – Creator and game designer, Thomas Was Alone (@mikeBithell)
• Jessica Curry – Composer and co-director, thechineseroom, Dear Esther (@jessicacurry2)
• Barry Meade – Co-founder and director, Fireproof Studios Ltd,  The Room (@Fireproof_Barry)

I’m really interested in the different ways that game dev teams collect and use feedback about play patterns in order to improve their designs.

So I grabbed the chance to ask the panel whether they had been able to look at how people played their games, and whether that had inspired them to change their designs.   I dropped in a reference to MBA’s swinging into the room on trapezes, equipped with machine guns, just to keep it real.  Just as I’d hoped – lots of fun ensued.

I’d summarise the reaction as follows – corrections welcome:

The Walking Dead  – used post-launch stats about which characters got fed in order to figure out which characters were most popular – so that they could later mess with your mind and change the story in response…

Thomas Was Alone  – did face to face  pre-launch testing with 10-20 people on each level, to see how it worked and what peoples’ reactions were, which was invaluable – “at the moment I’m just learning so I take all the feedback I can get”, was Mike’s reaction

Dear Esther  – set out to make a game that would have minimal possibilities for interaction and agency, and wouldn’t be replayed, but they found that people did replay it, sometimes 6, sometimes as many as 25 times – but that didn’t affect what they would do in the future, as they were pursuing a vision which was independent of people’s reaction to it, like all great art

The Room – looked at the stats but it didn’t tell them anything that they didn’t know – they knew people would get stuck on a certain level, and they did –  Barry’s take was basically that “we don’t have the type of game where metrics would help”.

Someone (Jessica or Barry, would be my guess 😉 –  said that anyone who changed something in the game because of a number was….. (I didn’t scribble down the exact phrase as I was too busy laughing,  but the gist of it was something a bit to the south of “not talented”).   I didn’t really take proper notes so there’s plenty I missed – but perhaps if BAFTA put the talk up online I’ll be able to add to this.    Corrections and comments welcome.

[Update March 6th:  highlights from the Q&A have been posted online on the BAFTA site.]

Advertisements

2 thoughts on “A fast, efficient process for dividing game designers into groups

  1. Argh!
    I’m crushed to hear that Mr. Meade (creator of my absolute favorite game) might not think facts could inform game design.
    I recently did a Delphi process of several dev teams with multiple choice answers to statements about why people would enjoy their game. It was intended to be a team building exersize but it revealed some dramatic and irreconcilable differences in the approach to game design. It was as though everyone on the team was building a different game. I then gave one team stats that showed a Simpson’s paradox in retention due to a quest system they’d just implemented. The group divided even further by interpretation of these data.
    What I had thought could inform ( not dictate) design was no match for the sanctity of personality in the design process. I really can’t pull off that GDC swagger that makes all game developers genuflect, but maybe a trapeze!?

    Like

    • I find that a compelling enough high wire act stops conversation for a few moments, but the effect is transient: designers are just as good at generating lots of opinions insights from the same set of facts as economists and psychologists are 😉

      Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s