My Message close
GAME JOBS
Latest Blogs
spacer View All     Post     RSS spacer
 
May 19, 2013
 
Making 2D Games With Unity [1]
 
All You Need is Love [3]
 
Students: Tips for Learning Game Development Over the Summer [2]
 
All Your Nintendo Let's Plays Are Belong To Nintendo? [85]
 
Even Further Down the Curation Rabbithole [12]
spacer
Latest Jobs
spacer View All     Post a Job     RSS spacer
 
May 19, 2013
 
Sony Computer Entertainment America LLC
Sr. Network Systems Engineer
 
Treyarch / Activision
Technical Animator
 
Amazon Game Studios
Sr. Game Designer
 
Amazon Game Studios
Quality Assurance Manager
 
Amazon Game Studios
Lead 3D Environment Artist
 
Amazon Game Studios
Game Graphics Engineer
spacer
Latest Press Releases
spacer View All     RSS spacer
 
May 19, 2013
 
Zeeek and The Secret of
Space Octopuses heading
to...
 
Battle bad 'bots in Bad
Bots, available now on...
 
Temple Run 2 Adds New
Terrain and Obstacles
in...
 
Little Amazon runs
through Android
 
Command Ops gets a
Massive Update!
spacer
About
spacer Editor-In-Chief:
Kris Graft
Blog Director:
Christian Nutt
Senior Contributing Editor:
Brandon Sheffield
News Editors:
Mike Rose, Kris Ligman
Editors-At-Large:
Leigh Alexander, Chris Morris
Advertising:
Jennifer Sulik
Recruitment:
Gina Gross
Education:
Gillian Crowley
 
Contact Gamasutra
 
Report a Problem
 
Submit News
 
Comment Guidelines
Sponsor

  The science of predicting MMO player churn Exclusive
The science of predicting MMO player churn
 

August 30, 2012   |   By Staff

Comments Post A Comment

More: Social/Online, Business/Marketing, Exclusive





In a new feature, MMO publisher Innova's head of analytics, Dmitry Nozhnin, writes that while it's harder to track when veteran players are going to quit your game, it's possible -- and worth it.

"For new players, defining churn was dead simple -- they just leave the game after a couple of minutes or hours. That's it. The last day of play was clearly defined, and data mining models on such churn factors were already well established," writes Nozhnin.

The question is much more confusing when it comes to veterans, though.

"Manual data investigation revealed that majority of churners have a 'long tail' of play days -- those occasional activity days during several weeks, or even months... They effectively stopped actively playing the game, but still log in from time to time."

These players may jump in to check auctions, chat with friends, or even pass their account along to guildmates for use -- but if you detect their behavior that late, they're already gone.

It turns out, however, that it's simpler than you'd expect to figure out when those players will quit.

"Most fascinating is the fact that final data mining models with best precision were entirely based on derivatives and calculations of only two metrics -- days of activity and daily playtime!" writes Nozhnin.

The full feature, in which he goes into his methodology for creating the analytics -- is live now on Gamasutra.
 
 
Top Stories

image
The laws behind Nintendo's Let's Play crackdown
image
New layoffs reach Trion
image
How developers mess up immersion (you might be doing it wrong)
image
Steam Trading Cards: The next-gen of achievements?


   
 
Comments


none
 
Comment:
 




 
UBM Tech