[casual_games] New Issue of Casual Games Quarterly is available

[James C. Smith]

>> Interesting data from James Smith's Real Arcade article... 
>> only there's a fundamental flaw in the presentation of that data... 

You bring up some good points. Those are things I did consider but didn't
have any good solution to.  
You can do you own analysis along those lines using all the data I made
public.  If you go to www.game-sales-charts.com you can run the queries your
self on the current data and tweak some of the options. Or you can click on
the IGDA Article link on the site and download the full data used for the
article in an excel file.

On the issue of a games position in the top 10, if you go to the web site
and run the "Top Games" query it will have a column for 'number of weeks'
and also a column for 'score'. There is an option to rank the games by
either number of weeks or score.  The "score" is a simple formula that gives
more points to game that were ranked higher.  But the problem is there is no
clear answer for how those scores should be weights.  Is 2 weeks at #1 worth
more then three weeks at #2? What should the point difference be between #1
and #2?  Or between #2 and #3?  My "score" system is just a very simple
linear scale with 10 point for each week spent at #1 and 1 point for each
week spend at #10.  But for the article I chose to rank the games by weeks
rather than score because it is a more transparent method.  It is less
subjective. Any "score" system I devised could be tweaked to favor one type
of game over another.  More importantly, it just didn't make much
difference.  Most games fallow a similar pattern of quickly climbing to the
top of the list and slowly falling down.  The longer a game was on the list,
the more likely they spent a long time near the top.  When I compared the
rankings based on weeks vs. the ranking based on 'score' I ended up with
very similar rankings.  I decided it wasn't worth having people questioning
the score system and instead just published the raw number of weeks since it
ended up with nearly the same results anyway.


Your second point about the competition at the time of the games release is
also very relevant but has no clear solution. How much weighting should be
applied? It is more useful to look at a games performance compared to the
other games available at the time of it's release. For example, for a game
released in 2005, only compare it to other games also released in 2005.  You
can do things like that using the options on the web site or by downloading
the data and doing your own analysis. But I presented a simple summary in
the article to peek people's interests and let them dig deeper themselves.

There are a dozen different ways to summarize this data. I presented a few
in the article and more several more available on-line.  It is not perfect
but it is a great start.  It is wonderful to have access to all this data.
I hope you would share any results you get from analyzing the data yourself.

James C. Smith
Webmaster: www.game-sales-charts.com
Producer: Ricochet xxx, Big Kahuna xxx

PS: I am sorry my game-sales-charts.com web site is not very reliable
lately. I will be moving it to a new host soon.

-----Original Message-----
From: casual_games-bounces at igda.org [mailto:casual_games-bounces at igda.org]
On Behalf Of SCOTT HANSEN
Sent: Monday, January 09, 2006 12:49 PM
To: IGDA Casual Games SIG Mailing List
Subject: RE: [casual_games] New Issue of Casual Games Quarterly is available


Interesting data from James Smith's Real Arcade article... only there's a
fundamental flaw in the presentation of that data... 

First, it doesn't take into account the ranking within the top ten over the
life of the game... Game A can spend 10 weeks at number 1 and Game B can
spend 10 weeks at number 10, yet both would receive the same overall 'rank'.


Second, it does not factor in the increasing number of games. Earlier games
lasted longer on the top ten because there were fewer games to push them
off.

Wouldn't it be more relevant to have some sort of weighted average which
takes into account the proportional relevance of the rank within the top ten
over the life of the product? It can even be 'inflation adjusted' to take
into account the churn rate at the time it was active on the list... 

Scott P Hansen
Game Designer, MumboJumbo
shansen at udgames.com



-----Original Message-----
From: casual_games-bounces at igda.org [mailto:casual_games-bounces at igda.org]
On Behalf Of Wade Tinney
Sent: Monday, January 09, 2006 9:20 AM
To: 'IGDA Casual Games SIG Mailing List'
Subject: [casual_games] New Issue of Casual Games Quarterly is available


There is a new issue of the Casual Games Quarterly now available at
http://www.igda.org/casual/quarterly/1_2/

Please don't hesitate to send feedback, ideas for the next issue (which will
focus on game design), or questions. 

_______________________________________________
Casual_Games mailing list
Casual_Games at igda.org
http://seven.pairlist.net/mailman/listinfo/casual_games