[MUD-Dev] Codename Blue & Facets - Nick Yee's new studies

Dave Rickey daver at mythicentertainment.com
Thu Apr 25 11:21:56 New Zealand Standard Time 2002

From: Koster, Raph <rkoster at soe.sony.com>

>  The stickiest factor is Relationship, NOT Leadership (which as I
>  mentioned, has a lot of "cooperation" involved in it). Both EQ
>  and DAoC score higher on the Leadership factor, because UO
>  doesn't offer plane raids or relic raids.  And DAoC, which shows
>  evidence in the study of having the poorest likely retention, is
>  actually the top game for Achievement.

I actually found our high score on Achievement surprising, we
weren't really trying to make an achievement-oriented game.

>  Lesson for us developers: Build relationships. Not teams, and not
>  advancement ladders.

I think there were some factors that skewed some results.  He took
the samples concurrently, which means that for UO he was sampling a
4.5 year old community, for EQ a 3 year old community, and for
Camelot a 6 month old community.  Without equivalent data points for
the other games, it's hard to say how much the time-correlated
factors (retention, proportions of players intending to stay
long-term, female/male ratios, and secondary effects of all of these
such as their impact on "relationship" weighting, etc.) were
affected.  Also, the first few months of an online game's existence
tend to be atypical in a variety of ways, EQ had a much lower
retention rate in it's first six months than it has now, and UO's
was less than 50% for the equivalent period, from what I have been
told (Camelot's is much higher).

I'm not trying to be an apologist here, I just think that it might
be very dangerous to draw too many conclusions from a comparison of
data sets that are not equivalent.  I genuinely believe that once it
shakes out, "building teams" will turn out to be a *very* effective
means of building relationships.

Just as an example: A year or so ago I made the statement that I had
never seen cohesive player organization beyond the 150 player range.
I can no longer say that, the last couple of months have seen the
emergence of cohesive player guilds in Camelot exceeding 400 members
(discrete accounts, not characters).  This can't be accounted for as
players with multiple accounts, his data indicates only 1 in 10
Camelot players has multiple accounts, so these guilds are well over
300 players.  In addition, we are seeing the formation of *large*
"Alliances" (game-supported links between guilds), some containing
nearly 5000 players.

Unlike the 150+ plus player guilds in EQ, and the 2000+ player
monarchies in AC, these organizations seem to be *stable*
formations, rather than extreme examples pushed way out the entropy
curve and in constant danger of dissolution.  In fact, the limit of
21 guilds in the alliances is causing a steady trend towards larger
guilds, in order to get more players into the same alliance.


MUD-Dev mailing list
MUD-Dev at kanga.nu

More information about the MUD-Dev mailing list