Games from 2000: 32
Games from 2001: 31
Games from 2002: 16
Games from 2003: 12
Games from 2004: 8
Games from 2005: 1
I'm trying to reconcile two ideas:
- The Windows PC gaming market is not dying, may in fact be growing.
- The bulk of a game's sales are made in the first 3 months after launch.
However, the explanation at the end of the list says that "[g]ames are ranked by units sold; then by revenue generated" and that "[t]his tends to favor games released earlier, rather than later, and games which have spent a long time at a lower price point". So maybe the old rule of thumb that a game gets most of its sales in the first 3 months after launch just isn't true. Perhaps in the long run more games are profitable than it appears just from what they recoup in their first year, or even two years?
A couple more factors to figure in (which may not be readily available information):
* How Many PC games actually tracked by NPD data were released each year? It's possible that while the hard numbers of units in the list drop dramatically, the percentage of PC games released for that year that make the top 100 aren't dropping quite as dramatically.
* What I'm guessing just eyeballing the data is that a lot of those 2000 / 2001 era games are actually more "evergreen" casual / non-hardcore titles that do not follow the three-month model. Jeopardy, Who Wants To Be A Millionaire, Mahjong, Deer Hunter... these are games that are not "front-loaded" like core games. I imagine they probably sold almost as well their second and third year as their first year --- and at the same price (not discounted). And these look like they are making up a lot of the 2000 / 2001 game sales. Separating the games into "core" and "non-core" games (which might take a judgement call) might reveal interesting results.
* Weighting the games by total revenue generated might reveal interesting trends, as well.
I think the whole "PC Gamed are dying / dead" thing is in part a self-fulfilling prophecy --- enough people believe it (especially publishers), and it becomes true.
It is interesting to me that the NPD recently revised it's sales figures for PC games to include income from subscription-based games and (some) downloadable game sales, and their total sales revenue for 2005 jumped up by almost 50% (from less than 1 billion to 1.4 billion).
I believe PC games are transitioning, not dying.
Coyote: All good points. I don't have access to the data necessary to answer this question definitively and I could have been more clear that I was just thinking out loud. My gut feeling from the data is that the 2003 and later numbers should be higher than they are, but factors, such as those you've noted, would certainly skew things especially for what you've called evergreen titles. Even a game like The Sims has had long-term appeal, it would seem.
The three-month deal likely still holds, and might be a good measure of how long it takes for people easily influenced by advertising to act on that influence, or how long it takes for the influence to wear off once it's delivery is stopped.
Wish I had the whole list in front of me... wonder if there are any Maddens, and how the extra time adds to sales, if any.