Flash of Steel header image 2

Top Selling PC Games of the new century

September 6th, 2006 by Troy Goodfellow · 3 Comments · Uncategorized

We are halfway through this decade, so it makes sense to have a list like this. Doing it by sales is certainly not as iffy a proposition as doing it by opinion or Gamerankings, but it’s certainly not perfect. The NPD numbers are the best we have, but are still not completely reliable, especially for online transactions. And Next Gen decides to start the century in 2000 instead of the correct 2001. (I’m feeling pedantic today.)

Sure, they cheat a little by only including the best selling game of a particular franchise. (If Age of Empires III hasn’t outsold Ultimate Mahjohng, Microsoft must be very unhappy.) It would be interesting to see the whole list laid out, though the top 12 would all be Sims games.

No major surprises for strategy gamers on the list. The sales of Civ III (No. 21) apparently outdid those of Civ IV, since the former makes the list and the latter is just lumped in with “other franchise titles” (which curiously includes the Activision Call to Power games.) Yeah, it’s only US sales and a sale based list will favor games that have had price reductions – the long tail of sales.

A look at the top ten reveals how lucrative the strategy and sim genre is. Six of ten are either Tycoon games, RTS or The Sims – which is in a category all its own. The impression of many that Age of Mythology was not a well-received mass hit is revealed as false. It was huge. Bigger than the Age of Kings expansion. Step back to do a top twenty look and we only get two more RTS added to the list (the aforemention Conquerors add-on to AoK and Command and Conquer: Red Alert 2.) But they’re 11 and 12 – not 16 and 19.

It’s also worth noting how few of the top 50 are what you would call “bad games” (judging by the collective wisdom via Gamerankings). The successful games that score under 75 percent (the bottom line of good for many people) are movie tie-ins or riding on the crest of a cultural touchstone like Who Wants to Be A Millionaire. Find a list of the top 20 grossing movies for this century and you will see a lot of crap. Same with books. Same with TV. Maybe it’s because games are really hard to screw up so bad that your GR score will dip below 70 – lots of sites don’t venture even that low unless it’s time to pick on Deer Hunter (no. 81 on the list, GR score of 71.)

I won’t repeat their entire list – that would be wrong, and a waste of my time since you can read it over there. Children’s games do well, game show games do well (Survivor is no. 77) and the hits you expect are there.

Reflect, ruminate and comment below.


3 Comments so far ↓

  • Jozef

    The statistics guy to the rescue! I’ve run some statistics on the top games, and found some interesting results. Not necessarily insightful, but still interesting.

    The top 100 games accounted for only 29.7% of industry sales over the past five years. The top 10 games accounted for only 9.8%. For me, this indicates that the industry is still very fragmented. Unlike movies or music, which already reached the 80-20 rule, PC games have still a long way to go.

    What is selling games? It certainly doesn’t seem to be reviews. The correlation between reviews and game sales is a mere 19.96%. Still, higher ratings cause higher sales, but only barely. The price of games was a slightly better indicator, with a correlation of 24.58%. Actually, this surprised me: I expected more unit sales for cheaper games, but that’s not the case. I assume that in this industry prices are still indicative of quality, and thus people are still willing to purchase games based primarily on perceived quality. However, what really seems to be driving the sales is an established franchise. Franchise sales (minus the sales of the Top 100 games) are correlated to game sales at 76.65%. This may be indicative of a McDonald’s model: customers tend to return to familiar experiences.

    However, without more detailed statistics, all three correlations are quite unreliable. For example, in the case of franchise sales the list often showed one of the game expansions, not the primary titles. The list also doesn’t take into account external tie-ins, such as movies of TV Shows, which could exmplain the popularity of games based on Spongebob or Barbie.

    On a side note, I found it interesting that the correlation between review ratings and the price of the game was 57.38%. It seems that price is indeed indicative of quality, or that reviewers perceive more expensive games as better. In either case, buyers don’t seem to be as enthusiastic about the perceived quality.

  • Troy

    Thanks for doing the leg work Josef. Of course, with only one title per franchise on the list, we really can’t read too much into those numbers. Civ IV, AoE 3, Baldur’s Gate: Throne of Bhaal…all huge sellers with huge ratings whose figures are just lumped into the generic “franchise sales” thing.

    Were you using the franchise sales numbers for your second paragraph or just the game statistics? Because then I think the skew would be quite different. Probably not 80/20, but closer than it is in your analysis.

    (I’m assuming you are using .05 statistical significance?)

  • Jozef

    I’d have to do some regression or compare means to work with a .05 significance, but don’t think I didn’t consider it ;)

    For comparing sales numbers in the second paragraph, I used the total revenues as presented in the article. Those revenues are only for the games on the Top 100 list. I was thinking about multiplying the franchiswe sales with the average sale price for each game, but I noticed that some of the franchise games were released in the last century. In addition, given that much of the franchise sales comes from expansions (especially true for The Sims), the average game price, as derived from the list, is wrong.

    But just for your amusement, doing this calculation yields 88.3% of all game revenues coming from the Top 100 games and their franchises. Pretty close to 80% ;)