This isn't the first time EA has seriously blown a launch due to infrastructure issues I realize. EA's main infrastructure server team runs things like a pre-dotbomb forture 5.
I'm slow to throw blame since EA's acquisitions chose to join EA. Technical Directors at Maxis are just as much to blame as well as product/project managers that fail to squeeze load testing into the busy schedule.
I do (sadly) agree that studios seem to lose their magic to us gamers once they join EA. Why?
I'll give you one guess.
Not quite correct. Maxis' shareholders decided to join EA; I doubt the average programmer, or even the average project manager or tech lead had any say in the process whatsoever.
One of my biggest disappointments with the games industry is that those editors never really made their way into the industry at large. One of my greatest hopes after Spore launched was that some company would license those editors and actually build a good game around them. A Dungeon-Keeperesque game with fully editable creatures and dungeon design was a particular dream. Sadly, it never really caught on, and Spore looks to be a mostly dead end.
http://en.wikipedia.org/wiki/Spore_(2008_video_game)
Unfortunately, the citation link is dead on the article. I don't know if this answers your question about commercial success. Maybe someone knows more about this data point in context of the industry.
I think the respective records of EA and a pre-EA Maxis ought to provide ample evidence of the likely culprit here.