I wondered about who's included in that 65% myself.
However, I think you're too much value to that 38% of American households own a console number and, it's households, not gamers (as you say in the last paragraph). According to the ESA numbers, it's 65% of households play videogames, so that 38% of households who own a console is actually ~58% of the gaming community (38/65). So, still a majority.
Besides the statistical inflation you've given to that number, you've also inflated it's importance. You're saying [console ownership == hardcore], but then again, we've got the introduction of the Wii, the casual gaming console of the now. I don't think the retirement homes qualify as hardcore, despite owning a Wii. So [Wii ownership != hardcore]. Also, (when you were still assuming that the 38% console ownership was of gamers, not households) you mentioned a platform to target that remaining 62% (actually 42%). The 42% of gamers who don't own a console, own a PC. That's the platform. And there are quite a few hardcore PC gamers. And they might give a care about Halo and God of War. Or, their PC ports, in any event.
Ultimately, I don't think you can draw any conclusions about hardcore gaming's future drawing power from the ESA studies. Not the way you've gone about it. The data does indicate that IF the ESA numbers are including "Peggle" fans and other casual gamers, then, despite that, PCs (which when combined with the Internet makes such gaming far more accessible) are still only a plurality of gamers (when we break down the 58% of gamers w/ consoles among Sony, Microsoft and Nintendo), not a majority. I'd actually suggest that's pretty cheering for Console gamers, because it means that people who focus on an exclusively console game (across all three platforms, though) have a larger market than an exclusively PC game.