I don't mean to be argumentative, but I still think you're making it sound a lot easier than it is. For example, I know more about iPhone dev than most people who aren't full-time iPhone devs, but I have no idea what is a "known winner" category or even how I'd go about guessing. AFAIK, Apple doesn't release the kind of numbers I'd need to figure out what categories are profitable. With a web app, we can use analytics, keyword tools, A/B testing, etc.

This feeling — that releasing for the iPhone is a crapshoot and that anything I release would probably be crowded out by the absolute flood of junk in the App Store — is why I've never bothered with the iPhone, despite having done Cocoa development since the early days of OS X.