He theorizes that pirates actually might promote the game. My Flurry numbers beg to differ.
My "unlocked" version of the game was pirated on July 15. Almost 2000 new users ran it that day, and all but about 10 of them were pirated copies.
So did I see my sales jump? I sold about 10 units that day, and 5 the next day. I don't think I've seen the numbers top 20 sales per day. If there are moral pirates out there, they didn't show up in sufficient numbers to make up for the devastation that followed.
How about downloads of the free version? Over the next two days, it crashed from consistently getting 750-900 downloads a day to about 350 downloads a day. And because of that drop, it fell off the charts and continued to crash down to 100-200 downloads a day.
So as nice as the theory is -- and as much as it might work for music, where the numbers are MUCH larger and there is some data supporting the theory -- it simply doesn't work for games.
In fact, it pretty much killed my game, because of the way the market rankings work -- the crash in downloads meant it lost position, which meant even fewer downloads, and a death spiral ensued.
So don't give me "piracy is actually good for your game!" when it killed mine. And it's still rated 4.5 stars, with tons of "Awesome game!" reviews, so don't tell me people don't like it.