There are probably a few levels. Originally, at small scale, you need to pick hits better than others (or get lucky). Next, you want to scale large enough tha

There are probably a few levels. Originally, at small scale, you need to pick hits better than others (or get lucky). Next, you want to scale large enough that you can make enough bets to amortize individual bet risk across a large portfolio. Then, once you're over that scale, you need to be back in the business of picking hits more reliably than the next VC.