I think the issue is the cost of #1 is non-linear. I think there's a space where yes, your impact on #1 is probably offset by gains in #2 and #3. But there's a

I think the issue is the cost of #1 is non-linear. I think there's a space where yes, your impact on #1 is probably offset by gains in #2 and #3. But there's also a space, where I'm at right now (and I suspect many new investigators are) where your impact on #1 either moves a staff scientist outside the realm of possibility, or simply dramatically increases the likelihood of stochastic funding failure unless you're literally god's gift to the research world.