> As you get older you've got students to teach/support, a department to run, grants to apply for I believe such problems are even more urgent in 20s and 30s.

> As you get older you've got students to teach/support, a department to run, grants to apply for I believe such problems are even more urgent in 20s and 30s. You're not established yet, you need to do a lot of things unrelated to your passion to make ends meet. Caring for young children is usually more time intensive than for older ones. By the 40s, academics have tenures and financial security. The children are older and don't require so much care. Yet statistically there's not much groundbreaking work appearing from older mathematicians. > The risks for going after something crazy instead of some iterative improvement / maintenance increase. That's not a big factor in mathematics where going after crazy things doesn't carry significant financial risks.