>1) Have one of the outputs be a box and whisker output showing the percentiles of ages the person will run out of money.
This is essential, and I wish it hadn't been overlooked. The only power of stochastic modeling over deterministic modeling is in quantifying the likelihood of outcomes in the outcome space.
Unfortunately, this requires either crazy stochastic math, or multiple computations (monte carlo methods). I suspect the op is a great guy and just hasn't been exposed to these ideas yet. When he does, he's going have that head-slapping moment we all have when we first encounter it.
1) Love the idea. I think that would be a great visualization.
2) Agreed, a simple function for this would be nice. You can currently doing it using some math based on the year, but it is a pain.
3) I think this would require collecting more demographics data to have a reasonable level of accuracy. I think the added complexity in UI might not be worth the insights.
Ideas to play with: 1) Have one of the outputs be a box and whisker output showing the percentiles of ages the person will run out of money.
2) Have a way a person could easily code a transition from an S&P500 portfolio to a fixed income portfolio over 5 to 10 years.
3) Add a generational mortality table on the back end and include a probability of death.
edited to clarify (2)