Brewer,

It comes down to hope for the best, plan for the worst. Or in this case, plan for a sort of middle ground.

Whether you want to accept that low predicted returns for the next few years is true or not, or instead believe that that is taken into account with the SWR studies, it can simply come down to it not being safe to plan on the best case scenario of what is safe based on high past returns.

I agree - how about if we do the opposite?

How about if we plan on the WORST case scenario of what is safe based on LOW past returns? Specifically, the worst 30-year period in the past century (and longer)?

If you do that, you get...

Intercst's safe-withdrawal-rate study.

The one you say is wrong.

The one that says an initial withdrawal rate of a bit under 4% (with the withdrawal amount adjusted for inflation in subsequent years) from an efficient-frontier portfolio, survived the worst 30-year period in the past century (and more) with the final-year withdrawal unmolested.

The one that doesn't talk about best case, and doesn't say much about average case, but is heavy on worst case.