This just shows how goofy the premise of FireCalc is. That a fundamental question like this could engender a long discussion with a lot of differing ideas makes this clear. Unless someone is a proven successful trader, then what counts is what Warren Buffet calls "look through earnings". When you retire you aim to replace your labor earning power with the earning power of your portfolio. What somebody is bidding for that earning power is meaningless, unless you are selling it to them at that price.
Goofy? I really don't see that, though clearly it is only a look at history. It provides information, what you do with it is another thing. I'll probably move to some sort of RMD-like weighted approach at some point, but right now, the historical reports from things like FIRECalc make me feel pretty comfortable with a conservative ~ 3%-3.5% WR
Also, the idea that Firecalc could contain some sort of ratchet mechanism that, like a come along, allows you to increase your spending but does not compel you to decrease it is very strange on the face of it.
Ha
Many things are strange on the face, until you dig in and understand the underlying premise. It took me some time to get my head around the 'ratchet up but not down' approach, and have the light bulb come on. But when applied to the historical data, it absolutely works. It
has to, by definition.
Another simple way to look at it, (again, I like to use a 100% success WR rate to keep this simple), FIRECalc is conservative, and reports what will succeed in
any period. A large number of periods will end up with as much or more than you started with, sometime much, much more. Each of those periods would have succeeded with a higher WR% than the conservative 'succeed all periods' number.
So if you started out in one of those many 'good periods', you can increase your WA (Amount not % - edit/add - but it is a higher % of the original portfolio), because we know that % succeeded every period in history, and this is now just one of those periods. And it has already passed, with any downturns that follow, with no reduction in WR%. Like I said, it has to, by definition.
In fact, if you use a tool that supports this mode, you will see that it is absolutely provides the best results of getting the most out of your portfolio w/o failing, and leaving the least 'on the table'. But that is also getting into data-mining territory, as it is really matching withdraws to that history.
But more pragmatically, for me, a conservative take on the data provided by these calculators tends to converge on any other conservative analysis. A 3-3.5% WR is pretty much in-line with a spend the distributions approach.
-ERD50