Nereo, I'm not sure how you're calculating your returns, but if I use cfiresim to look at the return someone earns on average over 30 years if they're annually saving, I get the following values for a 100% stocks portfolio:
Min 0.91%
25th 5.66%
Med 7.03%
75th 8.80%
99th 10.92%
...
Seattle - you seem to be doing something a bit different from what Andysandp and I are discussing (or at least what I am discussing).
I'm talking about the annualized real returns on a given 30 year period.
For data I've been using Robert Schiller's data for SP500 and for inflation (mostly out of convenience) the CPI. Dividends are reinvested.
This is going to be different than if you are saving annually. There's so many different inputs you could use with cFIREsim that you'll have to walk me through which you used in order for us to have a productive conversation.
From my calculations, the stretch that ends in 1920 (i.e. 1890-1920) returned -1.44% before dividends, and +3.47 with dividends reinvested.
If I misunderstood and you're looking at Jan 1891 until December 1920 the results are pretty similar: -1.67% without div, +3.3% with dividends.
In effect this is basically "real return on $1 after 30 years".
I'd love to understand what your inputs were and how you are arriving at such low values. REgardless, both of our data indicates that this is NOT the worst period in history. Of course we could both be making large mistakes somehow...