Looks very accurate to me. The biggest problem is it's based on a "die broke" theory -- and if you live longer than you estimated, you're out of money.
It gave me about a 5% SWR (given a 6% returns and 3% inflation assumptions), which is obviously high compared to the standard 4% SWR advice, but the standard 4% is based on the idea of you never running out of money (gives a 95% chance to not run out after 30 years). This will give you a higher potential failure rate, but allow you to spend at a higher rate (and/or retire earlier, needing a smaller nest egg at a 5% SWR) so that you die broke.
I'm not a huge fan overall, but it gives the ER layman a target to start.