If you're pulling $50K offa' a Million invested, that's 5% and a higher risk of failure than 4%.
That all depends on how you define "failure" doesn't it? A 4% SWR lasts, on average, forever. By which I mean that in more than 50% of the historical periods of all lengths, the portfolio survives without being depleted or requiring any spending reductions. A 5% SWR, by contrast, only
lasts an average of 49 years. If your life expectancy is less than 49 years, then shooting for 4% actually means you have worked too long. That's also a failure, isn't it?
The only way to "fail" retirement is to be forced into working more than you want to, and there are two ways to do that. 1) work too little, and then have to go back to work, or 2) work too much in the first place. We can't know in advance what our financially optimal retirement date is because we don't know what future market returns will be. If you retire before that date, then you'll need to work more in the future. If you retire after that date then you have worked more than needed in the past. Using a 4% SWR virtually guarantees (95% chance) that you fail the second way, and decide to work too long.
The closest way I can figure to finding that financially optimal retirement date is to look at the
50% success probabilities for each SWR in the past. That analysis suggests that someone like me, who is only planning to fund 23 years of retirement, should use a SWR of around 7% per year, and that someone planning a 30 year retirement should use about 6%. Those are the statistically correct SWRs to use, if you believe that the future market will look like the past market, recessions and all.
If the future turns out to be as good or better than the past, you will have picked right. If it turns out to be worse than the past, you'll have to go back to work or reduce your inflation-adjusted spending limits. I don't think either of those options are so bad that they're worth deliberately failing by working too long up front.