I’m not sure where your data is coming from, but according to the 2016 social security life expectancy tables, the probability of a 62 year old woman living to 92 is less than 25%. However this is not my main issue.
I probably didn’t explain my point clearly enough. Your post showed an interesting analysis that seemed to indicate one could spend more money over a 30 year retirement by taking social security at age 70 over age 62. However, if we look at historical returns since 1928 for a 60/40 portfolio, that has only been true about 23% of the time. (I actually did the calculations when I was trying to decide when to take SS). The reason your analysis looked different is because you used a 4% withdrawal rate, which is the historical worst case withdrawal.