[robocat]

“Input factors” have many imperfections, but at least they can lead to more objectively comparable metrics.

When comparing between populations, subjective measures such as surveyed happiness make for very poor and very noisy metrics - you often can’t even simply compare answers between different populations within a country. It is even dependent upon your age demographic https://archive.ph/sAzDA and there are some surprisingly counterintuitive properties to reported happiness: https://en.m.wikipedia.org/wiki/Hedonic_treadmill https://en.m.wikipedia.org/wiki/Easterlin_paradox

I would think that life expectancy is a poor metric, because: (1) the differences become smaller as it rises so it becomes dominated by noise, (2) it overvalues the last few years of life compared to the rest of a life, and (3) expectancy is highly dependent on how it is calculated along with parameters such as input factors, similar to those that you dislike!

In summary: intuition can often be a very poor substitute for research, and before relying on your intuition it is a good idea to look at the basics.