I couldn't find the ingredients list -- but because it is not fully understood what food contains there may be missing nutrients in soylent that a well-balanced diet would have.
This is somehow the same distinction between mortality and morbidity. IMHO, mortality stats are pretty useless and we should always look at morbidity stats.
Glib/nitpicky I know, but all stats are useless in some context - the trick is to get the right tool for the job. Dismissing an entire dataset as useless is just missing the point.
On an individual basis, I will rather be dead than unable to have an autonomous life, but this is completely a personal preference.
http://cdn.shopify.com/s/files/1/0421/5993/t/14/assets/files...