As a PhD student major in statistics, this question actually has bothered me for a long time. I am always not quite sure if it is appropriate to discuss this somewhat sensitive topic publicly. Finally I decided to bring it out because today when I was reading a proof of paper A, it cited one result of paper B (then made a very obscure conclusion which contains some notation never appeared before). OK, then I looked up this result in paper B, astonishingly, this result in paper B also has severe, if not fatal, mistakes. (Again, I am not sure if it's appropriate to list the papers and journals published them, if there are no ethical problems, I can give them later).
It is the fourth year of my PhD, and I have read numerous statistical research papers. However, honestly, only around 10% of them can I finish reading completely. And the reason I gave up reading is that they have too many typos or too big bugs that I can't get away with.
It might not be an exaggeration that almost every paper I have read has more or less typos --- small typos are fine since anyway papers are written by people and people made mistakes. Nevertheless, some papers involve essential logic mistakes/blurriness that are beyond readers' fix ability (like I encountered today). Monographs seem also suffer similar quality problems. For example, when I went through a monograph in my research field, I took down the typos and mistakes I met. It has around 300 pages, I managed to reading until page 187 and found 111 typos (then I quit), which means more than one typo appeared every two pages. The author of the book is a leading figure in this field and I believe the press is also prestigious.
To summarize, my question is: Why are there many (not immaterial) typos and errors in publications?