They say "Google doesn't like "old" content? That's not a thing!" But who knows, really? They run things to extract features nobody outside of Google knows tha

They say "Google doesn't like "old" content? That's not a thing!" But who knows, really? They run things to extract features nobody outside of Google knows that are proxies for "content quality". Then run them through pipelines of lots of different not-really-coordinated ML algorithms. Maybe some of those features aren't great for older pages? (broken links, out-of-spec html/js, missing images, references to things that don't exist, practices once allowed now discouraged...like <meta keywords>, etc). And I wouldn't be surprised if some part of overall site "reputation" in their eyes is some ratio of bad:good pages, or something along those lines. I have my doubts that Google knows exactly what their search engines likes and doesn't like. They surely know which ads to put next to those maybe flawed results, though.