Thanks, the one from 2016, although much greater in BTC terms, was around 4 times lower in USD terms than today's one.
The 2016 transaction was widely reported as a sender's error, so not sure why would you dismiss the possibility of someone making a similar mistake today. I can't imagine anyone willing to send a "donation" to a random mining pool this way. Just sounds way less probable than a human error.
And if my memory serves me well, I think in 2016 the pool agreed to return the fee (but I may be thinking about yet another situation).