That quadratic behavior is only true for high Reynolds numbers, because the drag coefficient in https://en.wikipedia.org/wiki/Drag_equation https://en.wikipedia

That quadratic behavior is only true for high Reynolds numbers, because the drag coefficient in https://en.wikipedia.org/wiki/Drag_equation https://en.wikipedia.org/wiki/Drag_equation is itself dependent on the Reynolds number (and hence on the velocity). The quadratic behavior actually sets in at _lower_ Reynolds numbers for less-streamlined objects. Now as it happens the Reynolds number for a large cargo ship is almost certainly quite large, because it involves the product of the ship's velocity and some linear measure of its size, and container ships are huge. https://en.wikipedia.org/wiki/Reynolds_number#Similarity_of_flows https://en.wikipedia.org/wiki/Reynolds_number#Similarity_of_... lists ~5e9 for a large passenger ship, and the quadratic behavior sets in somewhere in the 1e3-1e7 range depending on how streamlined you are. So we're well into the quadratic regime in this specific case. But assuming that you're there in all cases is not a great idea. For example, it's not clear to me that a walking human is in the quadratic regime (though I am pretty sure a running human typically is). Small fish are typically _not_ in the quadratic regime, I expect.