How many quarts are necessary?
Wild guess here, but it's probably tribal knowledge. I mean, an OEM doesn't just hire some recent grads and tells them to design a new engine. People on the teams have been doing this for years, and have done many tests over the years, and probably have reams of data going back years as to how their engines fare, in terms of oil temperature, bearing wear, and so forth. They can probably look at the desired specs for an engine and guess what the minimum amount of oil required will be, then work from there. Eventually it goes on a dyno and gets flogged, and if wrong, well, they either increase capacity or ask for a new oil spec.
That said, "How we've always done it" started someplace in past. Old engines needed more oil for various reasons. Like other things, sometimes what worked in the past gets copied forward, and either no one thinks about changing, or no one wants to put in the effort to change. 4 quarts versus 3, just how much effort would it require to test that out? Hours of dyno time isn't cheap.
