or is it the opposite?: Maybe it's that in baseball, BPA 2-5 yrs out is so impossible to determine with any degree of probability that it's pointless to go down the rabbit hole of trying to make razor fine distinctions to rank many similar players in the absence of anything like MLB relevant competition? Sure there are mere handful of guys that break away from the pack in a given year - but always far less than the number 1st round picks and thus only relevant to a small minority of the 30 teams. IIRC, when people have broken down the MLB draft by total WAR by draft position things get pretty flat after only about the 10th pick or so.
It would be interesting to see that same WAR vs draft position data reparameterized by team. It's often argued here that it's the teams doing drafting that is the better predictor of performance than draft position. Of course there are two possible explanations for that, which are not mutually exclusive: Some teams may do a better job with their internal BPA rankings than the conventional published wisdom (i.e. they are getting better players in lower rounds) or they simple do a better job at skill development - i.e they create their good players - at least to some degree.
And then the third thing that all the concentration on the US draft misses is that nearly half of the players are coming from the international system, and being good at working that system is a whole different set of competencies for ML teams.