Ah the BPA discussion again.
So when we say BPA are we referring to likelihood of playing in the NHL or likelihood of being a star? 6A or 8D? No boom bust guys? When the Leafs traded down for Dermott it was likely because they saw a few similar prospects with little to choose between the them, rather than he was a great hidden talent absolutely nobody else knew about. There is BPA by 1-5% and there is tangible BPA (Rielly vs Koekkoek or Galchenyuk vs Grigorenko). If we are going to discuss BPA vs need we should agree on how much bias the "need" people are actually suggesting.
Need is a factor for every team should at least consider. In an extreme example, if you have all small skilled danglers, or all big grinders with limited explosive offense, at some point the identicals are blocking each other in your system (and the road blocked guys can't show their stuff enough to have strong trade value) so you end up trading for need and being squeezed. They have to decide between highest floor and highest ceiling too, so how is that metric so different from need? We can all agree that clubs don't always go for highest upside on every pick so BPA then becomes some kind of mix between absolute star potential, and chance of at least being a pro. Is BPA the 6A or the 8D? I am sure that varies from team to team, especially factoring intangibles like how deep is their team and how good has their drafting been the last few years? Some teams can afford to swing for the fences while others will need to save their jobs or at least get some more home grown talent in their system.
Talented defensemen with size will always be in the highest demand so their market value will make them more valuable than a similar forward with the same chance of success. A team that has high end young talent at every position salted throughout their system can go absolute BPA with any pick but that isn't most of the clubs in the league. Some people seem to imply the failure to take the 80% guy on your scouting sheet over the 79% guy is going to somehow make your club weaker. There is only a 20% chance any of these guys succeed so how many years of drafting would be required for a statistically significant deficit? Except for the top few picks the scouts are wrong almost all off the time. Compare draft position with success in most years and its clear.
If we have Suzuki, Valimaki, Foote and Kostim all there at 17 the scouting lists available to us said they are all very comparable, decent picks. They are all BPA depending on who you ask so throwing a dart is as likely to get us the best guy as any other means. For BPA to mean anything, the first thing is, there has to be a significant difference, not a decimal point, because every scout involved for every team has seen much better guys selected five picks after their BPA, every year.
The Leafs selected Dermott ahead of Aho. Were they drafting for position or BPA? They took Valiev ahead of Brayden Point. What was their thinking there because the chances of being an impact player always favored Point who lacked NHL size or he would have been a 1st round pick?
Skill level and physical package are both considered in every selection. The trick is not to miss the necessary balance, plus to have the luck that your prediction on the future development is accurate. If Karlsson stays 5'10" 170, he would have had a much different career. Alone among the major sports, the NHL has to consider growth between 17 and 20 as a huge factor in most drafting.
BPA is different from club to club and from scout to scout within an organization. Not an exact science.