I assume I'm one of our "NHLe obsessed posters".
1. I've literally never argued for anything more than incorporating production into our scouting, as in scout players and use it to pick/avoid guys with seriously high/low production. The same way you wouldn't draft a guy whos an awful skater or has low hockey IQ high in the draft you avoid guys who haven't produced, especially in the middle rounds. I've also completely ignored production in some player evaluations in the past. For example I really liked Svechkov last year even though most models weren't high on him and I wasn't as high as most on Chaz Lucius, despite his very good production.
Additionally, I think I'm the only one who really had anything to say on our 5th round pick Cole Jordan (who absolutely is not an NHLe pick) but was loved by a few local WHL scouts who's opinions I trust.
TLDR: It's a tool, it should be incorporated into player evaluation. That doesn't mean I think it's wrong to take the 5th ranked player over the 2nd ranked player when your scouts agree the 5th ranked player has higher upside/is more likely to translate to the NHL.
2. Lundell is not an example of this. I'm pretty sure Bader's model had Lundell higher than where he was drafted, Lundell had an elite D-1 season and his draft season was better than Rantanen's. I specifically didn't like him because watching him in his draft year reminded me of Sean Monahan. I felt he was a poor skater, who developed early and capitalized on almost all of his chances despite not generating very many. Over the summer of his draft year he massively improved his skating and the talent is a lot more obvious. I just didn't like him in the top 10 at the time.
3. Owen Power was ranked 6th, that's clearly a little low but there were many scouts (not numbers guys) who had Power at 6 or even lower last year. It's not as if the numbers guys hated him and all the scouts raved about his game. He, like most players in the 2021 draft had/has question marks. Here's a scout (for elite prospects and Dobber) I follow even questioning how translatable his offensive game was.
4. McTavish is a bit of a weird one. His ranking in the model was largely based off of his D-1 season and 13 games in a Swiss league. I don't think that's fair to him or the model. His single biggest draft stock booster was the U18s which Bader's model doesn't incorporate. He also seems to have developed a lot over his draft season. Tony Ferrari (former head scout of Dobber prospects, now works for The Hockey News said in regards to McTavish: "I was down on him early because his early season games with Olten were *fine*. I circled back and watched a few of his other season games and there was something that clicked. He was really good as the season wore on. Just keeps impressing."
Many of the scouts who I follow on Twitter and who's opinions I trust had him in their top 3/5. Cam Robinson, mentioned earlier, was one of the biggest proponents of him and had him at 2. I also don't think he was disliked on HF. I think the U18s got a lot more viewing than they usually do and McTavish dominated at them. I saw almost entirely positive things following the U18s. Even then it seems clear to me that he's still making big leaps in his development, he looks far more fluid than he did even at the U18s. The Ducks have been drafting so well in the last ~3 years. The only bad thing I ever said about him IIRC is that he wasn't going to be there at 13.
TLDR: Nhle is a tool that should be incorporated and just because I use it doesn't mean it's the only thing I focus on. Just because a guy is ranked 6th by Bader doesn't mean I'm comfortable picking him at 6, might just mean he's a good chance to take in round 3. People don't have to use one or the other and I'm not Georgeleafer.