The predraft contributions here are great. I think those of you who have given us the benefit of your experience have raised the bar. It’s discussed with some nuance and people will explain their thinking which is helpful. I still remember your descriptions of the 2018 draft well and I recall being a but bummed that NJ didn’t land Miller instead of Smith but obviously it worked out well in the end. For some reason I recall your evaluation of Dylan Samberg and it was spot on. You have a good eye for defenders.
@Guadana is so good with the European juniors and explaining why they will or won’t fit into the western game. He has an excellent feel for that. There are several others who pitched in this season with you busy. I read some of Steve Kournianos and he’s had some podcasts. I like that he’s not biased despite being an open Rangers fan he’s been very complimentary of NJ prospects the last few years. I can understand why a Bader annoys those of you who actually watch the games and video. It’s like those guys who said Petterson was going to hit in 2017 based solely on their statistical model. He did but my guess is that they missed other picks with the same model.
The highest ranking of Pettersson was actually Kournianos, who ranked Pettersson #3 overall despite not using any statistical metrics in his rankings.
In 2017, when Pettersson was drafted, the big stat guys who do draft rankings now -- Bader and JFresh primarily -- were not yet ranking draft eligibles. But with Bader and the 2017 draft, we need to keep in mind his loudest opinion was what a terrible pick Colorado made with Cale Makar, who was not even worth a 1st round pick based on his statistical model.
The most analytics friendly draft ranking I can think of from 2017 was Scouching, who uses more of a hybrid analytics/scouting model in his rankings. You are correct that Scouching also nailed it with Pettersson, as he was the only ranker I know who also had Pettersson at #3 overall.
Scouching also hit a home run in 2017 with a world-high ranking of Jason Robertson at #14. But the risk you find with guys who weigh analytics heavily into their draft rankings was also on display, as he had a ridiculously low ranking on both Makar (#9, despite he was in every other top 3-5 I can recall) and Rob Thomas (#28, despite Thomas being ranked routinely in top 20s and often in the early teens).
But again, Scouching -- because he also researches religiously and actually scouts the prospects -- is light years superior to a Bader or a JFresh.
Ultimately, you're correct in your assumption that analytics-based draft models sometimes hit at the cost of far-more-frequently missing by country miles. My advice with the analytic models would be to pay attention to the guys who are anomalously high (such as a Pettersson or Stankoven or Zellweger) while ignoring -- completely f**king avoiding -- any negative opinions on the players the consensus has very high but analytics models hate, such as a Seider or Stutzle or Makar).