NCAA Seeding

[El-Jefe]

Didn’t we get the same thing? You said 59 of 80 kids seeded in the top 8 made all American and I said 21 kids outside the top 8 made All American. My question is if the 75 percent accuracy means the seeding was good or if 75 percent of top 8 seeds made All American because they had a favorable draw?

I should refund my Project Literacy scholarship.

But in any case: there will always be some top 8 guys not make the podium. To me, having 76/80 top 12 seeds make the podium says they seeded very well to within 1 win.

The counter-argument would be that 9 guys (almost 1 per weight) failing to reach the blood round means that a few too many guys got over-seeded.

 

[IwasAwful]

I should refund my Project Literacy scholarship.

But in any case: there will always be some top 8 guys not make the podium. To me, having 76/80 top 12 seeds make the podium says they seeded very well to within 1 win.

The counter-argument would be that 9 guys (almost 1 per weight) failing to reach the blood round means that a few too many guys got over-seeded.

I would love to use Berge as an example of being seeded well increases your chance of making the podium, but he didn’t make it and the kid with basically the same seed as him did.

 

[El-Jefe]

I would love to use Berge as an example of being seeded well increases your chance of making the podium, but he didn’t make it and the kid with basically the same seed as him did.

We all agree Berge's seed stunk on ice.

But cherry-picking that one example doesn't prove anything about seeding quality at the tournament level.

BTW, I think the entire seeding process is awful, and have pointed out several ways in which it could be improved. That may or may not impact placement metrics. It's quite possible that a 100% random draw would produce good placement metrics, but of course that would be ridiculous in so many ways.