If the only teams those four teams lost against were eachother. . . .

Posted by arssanguinus on 12/26/2011 3:31:00 PM (view original):
If the only teams those four teams lost against were eachother. . . .

Which isn't the case. NC Wesleyan is 3-1 against those four #1 seeds. 

get em girls

NC Wesleyan getting a 3 seed is a joke.  They should have been #1 overall and got #11.

And in rare form, none of the #1 seeds made it to the E8 in Knight. 

That's what happens when you give 7 and 8 loss teams most of the 1-seeds.

I was greatly shocked to see the S.CALI conf (as good as they were) get ALL four 1-seeds and two 2-seeds. I get they were attempting to reward the s.cali for such an above and beyond conf RPI, but when my team (NCW) got bumped to a 3!?!?!...I find that ludicrous

That being said I'm sure other well deserving teams got bumped down in a giant wave-effect

Posted by tianyi7886 on 12/26/2011 2:44:00 PM (view original):
Asher, going against a tough conf is one thing, but most of the teams in socal are .500-.600 against teams in rpi 1-50. You could say wow, that means socal is really strong and no one can dominate, but does it show that socal has the four best in teams in the country and deserve all #1 seeds? Hardly. If that logic applies, we might as well have all #1 seeds and #2 seeds in D1 belong to the ACC, and have 40-50 teams in the NT from the big 6 conf in D1. 

How is NC Wesleyan not a #1 seed with only 2 losses, and an 8-2 record against rpi 1-50? Sure, they didn't play 16-18 games against top opposition but they did go .800 in 10 games against top opposition. When all #1 seeds have this many losses, the system isn't working right. 

I feel like my clarity on what I mean has gone to zero the last month in the forums...

What I meant was that I don't know who deserves what seed, and since conferences that 'dominating' in the RPI, etc. at DIII are rare, it's hard to know if they really are going to wipe the floor in the postseason or flop on their faces (turns out we did the later).

I'm not sure where the balance should be struck between playing 18 games and having more wins against top 50 fits versus 10 games and 8 wins.  I was not arguing for or against what has happened, I was just saying that if the results were backed up in the tourney, then it wasn't an error.  As it sits, it's appears to be a flaw that needs to be adjusted.

I think it's not that difficult to strike a balance. Go with win % against top opposition once the teams have faced enough top rpi teams, set it at 8, 10, 12, or whatever seems logical. The current system is rewarding losses to top 50 rpi teams, which is plainly wrong. I have made a post about this issue in Tark seeding for D3, when a team with 4 more losses and worse record against top 50 rpi opponent (5-6 against rpi 1-50) takes the 2 seed while another team with more wins, higher rpi, and better % against top 50 rpi opponents (4-1) takes the #3 seed. 

Knight D3 further illustrates that the new system is rewarding teams that simply play good opponents, regardless of wins or losses. NC Wesleyan at 27 wins, CT championship, 8-2 against rpi 1-50 gets a 3 seed while 6 S. Cal teams with worse records, losing much earlier in CT, and much worse % against top 50 rpi gets the 1 and 2 seed? That's simply illogical. The NT results, which is your proposed metric to evaluate the seeding, further shows that teams are seeded incorrectly. I can't remember any NT in which none of the #1 seeds made it past the S16. 

good point.

Remember, too, that when a debate like this comes around, we naturally provide human logic and experience. In this case, we see what the real life selection committee would take into consideration. The HD formula, no matter how much it wishes to act like real life, can, and never will, process those human characteristics. We will always see anomolies like this one. Just part of the game.

Sure, but this system could easily be improved by not giving teams credit for losing games against good opponents, as it is doing right now. 

tianyi, I don't think that teams who play a bunch of good teams and lose all those games are being rewarded. Like in real life, teams are being rewarded for (a) actually winning some of those games and (b) playing a tough schedule.

Here's a perfect example of a team with a good SOS that played a solid # of NT teams, did poorly against them, and was left out ... when they clearly would've made it under the old system.

http://www.whatifsports.com/forums/Posts.aspx?TopicID=449098&TopicsTimeframe=30&TopicsPage=2

Posted by girt25 on 12/28/2011 9:14:00 PM (view original):
tianyi, I don't think that teams who play a bunch of good teams and lose all those games are being rewarded. Like in real life, teams are being rewarded for (a) actually winning some of those games and (b) playing a tough schedule.

Here's a perfect example of a team with a good SOS that played a solid # of NT teams, did poorly against them, and was left out ... when they clearly would've made it under the old system.

http://www.whatifsports.com/forums/Posts.aspx?TopicID=449098&TopicsTimeframe=30&TopicsPage=2

Check out this thread. Compare Drew and Hamline against Greensboro and Lynchberg. I would say that Greensboro is the weakest team among the 4, Hamline strongest by quite a bit. The system seeded these teams the opposite way. 

http://www.whatifsports.com/forums/Posts.aspx?TopicID=448462&ThreadID=9699112#l_9699112