How to fix the seeding process

[Stephen Turner]

This is the best thing I have read, and sums up the situation perfectly.  

[Steve M]

I'm glad you feel that way too.  I guess I misunderstood your position since I thought your article only talked about giving flexibility on assigning the locations of the #1 seeds.  I didn't know you also were in favor of allowing for swaps within bands to improve attendance.  Sorry to put incorrect words in your mouth.  Good job on the selection show.

[Steve M]

Why thank you. :-)   Are you the same Steve Turner I knew from the Class of '86 that was good friends with Paul H. (Mr. winning team, losing team 85-86)?

[Steve M]

Point well taken, and I know I use parentheses way too much. I'm an engineer after all, and as hard as I try, the ability to write direct, clear prose isn't one of my strengths.  At least I get the spelling right most of the time.  ;-)

[KeithK]

I wouldn't want to see attendance trump everything though.  As one of the articles on USCHO right now reminds us (Jack Parker's comments) the tourney three years ago where everyone stayed close to home, due to travel fears after 9-11, just wasn't a very interesting tournament.  I want to see as many East-West matchups as is reasonable and that means shipping some teams (presumably the lower seeds)  away from home.

BTW - I see no hypocrisy with my seeding reaction this year.  I love the east-west matchups in the West Regional. I just hate the fact that we have to go west. :-)

[KenP]

My only gripe about the seeding process is that the WCHA got 3 or the 4 top seeds.  No matter how dominant any one conference is, this is not good for the sport.  I'd like to see a cap.  One conference can have at most 2 teams as #1 seeds in the tourney.

This year I feel both Cornell and Michigan are more deserving of a #1 seed than Minnesota.  Even if Cornell was the #1 seed and still placed at Mariucci, I think this would be better.

[DeltaOne81]

Go ahead and fix the ranking system that gave Minnesota a #1 seed, but if 4 of the top 4 teams by an objective system are all from the same conference, then they all deserve a #1 bid.

[KeithK]

[Q]Go ahead and fix the ranking system that gave Minnesota a #1 seed, but if 4 of the top 4 teams by an objective system are all from the same conference, then they all deserve a #1 bid.[/q]I don't agree. Just because the system is objective doesn't mean that it's either the best objective system (RPI is completely objective but somewhat arbitrary) or the best thing for college hockey.  Now I do agree that the committee should stick with whatever criteria they establish and let the chips fall where they may (which they didn't in the Denver-CC case).  But I see no problem with modifying the criteria to achieve desirable outcomes, particularly in the case of seeding.

[DeltaOne81]

I don't see any problem with it, I just disagree with it.

I agree that no system can be perfect, but I think we should pick the best one that we can collectively find and go with it. 3 or 4 teams from the same conference being in the top 4 in the nation is entirely possible, and I don't see a reason why we'd need to fight that. Pick the best ranking system you can find and go with it.

Sure, its not going to be perfect, but its not any more or less flawed than saying that more than 2 teams from the same conference can't be #1s. If its more acceptable to have a flawed system that promotes some sense of fairness/spread the wealth then a flawed one that goes exactly by the numbers, well, that's a judgement call.

Say Keith, isn't this backwards from our occassional other run ins where I argue for spreading the wealth and you argue for letting the system (whatever system that may be) run itself? :-)

[KeithK]

I don't claim to be consistent! :-D  But sports is not a place for pure capitalism in my mind, so I do draw the distinction.

[DeltaOne81]

I didn't mean to take you to task in general, I just thought it was funny :-D. My sense of fair play is involved here, but I'd be more likely to agree with a system that said that more than 5 or maybe 6 teams from a conference couldn't make the tournament at all.

The #1 seed seems less important than a shot to play in the tournament at all, and have a chance to prove yourself. That i do agree should be spread.

I don't think getting a #1 is vital enough - especially as this 'home rink' bull seems to be starting to be phased out - hopefully Engelstad next year will be the last one ever.

With the sport getting big enough to put western regionals in places like Green Bay, Denver, Grand Rapids - potentially XCel, Detroit. And the east obviously can have a very good rotation with Worcester, Manchester, Rochester, Providence - potentially Syracuse, Springfield (not Amherest), Hartford, New Haven, even MSG, Philly, Buffalo, Boston, Portland (Me), etc.

[KenP]

The best simile for my seeding thoughts likens it to the NFL post-season.  In the NFL, the leaders get an auto-bid to the post-season.  After that you look at SOS, etc.  If one division has a 15-1 team, a 14-2 team and a 13-3 team, the 14-2 and 13-3 teams are a wild-cards, even if they have the 2nd and 3rd best overall records.

Going back to the NCAAs, I don't have any problem putting the 10 best non-conference-winning teams on the ice.  And I don't have a problem saying that 2 of the "best" 4 teams come from a single conference.  But you can't convince me that a team that is in 3rd place in their conference deserves seeding preference over 4 of the 5 other conference leaders.  That's a by-product of algorithms comparing conferences based on the unbalanced and sparce sampling of nc games, and should be explicitly addressed by imposing a #1-seed cap.

[Al DeFlorio]

[Q]DeltaOne81 Wrote:

 I didn't mean to take you to task in general, I just thought it was funny . My sense of fair play is involved here, but I'd be more likely to agree with a system that said that more than 5 or maybe 6 teams from a conference couldn't make the tournament at all.

The #1 seed seems less important than a shot to play in the tournament at all, and have a chance to prove yourself. That i do agree should be spread.

I don't think getting a #1 is vital enough - especially as this 'home rink' bull seems to be starting to be phased out - hopefully Engelstad next year will be the last one ever.

And the east obviously can have a very good rotation with Worcester, Manchester, Rochester, Providence - potentially Syracuse, Springfield (not Amherest), Hartford, New Haven, even MSG, Philly, Buffalo, Boston, Portland (Me), etc.[/q]
I think the old Coliseum in New Haven has been condemned as unsafe.  Bridgeport's Arena at Harbor Yard seats about 10,000.

[DeltaOne81]

[Q]Al DeFlorio Wrote:
I think the old Coliseum in New Haven has been condemned as unsafe.  Bridgeport's Arena at Harbor Yard seats about 10,000.[/q]
I don't know if its condemned, but you're right, it is closed, and they're trying to schedule it for demolition although there's a lot of objection as that place has a great deal of memories for people (me included).

I was just naming cities and didn't think of that, but Bridgeport is definitely an option. The point is there's plenty of eastern places to go where you don't need to do home rinks (though Amherst is as safe as any because UMass isn't exactly a power house - Minn/UND is definitely worse). And out west we're slowly moving to the same situation (I'll add Milwaukee to my previous list and hell, anyone want to bid with Chicago?)

[Steve M]

[Q]adamw Wrote:

 [Q2]Steve M Wrote:
I know Adam and Jayson wouldn't like it, because it would be tougher for USCHO to predict the brackets, but as long as they're allowing swaps such as 1b with 1c, they won't be able to predict the brackets perfectly anymore anyway.[/Q]
Not true. I am on record many times stating that I wish the committee would afford themselves more flexibility for common sense, because the numbers aren't accurate enough to rely so heavily upon them (after selection). And anyone who has been to regionals with small crowds knows how important it is to maximize attendance, and I have no problem with the committee doing so, so long as that is applied as fairly as possible (i.e. putting Michigan and Wisconsin in GR is OK, because Wisconsin can bring people too).

[/q]

Based on your article today, I'm going to have to retract my apology and stand by what I said above.  There is no comparison between swapping teams within a seed band and swapping teams in and out of the tourney, or even between seed bands for that matter.  The way I read them, the published NCAA rules don't require rigid seriatim ranking based on PWR, they just require their system to pick to field and break it into 4 seed bands.  The only stated reason to rank them further is to place the four #1 seeds, in order, in the regionals closest to home, which they technically did on the strict PWR rankings with CC at #2.  If you're so strongly against the DU/CC flip (which I agree was unnecessary, yet completely reasonable since the TUC and COPs were so close) how would you be able to defend multiple swaps, or tossing the 1-8, 2-7, 3-6 , 4-5 concept completely, to boost attendance?