NCAA bracketing affirmative action style -- loosen the rules?

Hockey seeding rules are so tight they hurt the Good of the Game. We're caught up in the absolutism of the four seeding bands (four 1-seeds, four 2-seeds, etcetera) when in fact the margin of error is probably one or two bands. Think how much Cornell's band and seed would flip if it played Michigan State at season's end when they're different teams, or if it was two wins, or two ties in the fall. Or if a Cornell team that was otherwise the same played Western Michigan not MSU in 2005.

What should the rules for seeding be:

1. Absolute rules:
- Host school stays home, meaning UMass-Amherst gets to play at UMass-Amherst if that's a regional site. Or Minnesota at Minnesota (oops).
- The four 1-seeds stay the 1-seeds, even though the difference between the fourth 1-seed and the top two 2-seeds isn't all that great.

2. There should be guidelines not rules:
- Avoid intra-conference matchups but balance that against what else happens if you force the four-entry WCHA into the four regions. Or four ECACHL teams should that happen.
- Higher seeds within bands should, not must, stay closer to home.
- In the East, "home" can mean either of two sites if it's a, say, 300-mile trip. Ditto for an Ohio State if Rochester or Detroit are regional sites.

3. Consider the good of the game guidelines:
- Add a weighting, maybe a tangent to power rankings, for a team's ability to promote the game through good attendance. Basically, this would keep a Cornell or Wisconsin in its home district most every year. Call it affirmative action, which is a still-shaky but decades-ago-established policy much beloved by colleges and a lot of people. We’ve got that already when an Alabama Huntsville makes the tournament. The repressed class being lifted up isn't any one college as much as all of college hockey. (Mostly we're trying to think of why we shouldn't be exiled to South Dakota when Cornell alone would half fill Pepsi Arena.)

4. Wildcard rules
- In lieu of the above, the seeding committee gets three wildcard choices. That is, it follows current rules, but it also has the power to make three exceptions in seeding, placement, and selection so long as the chairman goes on TV and explains them once at the outset of the tourney and once on Friday of Frozen Four weekend. Right, that No. 16 berth could be given to the No. 20 team if No. 20 had an awesome last month when its leading scorer came back. Make it fun to be on the committee once again.

But then there would be controversy. {insert sound effect here}

I'm pretty certain the whole reason for the strict pwr seedings is so the committee can say there wasn't any subjectivity.

I kind of like your ideas around where the teams should play. I'm not so big on the wildcard idea. In its current form, I think going strictly by pwr has something to be said for it. In your scenario, I'd tell #20, "Too bad. Play better next year." OTOH, I kind of like the idea of a 32 team tournament. If roundball can have 65 (what's up with that 64 vs 65 game, anyway?) I don't see any reason why hockey shouldn't trot out a tournament half as big. I bet the NCAA would make even more $$ that way, too.

Karl B. '77

PS Bill, we were at the Philly event but didn't bump into you. Next time.

heykb
OTOH, I kind of like the idea of a 32 team tournament..

I just don't think college hockey has the depth for a field that large. The RPI on USCHO only goes to 30, and that's including teams like 12-16-4 Notre Dame and Mercyhurst. A field of 32 would have over half of the D-1 teams in it. While a large tournament would be fun, a lot of the teams wouldn't deserve to be there.

heykb
But then there would be controversy. {insert sound effect here}

I'm pretty certain the whole reason for the strict pwr seedings is so the committee can say there wasn't any subjectivity.

I kind of like your ideas around where the teams should play. I'm not so big on the wildcard idea. In its current form, I think going strictly by pwr has something to be said for it. In your scenario, I'd tell #20, "Too bad. Play better next year."

When deciding the seedings, I would like to see the committee somehow factor in your record in the last 16-20 games. I think if you are playing well at the end of the season, it should matter. It shows that you have taken care of business in your conference and conference tourney. I believe basketball takes this into account in deciding which bubble teams will get in and also their seeding.

We know people in the WCHA will be against this as a "criteria" because "their league is so strong that any team can beat another on any given night" .

why not make it so that the host school cannot play at home??? isn't basketball that way?

Dpperk29
why not make it so that the host school cannot play at home??? isn't basketball that way?

Basketball is that way because they are assured of selling out every game at every regional no matter what teams are playing there.

Hockey is not because they are not.

Beeeej

Dpperk29
why not make it so that the host school cannot play at home??? isn't basketball that way?

Basically no. They just farm the teams out by seeding, with no more than a tweak so that #1's can play near their fan base.

heykb
If roundball can have 65 (what's up with that 64 vs 65 game, anyway?) I don't see any reason why hockey shouldn't trot out a tournament half as big.

There are 318 D-1 basketball programs. A 65 team tournament represents 20% of the country. There are 58 D-1 hockey teams. A 12 team field is 20%, 16 is 27.5%, and 32 is 55%. There aren't even 32 teams with RPI greater than .500. I like the idea that you have to have to be good to get into the tournament. 16 is a litte too big because it takes over 1/4 of the country, but it's ideal because there are no byes anymore. If there was a 32 team tournament this year based on RPI and the season ended today, Union would be in. I'm not even sure it would make more money. There a likely to be quite a few stinkers in the first round. They'd probably have to be campus sites with four regionals the next week. It's like a play-in game for everyone. Yuck.

Here's your 32 team field:
1 Minnesota
2 Wisconsin
3 Miami
4 Boston University
5 Michigan State
6 Nebraska-Omaha
7 Colorado College
8 Michigan
9 Maine
10 North Dakota
11 Cornell
12 Boston College
13 Denver
14 St. Lawrence
15 Harvard
16 Ohio State
17 Providence
18 Holy Cross
19 Alaska-Fairbanks
20 New Hampshire
21 Dartmouth
22 St. Cloud State
23 Ferris State
24 Vermont
25 Lake Superior
26 Northern Michigan
27 Colgate
28 Clarkson
29 Notre Dame
30 Mercyhurst
31 Sacred Heart
32 Union

Rita

When deciding the seedings, I would like to see the committee somehow factor in your record in the last 16-20 games. I think if you are playing well at the end of the season, it should matter. It shows that you have taken care of business in your conference and conference tourney.

I think that was the original logic of using the league tournament winner as the auto-entry vs the RS champ. Yes, it's each league's choice, but I believe that's the fundamental logic.

If more weight should be given to the last 16-20 games, you'd think savvy schools would start to tweak their schedules to give themselves a powder-puff non-conf game or two towards the end of the year. Everyone should start holding a Beanpot-like Feb tourney, hosting the Sisters of Mercy and the State School for the Hopeless. Then their "record in the last 20" is that much better.

Chris '03
...It's like a play-in game for everyone. Yuck.

You wouldn't want to see Colgate vs UNO? or Clarkson vs Michigan St? OK, there would be a few stinkers, but I see a lot of teams that are good enough to make the higher seed grumble about what a tough game they're playing in the first round.

I think it would be a hoot. Even if we wind up playing a UNH or LSSU in the first round. Not exactly a creampuff. There'd be a lot of pretty interesting matchups.

Rita
When deciding the seedings, I would like to see the committee somehow factor in your record in the last 16-20 games. I think if you are playing well at the end of the season, it should matter. It shows that you have taken care of business in your conference and conference tourney. I believe basketball takes this into account in deciding which bubble teams will get in and also their seeding.

We know people in the WCHA will be against this as a "criteria" because "their league is so strong that any team can beat another on any given night" .

I remember hating this feature more than any other back when it was included in the PWR (sorry, Rita) . It doesn't reward conference tournament performance too significantly (the first round loss in the Big East tourney is worth just as much as losing in the conference tourney championship), and specifically in hockey the different formats for conference tournaments would throw another variable into how many of last 16 truly "matter." As for the last handful of games played before the conference tournament, the only reason they mattered more than games at the beginning of the season was because the PWR specifically weighted them more with the "Last 16" factor (and I guess because they're more likely to be conference games). I would hate to see a PWR-mandated incentive for scheduling perennially poor teams at the end of the season. Also, the results of a consolation game should never ever matter.

The only reason for including this as a factor that I find compelling is that it will likely produce a better NCAA tourney field (as it will improve the chances of getting a "hot" team in). For me this isn't enough to further discount early-season games.

And for the record, I don't remember WCHA fans whining about this category more than any other fans, but maybe some older elynah denizen can remark on attitude shifts during the "Last 20" and "Last 16" days.

It wasn't too long ago that there was a "last-N Games" factor in PWR. It was discarded because it including almost entirely in-league games and thus favored weaker conferences.

heykb

Chris '03
...It's like a play-in game for everyone. Yuck.

You wouldn't want to see Colgate vs UNO? or Clarkson vs Michigan St? OK, there would be a few stinkers, but I see a lot of teams that are good enough to make the higher seed grumble about what a tough game they're playing in the first round.

I think it would be a hoot. Even if we wind up playing a UNH or LSSU in the first round. Not exactly a creampuff. There'd be a lot of pretty interesting matchups.

I'd love to see those matchups, but I think that's what the regular season is for. The purpose of the tournament should be to determine who the NCAA champion is. As great as March Madness is, it has really devalued the regular season and I would hate to see that in college hockey.

heykb
But then there would be controversy. {insert sound effect here}
If roundball can have 65 (what's up with that 64 vs 65 game, anyway?) I don't see any reason why hockey shouldn't trot out a tournament half as big. I bet the NCAA would make even more $$ that way, too.

Because hockey has less than 1/4 of the number of teams as squeakball, so having a tournament 1/4 is the size is actually already erring up. Having half of all D-I hockey teams in the tournament would be crazy.

I mean really, does anyone think that Vermont, Clarkson, and Notre Dame are worthy at an at large big based on their season as a whole? Please.


Now one by one:

[Q]1. Absolute rules:
- Host school stays home, meaning UMass-Amherst gets to play at UMass-Amherst if that's a regional site. Or Minnesota at Minnesota (oops).
- The four 1-seeds stay the 1-seeds, even though the difference between the fourth 1-seed and the top two 2-seeds isn't all that great.[/Q]

Already done.

[Q]2. There should be guidelines not rules:
- Avoid intra-conference matchups but balance that against what else happens if you force the four-entry WCHA into the four regions. Or four ECACHL teams should that happen.
- Higher seeds within bands should, not must, stay closer to home.
- In the East, "home" can mean either of two sites if it's a, say, 300-mile trip. Ditto for an Ohio State if Rochester or Detroit are regional sites.[/Q]

A matter of priorities only. First is basically already done (avoid interconfence is done unless it can't be), as is the last (unless we're talking, say, BC going to Worcester instead of Albany - there's a difference there).

The middle is an option, but it wouldn't stop screwing over, it would only change who gets screwed. That's because it would eliminate competitive balance (or whatever they call it).

If you were #1 (Minn, Wisc, etc), would you rather play #8 whoever (Cornell right now), or say, #5 because #5 is closer. Oh great, so I'm the #1 team in the country and my reward is to play a better team with more fans there, instead of a worse team with less who will make the trip. You can say being close to home is more important (and we would certainly think so after last year, and potentially this year), but the fact is that several teams will always get screwed. That being the case, competitive balance seems to me to at least be the most fair way to screw people.



[Q]3. Consider the good of the game guidelines:
- Add a weighting, maybe a tangent to power rankings, for a team's ability to promote the game through good attendance.[/Q]

Attendence concerns are already in there - but have taken a distinct back seat to about everything else by interpretation. You could give them higher weight, but again, you're just screwing different people. Sure, this one would help us, but you can't objectively say that that makes it better. Its just bias.


[Q]4. Wildcard rules
- In lieu of the above, the seeding committee gets three wildcard choices. That is, it follows current rules, but it also has the power to make three exceptions in seeding, placement, and selection so long as the chairman goes on TV and explains them once at the outset of the tourney and once on Friday of Frozen Four weekend.[/Q]

Awful, awful idea. Sure, instead of staying home because you didn't play well enough, you get to stay home because the committee member's brother's friend's cousin's husband is the coach of UMinn-Backwater, and ya know, they did pretty well down the stretch.

How will we feel about this one the first time #13 Cornell gets subjectively passed over for a WCHA club?


[Q]When deciding the seedings, I would like to see the committee somehow factor in your record in the last 16-20 games.[/Q]

They did this up until a few years ago (PWR category called L16 ("last 16" ). It was eliminated for 2 reasons. First, it didn't account for strength of schedule. Should Holy Cross really win a comparison because they went 13-3 down the stretch in Atlantic hockey over Wisconsin going 12-4 in the WCHA? Fix that and you're onto something.

Second, it meant that all games were not equal. Games at the end of the season mean more than games at the beginning. You can argue that makes sense. You can also argue its not just and all games should be equal. The NCAA sided with the latter.

Pete Godenschwager

heykb

Chris '03
...It's like a play-in game for everyone. Yuck.

You wouldn't want to see Colgate vs UNO? or Clarkson vs Michigan St? OK, there would be a few stinkers, but I see a lot of teams that are good enough to make the higher seed grumble about what a tough game they're playing in the first round.

I think it would be a hoot. Even if we wind up playing a UNH or LSSU in the first round. Not exactly a creampuff. There'd be a lot of pretty interesting matchups.

I'd love to see those matchups, but I think that's what the regular season is for. The purpose of the tournament should be to determine who the NCAA champion is. As great as March Madness is, it has really devalued the regular season and I would hate to see that in college hockey.

I appreciate your sentiment. But as long as the definition of national champion is "won 4 games in a row against opponents determined by the NCAA seeds" you could say the regular season is already devalued. We'd have to go back to the bad old days of 1-team-represents-a-conference to change that.

Last-n games overweighted means mostly league games and it favors the better team in a less competitive league. But for purposes of this discussion, we're talking improvements in fairness that are most advantageous to Cornell, not ways to help North Dakota or Quinnipiac. There should be a way to recognize a team that had, say, its two best players injured through mid-season or a freshman class that caught fire in February. OTOH, if it's a true national contender, they'd still be in there as a mid-pack seed, not out of the tournament. (Trivia: What is the lowest seed in a 16-team tournament to win the title? In basketball there are lots of first round upsets, but three of the final four are usually fairly high seeds. OK, there was Pittsburgh in pro football this year.)

Expanding beyond 16 teams would be a stretch as noted because 16 is already a greater percentage of D1 hockey than 64 basketball teams is a fraction of D1 basketball. Also, expanding the field increases the odds that the better team will have one off night, lose, and not make the title game. When Cornell won its national titles, the NCAA playoff was 4 teams (out of what, 30 that played seriously at the time?) vs. 16 now out of 60. Ned Harkness and Dick Bertrand and John Hughes did not have to worry about getting past Nebraska-Omaha or Miami or Ohio State on the tournament's first weekend.

Others have pointed out with finality: The NCAA tournament doesn't determine the best team in the country. It's tautological: The tournament defines the tournament winner.

We're grousing in part because we've had a run of bad luck in seedings and placements -- in 2002-03, didn't we get a tougher first round opponent than we should have in order to make the tournament allegedly fairer in some other way? But then, every year 15 hockey schools and 63 basketball schools are playing if-only in April ... we've been one of the if-only schools for 35 years running now. Maybe eLynah counts as grief therapy.

[Q]I appreciate your sentiment. But as long as the definition of national champion is "won 4 games in a row against opponents determined by the NCAA seeds" you could say the regular season is already devalued. We'd have to go back to the bad old days of 1-team-represents-a-conference to change that.[/Q]

I see your point, which is why I was clear to say NCAA champion and not say best team. So, we're past 1-team per conference, and we're short of a "anybody who played hard at any point in the season gets in" field. Sounds about right the way it is then. Getting into the 16 team field is competitive enough that the regular season matters, yet it's not so exclusive that a good team cannot recover from a poorly played stretch of games.

Picking the field should be done following hard and fast rules that are known by the community in advance. That way no one can ever complain that they're team was screwed out of a bid - you know what it's going to take and you just have to get it done on the ice. (You can complain about the crieria used, but that's different).

Seeding on the other hand is a totally different thing. The committee should have plenty of flexibility here. After all, everyone who gets in has to win 4 games to win it all including several against tough teams regardless of draw. Sticking to arbitrary rules like the "bands" or the 1-16 PWR ranking is silly. If you can make a better tournament by switching 12 and 13 then do it. Same for 11 and 13, or 9 and 12. There just isn't that much difference between the 9th and 12th ranked teams or between 11 and 13. Now they shouldn't switch things around without a good reason and should try to stick to the order where possible, but tying your hands because of the numbers is silly, IMNSHO.

Unless, of course, it meeans screwing the Gophers. Then I'm all for it.

[Q billhoward]Last-n games overweighted means mostly league games and it favors the better team in a less competitive league. But for purposes of this discussion, we're talking improvements in fairness that are most advantageous to Cornell, not ways to help North Dakota or Quinnipiac.[/Q]

Well, at least you're honest. Somehow I don't think saying "hey, NCAA committee, here are some ideas to help Cornell's seeding" is gonna go over that well.

There should be a way to recognize a team that had, say, its two best players injured through mid-season or a freshman class that caught fire in February.

Great, so now we have the ncaa basketball process, with subjective judgement, background deals, and smoke and mirrors. That's what the entire point of PWR is to avoid. If you want that, go be a basketball fan.

Last n games is a silly criterion to use. The season is the season and games played in November should mean exactly the same as games played in March. You should have to play well all season in order to earn the right to play for the national championship. There's no reason to reward the "hot" team - if a team is really "hot" they'll have a better chance to win the tournament.

billhoward
But for purposes of this discussion, we're talking improvements in fairness that are most advantageous to Cornell, not ways to help North Dakota or Quinnipiac. [/q]

Am I alone in thinking that a tournament that is fair is better than one that unfavorably advantages Cornell?

[q] (Trivia: What is the lowest seed in a 16-team tournament to win the title? [/q]

There's a sample of 3 and they were: 1-seed Minnesota, 2-seed Denver, 1-seed Denver. I don't think any of them were the overall #1.

[q]We're grousing in part because we've had a run of bad luck in seedings and placements -- in 2002-03, didn't we get a tougher first round opponent than we should have in order to make the tournament allegedly fairer in some other way?

You're grousing.
In '03 Cornell (the overall #1) drew #14 Minnesota-Mankato instead of the CHA or AHA entry because there were two WCHA #1 seeds and two WCHA #4 seeds meaning that the only way to avoid first round intraconference games was to match the two #1 seed WCHA teams with the AHA/CHA entries. Thus, instead of Cornell being rewarded with a #1 vs. #16 games, it was stuck playing a #14 that was leaps and bounds better than #15 or #16. It's a matter of how the committee's priorities are ordered. Where one might think that having 1/16 and 2/15 is fairer and more in keeping with bracket integrity, the committee feels that avoiding intraconference matchups is more important as long as the teams are kept in their bands.

Chris '03
You're grousing.
In '03 Cornell (the overall #1) drew #14 Minnesota-Mankato instead of the CHA or AHA entry because there were two WCHA #1 seeds and two WCHA #4 seeds meaning that the only way to avoid first round intraconference games was to match the two #1 seed WCHA teams with the AHA/CHA entries. Thus, instead of Cornell being rewarded with a #1 vs. #16 games, it was stuck playing a #14 that was leaps and bounds better than #15 or #16. It's a matter of how the committee's priorities are ordered. Where one might think that having 1/16 and 2/15 is fairer and more in keeping with bracket integrity, the committee feels that avoiding intraconference matchups is more important as long as the teams are kept in their bands.

Yes, but where the committee really screwed up that year was in ignoring individual comparisons and using straight PWR to place teams into the bands, which was unlike what was done in previous years. If they had focused on comparisons among teams in the tournament rather than irrelevant comparisons with non-tournament teams, OSU would have been a 4-seed instead of SCSU, and Minnesota and CC could have played OSU and one patsy, leaving the other one for Cornell. I'll let someone else look up the Hockey-L and/or eLF posts about it.

[q]Am I alone in thinking that a tournament that is fair is better than one that unfavorably advantages Cornell?
[/q]Nope. Fair is fair, even if it doesn't help Cornell.

jtwcornell91
Yes, but where the committee really screwed up that year was in ignoring individual comparisons and using straight PWR to place teams into the bands, which was unlike what was done in previous years. If they had focused on comparisons among teams in the tournament rather than irrelevant comparisons with non-tournament teams, OSU would have been a 4-seed instead of SCSU, and Minnesota and CC could have played OSU and one patsy, leaving the other one for Cornell. I'll let someone else look up the Hockey-L and/or eLF posts about it.

I have a better idea... how about you invite some of the committee members over, and we can all watch you get into a 10 minute debate about it in which you confuse the hell out of them

Ah, good memories...

Chris '03

billhoward
[q]We're grousing in part because we've had a run of bad luck in seedings and placements -- in 2002-03, didn't we get a tougher first round opponent than we should have in order to make the tournament allegedly fairer in some other way?

You're grousing.
In '03 Cornell (the overall #1) drew #14 Minnesota-Mankato instead of the CHA or AHA entry because there were two WCHA #1 seeds and two WCHA #4 seeds meaning that the only way to avoid first round intraconference games was to match the two #1 seed WCHA teams with the AHA/CHA entries. Thus, instead of Cornell being rewarded with a #1 vs. #16 games, it was stuck playing a #14 that was leaps and bounds better than #15 or #16. It's a matter of how the committee's priorities are ordered. Where one might think that having 1/16 and 2/15 is fairer and more in keeping with bracket integrity, the committee feels that avoiding intraconference matchups is more important as long as the teams are kept in their bands.

Do you recall the interview after the pairings were announced and Schafer, I believe, expressed his displeasure ... somewhat delicately? Perhaps he, too, recognized it's hard to feel sorry for an overall #1 arguing it deserves #16 not #13.

Overall, if one looks at Cornell's NCAA opponents and locations, would an impartial observer (maybe not us) say Cornell got fair, neutral, or disadvantaged draws? The 1-vs.-14 seed was ever so modestly disadvantageous but most people would call it neutral. Vs. Minnesota on Minnesota's home Olympic sheet was disadvantageous and the NCAA could have made it fairer by forcing the Gophers to use the Excel Center instead. We got lucky that the 2003 Frozen Four was in Buffalo ... and were unable to do anything with it.

To be national champion, you have to win four NCAA games. But you might not be able to beat every one of the 15 others. Maybe we could have beaten the team that beat Minnesota.

KeithK

Chris '03

billhoward
But for purposes of this discussion, we're talking improvements in fairness that are most advantageous to Cornell, not ways to help North Dakota or Quinnipiac.

Am I alone in thinking that a tournament that is fair is better than one that unfavorably advantages Cornell?

Nope. Fair is fair, even if it doesn't help Cornell.

That was meant to be a wry comment, just without the <g> after it. Fanatics could argue Cornell deserves a make-good for the alleged wrongs of previous years. No winner thinks the tournament was unfair.

Does anybody else think that the NCAA tournament should look a little bit more like the ECACHL tourney? I think that the initial two teams played (before the frozen four), should be played as a best of 3 games. I feel that this would help eliminate some spoilers and would guard against flukes. I know this doesn't specifically favor Cornell, which is why I brought it up. Personally, I've always hated the football playoffs for this reason and prefer the NHL and NBA method (Though, not best out of 7). It's such a shame to have a great season going and then in one game, lose to Mercyhurst and the like because your goalie ate something funny for lunch. Any thoughts?

-I do feel that the frozen four should continue as played, though, with a one game per opponent method (as per ECACHL again).

canuck89
Does anybody else think that the NCAA tournament should look a little bit more like the ECACHL tourney? I think that the initial two teams played (before the frozen four), should be played as a best of 3 games. I feel that this would help eliminate some spoilers and would guard against flukes. I know this doesn't specifically favor Cornell, which is why I brought it up. Personally, I've always hated the football playoffs for this reason and prefer the NHL and NBA method (Though, not best out of 7). It's such a shame to have a great season going and then in one game, lose to Mercyhurst and the like because your goalie ate something funny for lunch. Any thoughts?

-I do feel that the frozen four should continue as played, though, with a one game per opponent method (as per ECACHL again).

I have never, ever, ever understood that rationale. You want to avoid fluke outcomes - except where the outcomes are really important?!

The idea of the tournament is for whichever team is playing best at any given moment to move forward to the next level. The win-or-go-home format serves that quite well, and if your opponent beats you, you deserve to go home.

Plus, if you went to best-of-three at the regional level, you'd have to have two weeks' worth of regionals instead of one - and/or put them at campus sites. The main reason the early rounds of conference playoffs work well as best-of-three is that they're on campus.

Beeeej

The reason for not doing it for the frozen four is that it is assumed at that point that the abilities of the teams are similar. Sure, the "outcomes are really important," but a loss in one game to the #4 team is not so "flukey" as a loss to the #16 team.

I do understand the concern for a lengthy tournament and hotel stays, etc. I had thought about this as well, but i didn't know if some trade-offs were worth it.

In regards to always playing your best, there is a reason why the NHL, NBA, and MLB use a best of 7 game series for the playoffs. Their rationale seems to directly contradict yours as to how a tournament champion should be decided. I mean, why not just drop the puck and say first one to score wins (read with sarcasm)??? Also, if football were such a physically demanding sport, I'm sure they'd the multi-game method too. It is the most fair, and the only negative I see is the cost and inconvenience of longer stays.

canuck89
The reason for not doing it for the frozen four is that it is assumed at that point that the abilities of the teams are similar. Sure, the "outcomes are really important," but a loss in one game to the #4 team is not so "flukey" as a loss to the #16 team.

If you want to be reasonably sure the top four teams make the Frozen Four, go back to a four-team tournament.

In regards to always playing your best, there is a reason why the NHL, NBA, and MLB use a best of 7 game series for the playoffs. Their rationale seems to directly contradict yours as to how a tournament champion should be decided.

Their rationale contradicts nothing I said. Every round in the pro hockey, basketball, and baseball playoffs is multiple games. They have the luxury of scheduling such long playoffs to avoid flukes, build excitement, and add revenue, and so they do. If college hockey were to have the luxury of a three-game series for every level of the NCAA tournament, I'd think it was a little silly just in terms of duration, but I'd shrug and go along with it. It's saying you want to avoid flukes only in the earlier rounds that I find bizarre - especially when there is always the chance that a Mercyhurst will still get through a three-game quarterfinal to the Frozen Four. Suddenly, Minnesota's terrified of another fluke, and saying to themselves, "Why isn't this round three games, too??"

Beeeej

Lets be honest here. The primary reason that professional leagues go with long series is revenue. Yes, there are plenty of fans/commentators who argue that a longer series is a better test - e.g. those who say the first round of the baseball playoffs should be 7 games. But the reason the owners go to longer series and a larger number is revenue.

I'm not sure that football would necessarily go with a a series if it were possible. I can't imagine the Super Bowl being such a big event if it were a series of games.

Chris '03
Where one might think that having 1/16 and 2/15 is fairer and more in keeping with bracket integrity, the committee feels that avoiding intraconference matchups is more important as long as the teams are kept in their bands.

Everyone on the planet who follows college hockey understands this. Some of us just think the NCAA has their priorities backwards.

Why is it a problem if Mankato plays Colorado College in the first round? Why is it preferable to screw a #1 overall seed in order to avoid such a match-up?

On the one hand people are arguing above the desirability of maintaining the precise PWR rankings ("Why should #1 have to face #5 instead of #8?", but on the other seem perfectly willing to accept this silly restriction.

And I'm with Beeeej when he writes: "It's saying you want to avoid flukes only in the earlier rounds that I find bizarre - especially when there is always the chance that a Mercyhurst will still get through a three-game quarterfinal to the Frozen Four." It still makes no sense to me that the first two rounds of the ECACs are two-out-of-three when the three "money" games are lose-and-go-home.

Remember the old riddle, "Why does a dog lick himself? ... "

In every sport where it's possible to play back to back games (just about everything except football and Daytona (and NASCAR yellow flags are effectively second and third chances)), the pros play them as best-of-seven series for the same reason as the dog.

More games bring in more revenue and the players don't have to go to class Monday-Friday.

The best-of-seven series does provide some fairness for a team that has a bad day in the first game. It can recover. OTOH the number of series that go seven games shows how closely matched the two teams are, or how sports are random, too. (Trivia: Of all the world series and NBA and hockey playoffs ever played, how many went the full seven games?)

What college sports have a second chance? Rowing (repechage)? CWS baseball? Other?

billhoward
What college sports have a second chance? Rowing (repechage)? CWS baseball? Other?

Wrestlebacks in tournament wrestling. But you can't win or finish second in your weight class going through the wrestlebacks.

Al DeFlorio
Everyone on the planet who follows college hockey understands this. Some of us just think the NCAA has their priorities backwards.

Why is it a problem if Mankato plays Colorado College in the first round? Why is it preferable to screw a #1 overall seed in order to avoid such a match-up?

On the one hand people are arguing above the desirability of maintaining the precise PWR rankings ("Why should #1 have to face #5 instead of #8?", but on the other seem perfectly willing to accept this silly restriction.

I think avoiding inter-conference matchups, especially in the first round, is a very important factor. I don't like the idea of teams who just played in a conference tournament final or semi turning around and playing again six days later in round 1. But in my mind the problem is not that they have this rule. It's that they apply certain rules - the banding - very strictly when it doesn't make sense. Everyone who was paying attention in 2003 knew that the committee could have avoided both first round inter conference matchups and preserved equitable seeding by switching the #12 and #13 seeds, who were certainly much more interchangeable than #14 and #16. But their strict adherence to the bands prevented this solution.

Thankfully we won both games that weekend anyway.

KeithK
Everyone who was paying attention in 2003 knew that the committee could have avoided both first round inter conference matchups and preserved equitable seeding by switching the #12 and #13 seeds, who were certainly much more interchangeable than #14 and #16. But their strict adherence to the bands prevented this solution.

Good point, not previously made so well: The #16 seed is often not a top twenty team. It got an automatic qualifier berth. It should be the reward for the overall #1 team to feast upon. If you were the #1 overall and you had an ailing defenseman or an overworked No. 1 goalie, you could pretty safely rest him in game one of the first weekend.

Some year, when the overall #14 gets swapped for #16 in the name of some other tournament integrity rule and takes out the overall #1, while the #2 team shuts out #16, there'll be squawking.

billhoward
Some year, when the overall #14 gets swapped for #16 in the name of some other tournament integrity rule and takes out the overall #1, while the #2 team shuts out #16, there'll be squawking.

...unless the eliminated #1 is Cornell, in which case mostly there'll be a lot of laughing, and it'll be fuel for several years' worth of "EZAC" posts.

Which is another reason it's a really good thing we didn't lose that game.

Beeeej

In response to Beeeej, I wanted the best of 3 only in the first two rounds because of the reasons you've stated. A series is more expensive and takes its toll on the players. Therefore, I thought a compromise would be in order. The positive of a series is that the better team has a better chance of winning. Following that thought, a best of 3 is more necessary (to avoid flukes) in the first two rounds than the last two (F4). I know the rationale seems weird, but it was a compromise between convenience and fairness. I do understand that other college sports don't do this either, however, someone has always got to be first.

Now to be honest, I don't expect this to happen soon, or at all for that matter. Instead, I was just asking what people thought if in fact this system could be pulled off and made to work.

I find myself wondering if you're aware that it's actually how they used to do it, for a few years at least, back in the earliest days of the 12-team field. Cornell beat Michigan in the first game of the NCAAs in 1991, but it was a best-of-three series and they lost the other two games. The quarterfinals was also a best-of-three.

Beeeej

...and before the best of three series we had the infamous "total goals" two game set. *cringe* Nothing like a six period game over two days!

...or how about the "mini-game" in the ECAC tournament, 1983-1991?

billhoward
Remember the old riddle, "Why does a dog lick himself? ... "

Good God, Bill. Among other things, you're becoming the Dan Rather of this board as well.

heykb

Rita

When deciding the seedings, I would like to see the committee somehow factor in your record in the last 16-20 games. I think if you are playing well at the end of the season, it should matter. It shows that you have taken care of business in your conference and conference tourney.

I think that was the original logic of using the league tournament winner as the auto-entry vs the RS champ. Yes, it's each league's choice, but I believe that's the fundamental logic.

If more weight should be given to the last 16-20 games, you'd think savvy schools would start to tweak their schedules to give themselves a powder-puff non-conf game or two towards the end of the year. Everyone should start holding a Beanpot-like Feb tourney, hosting the Sisters of Mercy and the State School for the Hopeless. Then their "record in the last 20" is that much better.

I think everyone's regular season ends this weekend, going back 10 weeks and assuming 2 games per week that puts you at December 24/25th. Thus if you take the last 20 games, that will give you the holiday tourney and allows Minnesota to count their Holiday tourney game v. sacraficial lamb Union and Denver's loss against Princeton (who didn't get the sacraficial lamb memo). But I should not be so snarky, considering I was in Estero for our win against 3-21-6 Northeastern. I do not think many teams have any open dates from mid-January on to pad their schedules, so unless their was a major shift in scheduling I don't think this would be a problem.

Teams do not have any control over who their conference opponents are, unless we really want to be like football and basketball (ACC in particular) and bribe/ steal teams from other conferences to make your conference more "marketable". No reason for hockey to do that, each conference can and does have their "big" tournament game .

I'll still throw this out there even though practically everyone hates the idea of the L20/L16. Is there a mathematical way to that take into account some RPI factor of your league to compensate for the fact the top to bottom the Atlantic hockey is not as strong as the WCHA? I know the RPI does this but it brings in your opponents opponents etc, things that you can't control. They have the bonus factors (.3/.2/.1 or .33/.22/.11 don't the numbers vary from year-to-year?) for quality wins, how much more difficult would it be to add in a bonus based on winning percentage/conference strength?

Yes, if you are a really hot team at the end of the season you will win your conference tourney and take the automatic bid. But if you start the season 6-6-2 and finish 15-4-1 with the 4th loss being in the conference tourney finals you are now 21-10-3. Good enough for an at large bid? Maybe, maybe not (North Dakato got in at 22-14-5) depends on what conference you are in, who your opponents played and beat, and other math that baffles me every February/March .

OK, I'll go back to my corner now and won't bring up L16/L20 anymore because I don't have time to do the math to see who would have benefited or been screwed by such a factor.

I think only the ECAC and CCHA are done this weekend.

[q]I'll still throw this out there even though practically everyone hates the idea of the L20/L16. Is there a mathematical way to that take into account some RPI factor of your league to compensate for the fact the top to bottom the Atlantic hockey is not as strong as the WCHA?[/q]You certainly could. I believe John's "You Are the Committee" scripts offered some version of this back in the day or at least we talked about it. I still think it's not a relvant criterion.

Rita
Is there a mathematical way to that take into account some RPI factor of your league to compensate for the fact the top to bottom the Atlantic hockey is not as strong as the WCHA?

Yeah, its possible. But the NCAA seems to be very hesitant to make the math more complicated than it already is.

Rita
OK, I'll go back to my corner now and won't bring up L16/L20 anymore because I don't have time to do the math to see who would have benefited or been screwed by such a factor.

Guess what this lets you add? [www.slack.net]

In fact it even lets you add it KRACH weighted.

You can thank JTW later

DeltaOne81

Rita
Is there a mathematical way to that take into account some RPI factor of your league to compensate for the fact the top to bottom the Atlantic hockey is not as strong as the WCHA?

Yeah, its possible. But the NCAA seems to be very hesitant to make the math more complicated than it already is.

Rita
OK, I'll go back to my corner now and won't bring up L16/L20 anymore because I don't have time to do the math to see who would have benefited or been screwed by such a factor.

Guess what this lets you add? [www.slack.net]

In fact it even lets you add it KRACH weighted.

You can thank JTW later

Unfortunately, I don't have time to fiddle with John's cool programs (I need to put together a poster for a conference), but I'll buy John a beer if he is in Estero in December.

RichH
Good God, Bill. Among other things, you're becoming the Dan Rather of this board as well.

Better that than Bill O'Reilly.

As a matter of fact, I was not aware due to my age. Thank you, though, for letting me know!

1) Get rid of the auto bids. Return the emphasis of the conferences, and of college hockey in general, to the *conference* tournaments. The worst thing that has happened to college sports is ESPNization -- obsession with the national championship. Return the NCAA tourny to its status as a "cherry on top," after the real business of the sport, the determination of the champions among the natural and historical rivalries, is accomplished.

2) Get rid of any non-deterministic seeding criteria. The .003/.002/.001 nonsense just leads to charges of obfuscation in order to play to favorites. Pick one deterministic ranking system (KRACH; whatever). Disband the "selection committee." The day the last conference final is played, everyone will be able to generate the field.

3) Get rid of the regionals. As long as there are regional sites, there will be biased and revenue-driven exceptions and policies. "The only way to prevent corruption in high places is to get rid of the high places."

4) Allow intra-conference meetings in any round. Strict 1/16, 2/15, etc...

5) Re-seed after each round. No fixed brackets -- the best survivor of the first round meets the worst survivor in the QF.

6) Make the 1R and QF best-of-three at the home seed.

7) Split all NCAA gate and rights fees equally among all D-1 programs.

Trotsky
Return the NCAA tourny to its status as a "cherry on top,"

Just like in 1985. A noble goal.

Trotsky
7) Split all NCAA gate and rights fees equally among all D-1 programs.

Would that include the non-D-I schools playing D-I hockey whom, last I heard, did not partipate in sharing receipts? (Maybe it's only the D-III schools.)

Your nom de plume (nom de guerre?) is apt. You dreamer! Take the revenue out of venue? Make the NCAA title a nice endinig to the season? Split gate fees with the have-nots of the sport so they'd get better? Next you're going to want students in class Monday to at least Thursday? How's that going to play in Miami and Nebraska?

I do like the idea of reseeding going into the Frozen Four. (As the ECAC does.) That way you could tinker with some initial seedings and give the #1 seed (#1 survivor) an advantage going into the final weekend that it deserves if it's the #1 surviving ream. That's sort of an advantage to a team that did well during the RS.

OTOH, there's something in favor of the new status quo: We've grown accustomed to made for TV and extreme sports becoming real sports. And the televised championship showdowns that go with them. (American Idol is the same thing only different. Pure trash. Plus the cruelty of seeing the pain on someone else's face when they're told on national TV that they can't sing worth a lick. Something they should have been told a couple years earlier in private.) There are probably people on the forum who remember when ski slops banned snowboards. Now snowboarding is an Olypic sport and snowboard cross is not just a sport, it gives gold medals. And I think it's pretty cool to watch.

billhoward
OTOH, there's something in favor of the new status quo: We've grown accustomed to made for TV and extreme sports becoming real sports. And the televised championship showdowns that go with them. ... There are probably people on the forum who remember when ski slops banned snowboards. Now snowboarding is an Olypic sport and snowboard cross is not just a sport, it gives gold medals. And I think it's pretty cool to watch.

Snowboarding is a good example, though, since it is only in the Olympics because Americans only watch the Olympics if Americans are winning gold medals, and we suck in all the real sports, so we had to rig the system. (See also: allowing in the professionals and destroying the entire point of the Olympics).

I actually heard someone say, with a straight face, that because the TV ratings are low in the US for the Olympics that the Olympics are "in trouble." That's exactly ass-backwards. An event becomes important, *then* it attracts TV. If the event can't support the TV audience, fine, get rid of TV and let the event settle back to what it was before, don't change the event. The current philosophy is the equivalent of saying, "nobody wants to stay up until 3 am to know who wins the presidential election, so let's close the polls in California at noon."

TV is a fine and wonderful thing, but it is essentially parasitic, and like any parasite, if it doesn't exercise restraint, it kills off its host. The natural evolution of something governed by marketing and mass media is the evolution of TV news from 1960 to today. That's not a model *anything* else should adopt.

We now return you to the 18th century, in which I would have been much more happy.

Trotsky
We now return you to the 18th century, in which I would have been much more happy.

Except of course that in the 18th century, your laptop's keyboard would be lit by candlelight.

Beeeej

DeltaOne81

jtwcornell91
Yes, but where the committee really screwed up that year was in ignoring individual comparisons and using straight PWR to place teams into the bands, which was unlike what was done in previous years. If they had focused on comparisons among teams in the tournament rather than irrelevant comparisons with non-tournament teams, OSU would have been a 4-seed instead of SCSU, and Minnesota and CC could have played OSU and one patsy, leaving the other one for Cornell. I'll let someone else look up the Hockey-L and/or eLF posts about it.

I have a better idea... how about you invite some of the committee members over, and we can all watch you get into a 10 minute debate about it in which you confuse the hell out of them

Ah, good memories...

Yeah, if I'd realized I was going to take the podium, I would have composed my thoughts better.

My pleasure, and I thought you might not - because having known it, you might have assumed it was ditched for good reasons. At least that's what I assume. And the reasons can't have amounted only to "Now that it's 16 teams there's no time for three-game early rounds," because it was still 12 teams from 1992-2002 and the first round was only one game.

(Which sucked for its own reasons - not that Cornell could've necessarily beaten eventual champion North Dakota on the second night in 1997, but they'd at least have had a fair shot at it if they hadn't played Miami the night before while UND was into the honor bar.)

Beeeej

Beeeej

Trotsky
We now return you to the 18th century, in which I would have been much more happy.

Except of course that in the 18th century, your laptop's keyboard would be lit by candlelight.

Whale tallow. Don't exagerrate.

The primary problem of the 18th century (other than dentistry), is that Cornell didn't exist yet but Harvard did.

Bet we'd still have more fans, though.

jtwcornell91
Yeah, if I'd realized I was going to take the podium, I would have composed my thoughts better.

John, no matter how well you organized your thoughts, the panelists would have been confused. Them trying to rationalize a system of which they had no comprehension was the greatest "Emperor's New Clothes" moment I've ever witnessed.

Ah, another NCAA anachronism: I believe Final Four teams once played one semifinal game Thursday and one game Friday in advance of the Saturday title game.

billhoward
Ah, another NCAA anachronism: I believe Final Four teams once played one semifinal game Thursday and one game Friday in advance of the Saturday title game.

Cornell knows this. All. Too. Well.

OK, I'll go back to my corner now and won't bring up L16/L20 anymore because I don't have time to do the math to see who would have benefited or been screwed by such a factor.

Guess what this lets you add? [www.slack.net]

In fact it even lets you add it KRACH weighted.

You can thank JTW later

See also [slack.net]

Trotsky
Snowboarding is a good example, though, since it is only in the Olympics because Americans only watch the Olympics if Americans are winning gold medals, and we suck in all the real sports, so we had to rig the system. (See also: allowing in the professionals and destroying the entire point of the Olympics).[/Q]Give me a break. Snowboarding is a sport because more young people do that than ski. Sure we are good at it but it's all over the world. If we didn't allow it skiing may have died out as the youngers just left. Now at least we can go to the same slope.

[Q]We now return you to the 18th century, in which I would have been much more happy.

But maybe dead by now considering the state of medical care then.

Jim Hyla
But maybe dead by now considering the state of medical care then.

Cheap health insurance, though.

Jim Hyla

Trotsky
Snowboarding is a good example, though, since it is only in the Olympics because Americans only watch the Olympics if Americans are winning gold medals, and we suck in all the real sports, so we had to rig the system. (See also: allowing in the professionals and destroying the entire point of the Olympics).[/Q]Give me a break. Snowboarding is a sport because more young people do that than ski. Sure we are good at it but it's all over the world. If we didn't allow it skiing may have died out as the youngers just left. Now at least we can go to the same slope.

[Q]We now return you to the 18th century, in which I would have been much more happy.

But maybe dead by now considering the state of medical care then.

Whenever I question if the US is "rigging the system" I look at how many medals are awarded in cross country skiing and Nordic combined.

nyc94

Jim Hyla

Trotsky
Snowboarding is a good example, though, since it is only in the Olympics because Americans only watch the Olympics if Americans are winning gold medals, and we suck in all the real sports, so we had to rig the system. (See also: allowing in the professionals and destroying the entire point of the Olympics).[/Q]Give me a break. Snowboarding is a sport because more young people do that than ski. Sure we are good at it but it's all over the world. If we didn't allow it skiing may have died out as the youngers just left. Now at least we can go to the same slope.
[Q]We now return you to the 18th century, in which I would have been much more happy.

But maybe dead by now considering the state of medical care then.

Whenever I question if the US is "rigging the system" I look at how many medals are awarded in cross country skiing and Nordic combined.

U.S. hopes in biathlon were dashed with the unavailability of the vice-president.