2012 NCAA Tournament: Division I Bracketology

[dave1381]

Re: 2012 NCAA Tournament: Division I Bracketology

Why is it that all of the peculiar and unusual swings benefit teams from the East and always at the expense of teams from the West? Why does BC seem to be glued in at number 4 no matter what happens? I understand the impact of BU and NU winning and the domino effect, but how does BC also lose and they are still at number 4. UMD and UND are better teams than BC, BU, Harvard, Dartmouth, NU, or Mercyhurst. But I can see the handwriting on the wall, there will only be 3 WCHA teams in the NCAA mix and the only way UMD gets in is if UND drops a couple games down the stretch. It really is unfortunate that the NCAA can't find a way to ensure that the best 8 teams get an opportunity to compete for the title. Maybe they will surprise us and put UMD in to save money on travel.

And I wouldn't be shocked if only Minnesota and Wisconsin make it.

The reason why these "swings" are benefiting eastern schools is the swings are typically caused by the revolving door at the bottom of the top 12 in the RPI. It's good for leagues to get teams in the top 12, because their teams have generally posted better records against these marginal top 12 teams than the top teams in their league (particularly for the WCHA). But the revolving door has been between ECAC and Hockey East teams for the most part. OSU and Bemidji are stuck down at 15th & 16th in the RPI now, and have not been part of the revolving door as of late.

 

[DC78-82]

Re: 2012 NCAA Tournament: Division I Bracketology

UMD and UND are better teams than BC, BU, Harvard, Dartmouth, NU, or Mercyhurst. But I can see the handwriting on the wall, there will only be 3 WCHA teams in the NCAA mix and the only way UMD gets in is if UND drops a couple games down the stretch. It really is unfortunate that the NCAA can't find a way to ensure that the best 8 teams get an opportunity to compete for the title. Maybe they will surprise us and put UMD in to save money on travel.

Every time I think that the Leagues in the East should bow out and retire gracefully home to start a sub par Championship in the East because we're clearly not good enough or worthy of a shot at beating your WCHA queens, I just look back to last year when BC smoked Minny to get to the Frozen Four, and BU beat Mercyhurst, and somehow Cornell ran the gauntlet and made it to the dance too.

You don't know who is the best team until they play the game, and frankly, while I agree that Wisconsin is the clear leader again this year, I'm growing quite weary of the pompous attitude of (just a few) westerners. Every team has vulnerabilities, and they have all season to work on eliminating them and positioning themselves to vie for the NCAA's. In a one game elimination, the better team is the one who exploits the opponents' weakness while containing their own mistakes.

There, I said it, and now maybe my tongue will feel better now that I've stopped biting it.


btw - BC did not lose this week. The game ended in a tie, and for purposes of eliminating a team from the finals, a shootout was employed. They lost a shootout to a ranked team.

 

[dave1381]

Re: 2012 NCAA Tournament: Division I Bracketology

Every time I think that the Leagues in the East should bow out and retire gracefully home to start a sub par Championship in the East because we're clearly not good enough or worthy of a shot at beating your WCHA queens, I just look back to last year when BC smoked Minny to get to the Frozen Four, and BU beat Mercyhurst, and somehow Cornell ran the gauntlet and made it to the dance too.

You don't know who is the best team until they play the game, and frankly, while I agree that Wisconsin is the clear leader again this year, I'm growing quite weary of the pompous attitude of (just a few) westerners. Every team has vulnerabilities, and they have all season to work on eliminating them and positioning themselves to vie for the NCAA's. In a one game elimination, the better team is the one who exploits the opponents' weakness while containing their own mistakes.

I agree that I've been annoyed in the past when WCHA supporters to make all kinds of unjustified claims about their conference's greatness based on the national championship record. But in the context of the current state of the sport, WCHA supporters have reason to be upset. They are being screwed by the current system.

You can say "you don't know who is the best team until they play game" but the problem is some WCHA teams are going to not get the chance to play that game, or they might not get home ice when they deserve to be one of the top 4 teams.

 

[DC78-82]

Re: 2012 NCAA Tournament: Division I Bracketology

I agree that I've been annoyed in the past when WCHA supporters to make all kinds of unjustified claims about their conference's greatness based on the national championship record. But in the context of the current state of the sport, WCHA supporters have reason to be upset. They are being screwed by the current system.

You can say "you don't know who is the best team until they play game" but the problem is some WCHA teams are going to not get the chance to play that game, or they might not get home ice when they deserve to be one of the top 4 teams.

Every year the selection discussion includes a team or teams that are being shafted by the current system in some way. I have nothing at all against UMD, and in fact the utmost respect for a program that always finds a way to be a contender, but to say outright that BC is not as good as UMD, when head to head this year they split. Sure, that was a long time ago, but at the end of the day, BC has managed to find a way to stay at the top of HE, and with a lot of competition. Northeastern has won the lion's share of games they've played against teams both non con and in HE, and against teams that are in contention for the top 8, and despite the slump in December, BU is healthy and really playing some great hockey. They toyed with Harvard on Tuesday, and the game only looked a little close when the top 6 forwards were being rested and they played the majority of the 3rd shorthanded. They split with NoDak, have wins against other teams who were ranked at the time.

I just have a problem with the blatant statement that they are all clearly inferior. The system may be flawed, but everyone has the potential to be on the outside looking in because of it, or unjustly sent to the visitor's slots, not just the teams from the west.

 

[yungblud]

Re: 2012 NCAA Tournament: Division I Bracketology

By Team v. HEA v. ECACH v. CHA v. Ind. Total
Bemidji State 3-0-1 0-0-0 2-0-0 1-0-0 6-0-1
Minnesota 2-0-0 3-0-0 1-0-0 0-0-0 6-0-0
Minnesota Duluth 3-1-0 0-0-0 0-0-0 0-0-0 3-1-0
Minnesota State 0-0-0 0-0-0 2-2-0 2-0-0 4-2-0
North Dakota 3-1-0 0-0-0 0-0-0 1-0-0 4-1-0
Ohio State 0-0-0 0-0-2 3-0-1 0-0-0 3-0-3
St. Cloud State 0-0-0 1-0-0 0-3-2 0-0-0 1-3-2
Wisconsin 2-0-0 2-0-0 0-0-0 2-0-0 6-0-0
Totals: 13-2-1 6-0-2 8-5-3 6-0-0 33-7-6
(.844) (.875) (.594) (1.000) (.783)

I would say those numbers show some proof towards the WCHA being a little bit better. Just thought I would add a little gasoline to the fire.

 

[brookyone]

Re: 2012 NCAA Tournament: Division I Bracketology

Curious, how many...anyone think this is a better mouse trap besides me? Would fans be more amenable to a tournament dictated by the Rutter rankings?

http://board.uscho.com/showthread.p...ter-Rankings&p=5337315&viewfull=1#post5337315

And yes, I think I can accurately predict a couple likely opinions. However, not to include Ice.

 

[ManBehindTheCurtain]

Re: 2012 NCAA Tournament: Division I Bracketology

At the beginning of the season everyone, EVERYONE knows what the end of season selection criteria are going to be. Everyone is in control of their own destiny. I used to say that if you want to be undefeated you have to win all of your games. The same applies to the tournament, if you want to play in the tournament you have to win the games that will assure that you have met the criteria for playing in the tournament.

Every one of the eight teams that eventually gets selected will be a DESERVING entrant. Even so, it goes without saying that those teams that narrowly miss will be disappointed and will point to would have, could have, should have.

Coulda, woulda, shoulda. Didn't.

 

[ARM]

Re: 2012 NCAA Tournament: Division I Bracketology

The system may be flawed, but everyone has the potential to be on the outside looking in because of it, or unjustly sent to the visitor's slots, not just the teams from the west.

That's a key point. I'm not against the system because WCHA teams are being hurt by it. I don't consider Mercyhurst to be a west team, and they surely aren't being hurt by the current system. I want the teams that have played the best hockey to be the ones that are chosen and seeded the highest. The current system has too many places where it can't differentiate between the top team in the country and the tenth best.

BC earned the right to host last season by winning its conference championship, while Minnesota did not. After the semis, Minnesota was sitting fourth. BC jumped ahead of UM on the strength of beating 16-12-8 Northeastern, while Minnesota lost in OT to 33-2-2 Wisconsin. My frustration isn't that Minnesota got the short end of the stick last year -- they had the chance to compensate for the system's flaws and failed. What bothers me is that this system is often ill-equipped to differentiate between playing a team that is four games over .500 and one that is 31 games over.

I don't want to see anyone sent somewhere unjustly, or live with a flawed system. Every system will have flaws, and no justice will be complete. However, I have trouble with any system that penalizes a team more harshly for losing to Wisconsin than it does for losing to Maine or Quinnipiac. While I'm not the world's biggest UMD fan, they are in PWR trouble mostly because they went 1-7 against the top two teams in the rankings. That's why models that are designed with more thought have the Bulldogs considerably higher than the PWR does.

I'd like to see the system improved. And ideally, it would be nice if fans from the East don't wait until the East is much stronger than the West and some deserving team from Hockey East is left out before they agree.

 

[dave1381]

Re: 2012 NCAA Tournament: Division I Bracketology

Every year the selection discussion includes a team or teams that are being shafted by the current system in some way. I have nothing at all against UMD, and in fact the utmost respect for a program that always finds a way to be a contender, but to say outright that BC is not as good as UMD, when head to head this year they split. Sure, that was a long time ago, but at the end of the day, BC has managed to find a way to stay at the top of HE, and with a lot of competition. Northeastern has won the lion's share of games they've played against teams both non con and in HE, and against teams that are in contention for the top 8, and despite the slump in December, BU is healthy and really playing some great hockey. They toyed with Harvard on Tuesday, and the game only looked a little close when the top 6 forwards were being rested and they played the majority of the 3rd shorthanded. They split with NoDak, have wins against other teams who were ranked at the time.

I just have a problem with the blatant statement that they are all clearly inferior. The system may be flawed, but everyone has the potential to be on the outside looking in because of it, or unjustly sent to the visitor's slots, not just the teams from the west.

Ok, I agree that asserting that UMD is light years ahead of BC (or even BU) is highly obnoxious.

I don't agree with the "every year is the same, someone always gets shafted" point of view though. There's a huge difference in the way the 3rd-6th place WCHA teams are being shafted now.

In past years where we've had controversy, it's been more about interpreting the criteria or deciding what the criteria should be or how to weight the criteria. Like with the Clarkson-Dartmouth decision in 2008, Dartmouth was ahead in some criteria, Clarkson was ahead in others. The decision was open to interpretation.

But both this year and year, we have had selection criteria that simply do not do what they say they're supposed to be doing. They're supposed to be using a ratings percentage index that adjusts appropriately for strength of schedule. They're supposed to value success against other top teams. They are not. With the balance of power tilted to the WCHA, the WCHA teams are at a significant disadvantage for as long as this imbalance exists. It's not like this Clarkson-Dartmouth decision, where each team is equally likely to miss out in any given season. It simply isn't true that eastern teams are equally likely to be at a disadvantage. The goal is to take the best teams when you do have better methods to determine the best teams based on the criteria you've used for the last decade. I hope people can understand the difference.

 

[dave1381]

Re: 2012 NCAA Tournament: Division I Bracketology

At the beginning of the season everyone, EVERYONE knows what the end of season selection criteria are going to be. Everyone is in control of their own destiny. I used to say that if you want to be undefeated you have to win all of your games. The same applies to the tournament, if you want to play in the tournament you have to win the games that will assure that you have met the criteria for playing in the tournament.
Every one of the eight teams that eventually gets selected will be a DESERVING entrant. Even so, it goes without saying that those teams that narrowly miss will be disappointed and will point to would have, could have, should have.
Coulda, woulda, shoulda. Didn't.

Well, yes and no. The committee has tweaked things every now and then in fairly unpredictable ways. No one really had a clue how the committee would solve the logjam between teams 3 through 6 or whatever it was last season. No one had a clue that the NCAA would never fly four teams under any circumstances last year, and sent UMD to Wisconsin and Dartmouth to Cornell.

You're right, that as a coach or player, you have to make the most of the hand you're dealt in any one season. But in the longer run, you want to change the system in a way that allows the best teams to be recognized and makes for the best tournament possible.

 

[dave1381]

Re: 2012 NCAA Tournament: Division I Bracketology

Curious, how many...anyone think this is a better mouse trap besides me? Would fans be more amenable to a tournament dictated by the Rutter rankings?
http://board.uscho.com/showthread.p...ter-Rankings&p=5337315&viewfull=1#post5337315
And yes, I think I can accurately predict a couple likely opinions. However, not to include Ice.

I certainly agree it would be an improvement to replace the RPI with KRACH or Rutter.

I prefer KRACH over Rutter because I think KRACH is simpler and easier to interpret (the probability of team with rating A beating team with rating B is A/(A+B)) and I've never noticed any significant difference in the results between the two rankings. Probably the main difference is that Rutter formally models ties while the KRACH treats a tie as half a win or loss. That matters if you care about predicting whether individual games will end in ties, but I don't think it matters much for the purposes of an NCAA selection tool.

I do think the NCAA is wise to include something akin to the "Results vs. RPI top 12" criteria because I do think a tournament should place a premium on teams that perform well against the top 12. So for example, BU is far behind UMD in KRACH/Rutter because BU has some bad Hockey East losses, but BU has done pretty well against teams like BC & UND and would probably be closer to UMD in a ranking that gave more weight to games against the better teams. But the problem with the "Results vs. RPI top 12" in the current PWR are (1) it uses the wrong top 12, (2) it doesn't adjust the record for strength of schedule, (3) the rankings are subject to wild swings as teams move in and out of the top 12.

I believe the best solution would be to compute a weighted KRACH where you give more weight to the games against teams ranked higher in the original KRACH. E.g. you count the games against Wisconsin 12 times, the games against Minnesota 11 times, all the way down the list. That way you avoid this steep drop when 1 team moves in and out of the top 12. Though I admit I don't know quite what the best solution is here.

 

[ARM]

Re: 2012 NCAA Tournament: Division I Bracketology

I believe the best solution would be to compute a weighted KRACH where you give more weight to the games against teams ranked higher in the original KRACH. E.g. you count the games against Wisconsin 12 times, the games against Minnesota 11 times, all the way down the list.

No! That retains the problem that WCHA teams already have when the best two teams are from that league and everybody has to play them at least four times apiece, plus often a fifth time in the WCHA tournament. Now you've made it worse by putting this huge multiplier on those games. No matter how dominant Cornell gets, at least any team that is in NCAA consideration will never have to play the Big Red more than three times. If you want to avoid the wild swings, then put a fractional multiplier on games against 13, 14, 15, 16, where those games get less and less important.

 

[OnMAA]

Re: 2012 NCAA Tournament: Division I Bracketology

At the beginning of the season everyone, EVERYONE knows what the end of season selection criteria are going to be. Everyone is in control of their own destiny. I used to say that if you want to be undefeated you have to win all of your games. The same applies to the tournament, if you want to play in the tournament you have to win the games that will assure that you have met the criteria for playing in the tournament.

Every one of the eight teams that eventually gets selected will be a DESERVING entrant. Even so, it goes without saying that those teams that narrowly miss will be disappointed and will point to would have, could have, should have.

Coulda, woulda, shoulda. Didn't.

Agree, and it is even simpler than that. With the exception of MC, ALL teams in the top 16 or so (maybe even top 20) have a chance to go on a run in the playoffs and win an autobid...SOOOOO...All those teams still have their destiny in their own hands. Keep winning in the playoffs and you will go to the dance, it is really as simple as that.

P.S. You Go DC. Finally someone sticking up for the East.

 

[dave1381]

Re: 2012 NCAA Tournament: Division I Bracketology

No! That retains the problem that WCHA teams already have when the best two teams are from that league and everybody has to play them at least four times apiece, plus often a fifth time in the WCHA tournament. Now you've made it worse by putting this huge multiplier on those games. No matter how dominant Cornell gets, at least any team that is in NCAA consideration will never have to play the Big Red more than three times. If you want to avoid the wild swings, then put a fractional multiplier on games against 13, 14, 15, 16, where those games get less and less important.

Ah, but you're thinking still in terms of the current system. Losing to Wisconsin and Minnesota doesn't really hurt your KRACH rating too much, provided those teams are winning 90% of their games.

For a concrete example, let's consider North Dakota. According to current KRACH, their expected win pct. is 15% against Wisconsin and 23% against Minnesota. They went 0-3-1 against Wisconsin (12.5%) and so far 1-1 against Minnesota (50%). Whether weighting these games more on the margin is going to improve their rating depends on whether their actual win pct. in these games exceeds the expected win pct. So since they performed close to expectations against Wisconsin, and clearly better than expected against Minnesota, North Dakota would be better off and we gave more weight to performance against Wisconsin and Minnesota.

In fact, the sum of your expected wins against your actual schedule has to equal your actual number of wins, so doing any kind of weighting really is going to simply measure whether your KRACH rating is more about performing well against the best teams or more a result of how you performed against the worst teams.

The weighting scheme you suggest could also work... maybe I would do weight 1 on teams 1-8, then descending weights on the bubble teams 9 through 12.

 

[ARM]

Re: 2012 NCAA Tournament: Division I Bracketology

With the exception of MC, ALL teams in the top 16 or so (maybe even top 20) have a chance to go on a run in the playoffs and win an autobid...SOOOOO...All those teams still have their destiny in their own hands. Keep winning in the playoffs and you will go to the dance, it is really as simple as that.

Sure. And if the ref makes a couple of bad calls to put your team down 5-on-3, you don't have to let them score on the PP. It would just be nice if we're going to go to the trouble of having referees, that they make the correct call while they are there, and if we're going to have this elaborate process to pick teams, we pick those that have played the best hockey.

 

[dave1381]

Re: 2012 NCAA Tournament: Division I Bracketology

Agree, and it is even simpler than that. With the exception of MC, ALL teams in the top 16 or so (maybe even top 20) have a chance to go on a run in the playoffs and win an autobid...SOOOOO...All those teams still have their destiny in their own hands. Keep winning in the playoffs and you will go to the dance, it is really as simple as that.
P.S. You Go DC. Finally someone sticking up for the East.

Sure, but there's some excitement in being able to participate in an NCAA tournament and Frozen Four. It's not like everyone has the same shot at those opportunities.

If you're North Dakota now, a team that should be 4th or 5th and about on par with BC, and instead you're 7th and in serious danger of missing the tournament entirely -- AND you have to go through the top 2 teams in the country to get that autobid, which BC doesn't -- that means your postseason experience is likely not going to be as enjoyable as BC's, even though you're just as good as BC.

Now if the NCAA explicitly set out a mandate to encourage broader representation across conferences, then we could debate the merits of that. But what we have now is a system that claims to adjust for strength of schedule but does so incompetently, and claims it's taking the best remaining teams at-large when it is not. I want to see the East do well too, but not because the current system sucks.

EDIT: Of course the NCAA does have a broader mandate to encourage broader representation -- that's what the autobid is. But the current system makes it near impossible for any league to get more than 3 teams in, even if that league deserves them.

 

[ARM]

Re: 2012 NCAA Tournament: Division I Bracketology

For a concrete example, let's consider North Dakota. According to current KRACH, their expected win pct. is 15% against Wisconsin and 23% against Minnesota. They went 0-3-1 against Wisconsin (12.5%) and so far 1-1 against Minnesota (50%). Whether weighting these games more on the margin is going to improve their rating depends on whether their actual win pct. in these games exceeds the expected win pct. So since they performed close to expectations against Wisconsin, and clearly better than expected against Minnesota, North Dakota would be better off and we gave more weight to performance against Wisconsin and Minnesota.

Consider the concrete example of NoDak last year. They went 0-5 against UW, because mostly, nobody beat the Badgers. The 8th place teams are going to usually struggle against the most dominant team, and if you weight it too much at the top, you are rewarding the teams that mostly played teams around #10, which is the same flaw that the current system has with a different multiplier. Even though losing to Wisconsin doesn't hurt as much in KRACH, every ratings system will still punish losses. If everyone was playing the same teams, then no problem, but this system still punishes lower TUC teams that play a more difficult schedule. If parity against the top teams existed to the extent that it does in the men's game, then it would be less of an issue.

 

[northhockey]

Re: 2012 NCAA Tournament: Division I Bracketology

Consider the concrete example of NoDak last year. They went 0-5 against UW, because mostly, nobody beat the Badgers. The 8th place teams are going to usually struggle against the most dominant team, and if you weight it too much at the top, you are rewarding the teams that mostly played teams around #10, which is the same flaw that the current system has with a different multiplier. Even though losing to Wisconsin doesn't hurt as much in KRACH, every ratings system will still punish losses. If everyone was playing the same teams, then no problem, but this system still punishes lower TUC teams that play a more difficult schedule. If parity against the top teams existed to the extent that it does in the men's game, then it would be less of an issue.

As a NoDak fan, (last year) I thought we could have done better than a 7-1 loss in the NCAA tourney to the team that lost to the team that lost in the championship..

 

[ARM]

Re: 2012 NCAA Tournament: Division I Bracketology

As a NoDak fan, (last year) I thought we could have done better than a 7-1 loss in the NCAA tourney to the team that lost to the team that lost in the championship..

Me too, but I'm guessing that they'd have sent you to play Wisconsin instead. Although I'm not sure what the limit is for the NCAA where they have to start flying rather than busing teams. I know that the published distance used to be fairly small (300 miles?), but I think even Duluth was above that distance from Madison. Madison and Grand Forks are almost 600 miles apart, so that's a heck of a bus ride. If that is too far for the NCAA tournament, then no matter what else happens, if you get into the tournament, you're not in the top four, and Minnesota is, I'll bet that they send UND to Minneapolis.

 

[northhockey]

Re: 2012 NCAA Tournament: Division I Bracketology

If that is too far for the NCAA tournament, then no matter what else happens, if you get into the tournament, you're not in the top four, and Minnesota is, I'll bet that they send UND to Minneapolis.

..and I will be in the Cities, watching some great hockey!