Even though you favored pure bracket integrity, you did the best job of identifying all of the "work within the system" options. Appreciate your work. You too, TTT.
USCHO Fan Forum
This is a sample guest message. Register a free account today to become a member! Once signed in, you'll be able to participate on this site by adding your own topics and posts, as well as connect with other members through your own private inbox!
-
The USCHO Fan Forum has migrated to a new plaform, xenForo. Most of the function of the forum should work in familiar ways. Please note that you can switch between light and dark modes by clicking on the gear icon in the upper right of the main menu bar. We are hoping that this new platform will prove to be faster and more reliable. Please feel free to explore its features.
2024 Pairwise Predictor
- Thread starter TonyTheTiger20
- Start date
Even though you favored pure bracket integrity, you did the best job of identifying all of the "work within the system" options. Appreciate your work. You too, TTT.
Quick further comment on the frustration:
Position yourself at the Ohio State vantage point. Put on a pair of WCHA glasses. Let me believe that Saturday 3/9 was an aberration, and that the Buckeyes return to regular season form.
Now glance at the pairings. It's immediately apparent that OSU could win the tournament having played ZERO national games.
None of the games are locks. In a couple of weeks we could easily be giving kudos to Colgate or Clarkson for winning it all. But the fact that the Zero National Games scenario is a realistic possibility? It does make OSU shoulders sag.
Thank goodness we played some of the ECAC Powers in non-conference play.
Timothy A
Let's go RED!
No it's not for what you say. It's to seed a tournament to determine a national champion. That is it.
ARM
Fan of chipmunk-like mascots.
Minnesota had such a bracket last year: UMD in the quarters; UW in the semis; and had UM been good enough, they'd have seen tOSU in the final. The year before, UMD knocked us out in a quarter designed to create more regional interest, or however it was phrased. I've wished for more emphasis on East/West pairings for years, but it isn't a goal of the committee, so here we are. OTOH, you might get lucky and Connecticut and Clarkson can provide you with a path to the final with nary a WCHA team in sight.
Thanks for adding this history. Yes, I remember those brackets. And the history certainly contributes to my opinion that National Games = Good; Intra-Conference Rehashes = Not So Much.
I'd only add that now that the 2024 brackets are done deal, I personally don't care who we play next. Publicly or Privately. Debating what the ideal bracket should look like is a fun and interesting pastime, right up to the Selection Show. But from this point forward, it's lace 'em up and play the hand you're dealt.
Hockeybuckeye
Well-known member
I think Sierra makes a valid point.
QuasiCosmos
Member
With four teams each from only two leagues (this year WCHA and ECAC) in the top eight, it was going to be unlikely that the top eight in the pairwise would result in no intra-conference first-game-for-any-team match-ups when following the pairwise. The only way to guarantee no intra-conference first game when the top eight are from only two leagues is to just declare before the season begins, "When only two leagues are represented in the top eight, make the top pairwise team from each league be seeded with the fourth from the other, and similarly for the second and third."
I guess we're lucky Minnesota had such a good season. It would have been fun to see what permutations people would have proposed if WCHA teams were 1,2,7,8! Would number 5 have to play number 2, and would number 5 still get the first round bye?
Let's all cheer for UConn to defeat UMD and Stonehill to upset Cornell in the first round, then everyone can have their inter-conference first game! (Full disclosure, I will be cheering for Cornell to defeat Stonehill in a well-played game by both teams.)
I guess we're lucky Minnesota had such a good season. It would have been fun to see what permutations people would have proposed if WCHA teams were 1,2,7,8! Would number 5 have to play number 2, and would number 5 still get the first round bye?
Let's all cheer for UConn to defeat UMD and Stonehill to upset Cornell in the first round, then everyone can have their inter-conference first game! (Full disclosure, I will be cheering for Cornell to defeat Stonehill in a well-played game by both teams.)
One solution to the debates in this thread could be for the NCAA committee to actually somehow formalize these tradeoffs between the Pairwise criteria, reducing travel, and intraconference quarterfinals in forming pairings.
You use Pairwise Criteria to seed and select teams. Then you use the additional criteria to score the different possible brackets, and you define ex-ante how brackets would be scored. This would reduce concerns about committee discretion and lack of transparency.
As we've noted, these additional concerns of reducing travel and encouraging intraconference quarterfinals have always been there in the handbook, but leaving their application purely to committee discretion has reduced their legitimacy, especially in contrast to the Pairwise criteria which are perfectly transparent.
I don't know exactly how to formalize scoring the brackets, but I suspect that one could look retrospectively and find a system that rationalizes most of the past 20 years of quarterfinal pairing decisions. Probably not all violations of bracket integrity are equal and could be scored somehow (e.g., if comparisons are not all transitive, or hinge on very close RPI/NPI).
Probably the main reasons this has not yet been done are (1) the NCAA has historically always left these tradeoffs to the discretion of the committee across all sports, and (2) the committeees have incentive to preserve some discretion in some task in order to justify their existence.
You use Pairwise Criteria to seed and select teams. Then you use the additional criteria to score the different possible brackets, and you define ex-ante how brackets would be scored. This would reduce concerns about committee discretion and lack of transparency.
As we've noted, these additional concerns of reducing travel and encouraging intraconference quarterfinals have always been there in the handbook, but leaving their application purely to committee discretion has reduced their legitimacy, especially in contrast to the Pairwise criteria which are perfectly transparent.
I don't know exactly how to formalize scoring the brackets, but I suspect that one could look retrospectively and find a system that rationalizes most of the past 20 years of quarterfinal pairing decisions. Probably not all violations of bracket integrity are equal and could be scored somehow (e.g., if comparisons are not all transitive, or hinge on very close RPI/NPI).
Probably the main reasons this has not yet been done are (1) the NCAA has historically always left these tradeoffs to the discretion of the committee across all sports, and (2) the committeees have incentive to preserve some discretion in some task in order to justify their existence.
Well, no, in that scenario, one WCHA intraconference quarterfinal is completely unavoidable according to the current rules, and the committe would have no discretion to avoid an intraconference quarterfinal. No one advocating for avoiding interconference quarterfinals is saying it should always dominate all concerns.
Sure, it was unlikely, but I am still mildly disappointed since there was a reasonable solution with 4 interconference quarterfinals guaranteed. Also, I believe that the current selection criteria underrate UMD at No. 8 compared to reasonable alternatives, so I find the outcome additionally disappointing for that reason (while acknowledging the pairwise/NPI is what it is). But sure, it's a bracket consistent with what this sport has been doing for many years, and I'm not going to lose sleep over it.
QuasiCosmos
Member
One way for the Pairwise to better indicate the quality of teams (and thus not "underrate" some teams and "overrate" others) would be to increase the number of inter-conference games during the regular season. Non-Ivy ECAC teams and CHA teams play about dozen non-league games, whereas WCHA and NEWHA teams play half-a-dozen, and HE and Ivy League teams play seven.
For example, this year Penn State got to play pairs of games against this year's tournament teams Connecticut, Colgate and Cornell; and a pair of games against a strong Northeastern team. Colgate played pairs of games against Ohio State, Penn State, and Minnesota Duluth.
The way to accomplish this is to somehow have more games each season (unlikely?) or for leagues to decrease the number of their league games. Surely creative minds could devise strategies to devise a schedule that wouldn't require significantly more travel costs.
The fans would enjoy seeing a wider variety of teams during the season. I suspect the players would benefit, too.
For example, this year Penn State got to play pairs of games against this year's tournament teams Connecticut, Colgate and Cornell; and a pair of games against a strong Northeastern team. Colgate played pairs of games against Ohio State, Penn State, and Minnesota Duluth.
The way to accomplish this is to somehow have more games each season (unlikely?) or for leagues to decrease the number of their league games. Surely creative minds could devise strategies to devise a schedule that wouldn't require significantly more travel costs.
The fans would enjoy seeing a wider variety of teams during the season. I suspect the players would benefit, too.
BadgerPete
Active member
I've been pining for some version of a WCHA-ECAC showdown for years. The difference in conference sizes make it inelegant, but if you just drop the four bottom ECAC teams it makes for some fun matchups. I had a basic plan that the matchups would be based on the previous years ranking with half of the home games going to each conference in order.
For instance, next year would be:
Ohio State at Colgate
Clarkson at Wisconsin
Minnesota at St. Lawrence
Cornell at UMD
St. Cloud State at Quinnipiac
Yale at Minnesota State
St. Thomas at Princeton
Brown at Bemidji State
Those are some fun teams to see, and some fun connectivity.
The second way would be to host mini-tournaments like volleyball does where three teams travel to one site and play two games each. For instance, Ohio State, Wisconsin, and Clarkson would all travel to Hamilton. OSU would play Colgate, Wisconsin would play Clarkson, and the next night they would switch.
More good noncons would be good for the sport and for the Pairwise. Win win.
ARM
Fan of chipmunk-like mascots.
Another approach would be to give up on the PairWise altogether. We've seen many iterations over the years. Originally, there was a "Last 16" component that was discarded, followed by abandoning the Teams Under Consideration. RPI has been chucked in favor of NPI, or whatever it is. We still have a flawed metric where they need to ignore "bad" wins and "good" losses, and add bonuses for wins that are so good that someone is truly impressed. Adding 3-on-3 OT made it even more suspect, as we don't know how much to count results decided in this manner; sometimes, they count them only as 2/3 and 1/3, but the same rules don't always carry over to H2H and COp. Granted, 3-on-3 OT tarnishes any model trying to rank teams on the basis of typical hockey outcomes.
I'd be more at peace with trusting bracket integrity across the board if I believed more in the integrity of the bracket that the PairWise yielded. For example, Grant's Krach has a different ordering of teams 3 through 7: with 3=Minnesota; 4=Clarkson; 5=Colgate; 6=SLU; 7=Cornell. UMD at No. 8 is one key ranking where KRACH and PairWise agree. Is that better? Not when viewed as a way to reduce conference pairings, as the Clarkson & Colgate match would be set in stone. If Minnesota feels the PairWise has done them wrong, they at least have the chance to go to Potsdam and do something about it; and if not, then they have previous experience at having the fighting nurses end their season.
lol...spoken like a true Gopher fan!!!
robertearle
Well-known member
ARM
Fan of chipmunk-like mascots.
Good catch. I started off by looking for KRACH on USCHO but couldn't find any there, although that may just be a personal shortcoming.
Regardless of whether Grant's KRACH is current or outdated, or any particular ordering, I would prefer a mathematical model that didn't require as many adjustments over the years as the PairWise does, or at least, a better formula to use as its backbone. When you have to ignore results because they are somehow misleading, it suggests that the basic model is flawed.
I don’t really care and I’m not into the math of the pairwise. But my initial reaction was OSU somehow got the tougher path despite being number 1.
it’s funny because “the math says it’s true” is being used to defend the bracket but in 2021, “the math is wrong,” justified Penn State’s exclusion. Is it possible the math is wrong again and OSU should at least be in a bracket with only 2 wcha teams? Or is the committee considering the last 2 outings by OsU against Wisconsin and thinking Wisconsin is more of a number 1, deserving to not have to deal with the WCHA until the championship game. They are seeded, the math is being followed, but I wonder if some folks think Wisconsin is the real number 1 right now.
Last year prior to the tourney Grant was like the west is better, there will be 3 WCHA teams in the frozen four and he was right. Then sure enough all wcha final.
As a Mercyhurst fan every year I cheered for the Lakers to get a non wcha draw. Every time they made it to the frozen four it was through a non WCHA path. Even in 2020 when things were cancelled, they were on their way to #1 Cornell, who they had already gotten a decent result against that season. I had hope that I wouldn’t have had if they were off to a #1 (or any ) WCHA team. Same thing for BC a few years back. They got to host, but they got OSU. Womp womp. They got shut out, and Pokechecker said it as well as you can: “welcome to the wcha.”
This is probably a wake up call OSU needed. Should be a wake up call for Bilka and Barnes too. They’ve won with the national team, but have they ever been “the one” to will a team to a championship? If I’m Jenn Gardiner, I’m thinking “it might be me, it has been me before” and trying to find a way to round up this team and lead them across the line together.
Again, not really outraged at the bracket, lots of cool games, but 3 WCHA teams on the side with #1 surprised me, knowing how historically strong that conference is.
it’s funny because “the math says it’s true” is being used to defend the bracket but in 2021, “the math is wrong,” justified Penn State’s exclusion. Is it possible the math is wrong again and OSU should at least be in a bracket with only 2 wcha teams? Or is the committee considering the last 2 outings by OsU against Wisconsin and thinking Wisconsin is more of a number 1, deserving to not have to deal with the WCHA until the championship game. They are seeded, the math is being followed, but I wonder if some folks think Wisconsin is the real number 1 right now.
Last year prior to the tourney Grant was like the west is better, there will be 3 WCHA teams in the frozen four and he was right. Then sure enough all wcha final.
As a Mercyhurst fan every year I cheered for the Lakers to get a non wcha draw. Every time they made it to the frozen four it was through a non WCHA path. Even in 2020 when things were cancelled, they were on their way to #1 Cornell, who they had already gotten a decent result against that season. I had hope that I wouldn’t have had if they were off to a #1 (or any ) WCHA team. Same thing for BC a few years back. They got to host, but they got OSU. Womp womp. They got shut out, and Pokechecker said it as well as you can: “welcome to the wcha.”
This is probably a wake up call OSU needed. Should be a wake up call for Bilka and Barnes too. They’ve won with the national team, but have they ever been “the one” to will a team to a championship? If I’m Jenn Gardiner, I’m thinking “it might be me, it has been me before” and trying to find a way to round up this team and lead them across the line together.
Again, not really outraged at the bracket, lots of cool games, but 3 WCHA teams on the side with #1 surprised me, knowing how historically strong that conference is.
robertearle
Well-known member
If you put the names of the eight teams in a hat and picked them at random, there's about a 15% chance that Ohio St would wind up with two other WCHA teams on their side of the bracket.
Last edited:
Can you explain a bit? Do you mean it is indeed surprising to have 3 WCHA’s including the #1 on one side? I don’t know if 15% is high or low in this context.
Regardless I admit I didn’t watch much all year until I paid $15 for a 1 month pass that got me the CHA final, last OSU reg season series game, wcha playoffs, and I believe some NCAA tourney! ( what a deal).
So I won’t vigorously defend anything I wrote, just sharing it as one fans reaction.
TonyTheTiger20
#SOAR
Thanks for this! Looks like one of my last refreshes didn't go through despite me updating the date. Try it now.