Liza Kelly

[njbill]

Top 5 and 10 according to RPI
Head to head is interesting. But not even close to dispositive.
Uva beat notre dame. But notre dame beat unc and was ranked higher than uva. A better win.

As I said, it depends on what top five ranking you use. RPI has never meant much of anything to me. I think it is in essentially worthless system. I won’t repeat here the many criticisms I have written about it over the years.

Head to head is more than “interesting.” Agreed it is not always dispositive, but in any ranking system, head-to-head between two closely ranked teams is the most important factor, certainly more important than some mathematically generated numbers result.

UVA and Notre Dame have mixed results. Yes, that makes it more complicated for them. Denver’s results are not mixed; nor are they complicated.

[njbill]

A much better example is Syracuse getting the two seed. Which they deserve. Even though bc beat them in the last game of the regular season and bc won the acc tourney. But Syracuse clearly has the better resume and their rpi is meaningfully better. Bc won the head to head game. But based on the compete season, cuse should be seeded higher…

I agree with you on this.

So: Northwestern, Syracuse, Denver, BC.

[Madlax59]

A much better example is Syracuse getting the two seed. Which they deserve. Even though bc beat them in the last game of the regular season and bc won the acc tourney. But Syracuse clearly has the better resume and their rpi is meaningfully better. Bc won the head to head game. But based on the compete season, cuse should be seeded higher…

I agree with you on this.

So: Northwestern, Syracuse, Denver, BC.

Not sure why unc is 3?

[Womenslaxxfan]

Unc is 3 because their rpi is incrementally better than bc’s. And I mean incrementally. I’d seed bc higher bc the rpi and sos differences aren’t statistically significant and bc just won the acc title.

But unc and bc have rpi numbers that are meaningfully above Denver’s. And the unc example takes the “but Denver beat bc head to head in march” argument out of the conversation.

Unc simply has a better resume than Denver because they have played a much tougher schedule…

Rpi is meant to approximate the overall strength of a team relative to other teams based on the teams own results AND the results of its opponents (a sort of strength of schedule complement, though not exactly). It’s not perfect as we’ve seen many times including notre dames exclusion from the men’s tourney last year..

BUT it does provide a tool to help the committee keep in mind not only the teams overall record and their “best” wins, but also the impressiveness of their overall body work relative to other teams bodies of work, particularly for teams that don’t play each other head to head or have very few if any common opponents.

I am completely fine with the committee seeding teams based in large part on how impressive their body of work was. Denver has some great “highs” and is deserving of a top 5 seed. But bc and unc had stronger seasons to this point precisely because of who they play.

The head to head against bc just isn’t enough for me to ignore the “season gauntlet” factor, which is encapsulated in RPI and SOS ratings..

And this will all be academic in three weeks. I hope Denver shows they deserve better than the five seed. It would be a great may if that were to happen!!!

[njbill]

Unc simply has a better resume than Denver because they have played a much tougher schedule…

Rpi is meant to approximate the overall strength of a team relative to other teams based on the teams own results AND the results of its opponents (a sort of strength of schedule complement, though not exactly). It’s not perfect as we’ve seen many times including notre dames exclusion from the men’s tourney last year..

BUT it does provide a tool to help the committee keep in mind not only the teams overall record and their “best” wins, but also the impressiveness of their overall body work relative to other teams bodies of work, particularly for teams that don’t play each other head to head or have very few if any common opponents.

I am completely fine with the committee seeding teams based in large part on how impressive their body of work was. Denver has some great “highs” and is deserving of a top 5 seed. But bc and unc had stronger seasons to this point precisely because of who they play.

The head to head against bc just isn’t enough for me to ignore the “season gauntlet” factor, which is encapsulated in RPI and SOS ratings..

And this will all be academic in three weeks. I hope Denver shows they deserve better than the five seed. It would be a great may if that were to happen!!!

Feel free to advocate for and put weight on RPI. Many people do. I don't. It is simply one of many tools to analyze and rank teams. But it shouldn't be the determining factor in seeding. I actually don't think it should be involved at all in seeding.

UNC had a tougher schedule than Denver to be sure, but SOS is (to me) largely meaningless. It is what you do with your schedule, not what your schedule is. UNC lost four games (to NU, ND, Syr, and BC). To me the cumulative effect of those losses drags them down below BC and Denver.

I've used this example many times over the years re SOS. Assume a ten team league. Each team plays the others once. Team 1 goes 9-0. Team 10 goes 0-9. Which team has the weakest SOS and which team has the strongest?

[Womenslaxxfan]

SOS for both is identical. But team a will be seeded and team b will miss the tourney because they sucked

[njbill]

Nope. It’s an extreme, theoretical example, of course, but the SOS for team 1 will be the lowest and the SOS for team 10 will be the highest.

[Womenslaxxfan]

Got it. Makes sense. 0-9 played teams better than it. So sos better.

[Womenslaxxfan]

What I mean is team 1 played all the other teams team 2 played AND beat team 2–giving team 2 the better sos…
Tricky. You learn something new every day.

[PhanLax99]

We might need laxreference to come in and correct me, but from what I understand is that the SOS for the committee is calculated by the top 10 highest rated teams that a team has played.

According to the most recently updated LR:

1. Bc (4)
2. Md (6)
3. Michigan (13)
4. Marquette (14)
4. UCONN (16)
5. Villanova (26)
6. Stanford (27)
7. Jacksonville (33)
8. Ohio State (42)
9. Georgetown (43rd)
10. Niagara (48)

Not bad! But let’s take a look at ND who finished 5th in the ACC.

1. Northwestern (1)
2. Cuse (2)
3. BC (3)
4. UNC (4)
5. UVA (11)
6. Liberty (33)
7. Clemson (34)
8. Duke (35)
9. VT (38)
10. Louisville (51)

Average RPI of opponent for ND is 20.1 to Denver’s 27.2. This where Denver (and Stony Brook in the past) ends up as a 5/6 seed. Plus for ND, the majority of their ACC counterparts will continue to go up in RPI as more conference tourneys start play. I full expect the contingent that is ranked in the 30’s to continue to rise as the teams above them playing lower RPI teams in their conference tournaments (looking at the 20’s falling).

Just for fun (I know it was the first game of the season) but Denver beat Louisville 5-4, while Notre Dame beat them 15-11.

[Puck Swami]

When you consider that this Denver team went a perfect 19-0, beat BC head to head, beat Maryland away, won 12 road games in seven different states, won nine games in the eastern time zone, and never let any team score 10 on them, it is utterly amazing. They did everything they were asked to do and then some...

Screw the RPI.

The eye test says Denver is quite deserving of a top three seed.

I hope the committee rewards them.

[Madlax59]

When you consider that this Denver team went a perfect 19-0, beat BC head to head, beat Maryland away, won 12 road games in seven different states, won nine games in the eastern time zone, and never let any team score 10 on them, it is utterly amazing. They did everything they were asked to do and then some...

Screw the RPI.

The eye test says Denver is quite deserving of a top three seed.

I hope the committee rewards them.

Can’t see them going higher than #5

[Puck Swami]

Can’t see them going higher than #5

I don't think there are four teams in the country better than Denver is right now, and neither should the committee, who needs to look beyond a skewed scheduling system to reward true merit.

The perfect 19-0 record, the beating of BC and Maryland on the road, the 12 road wins, and nobody scoring more than nine goals on them all season long should more than enough of any eye-test evidence to seed them higher than fifth.

It's not as if Denver ducked playing good teams - the Pioneers played as many of them as they could schedule and they beat them all. They did everything they were asked to do, and should not be penalized for not having as many opportunities as Big 10 and ACC schools get to play each other and pile up higher SOS numbers.

In my view, the P5 schools teams get enough major revenue advantages over all other schools that they should not also get rewarded for losing to each other, too.

[Madlax59]

Can’t see them going higher than #5

I don't think there are four teams in the country better than Denver is right now, and neither should the committee, who needs to look beyond a skewed scheduling system to reward true merit.

The perfect 19-0 record, the beating of BC and Maryland on the road, the 12 road wins, and nobody scoring more than nine goals on them all season long should more than enough of any eye-test evidence to seed them higher than fifth.

It's not as if Denver ducked playing good teams - the Pioneers played as many of them as they could schedule and they beat them all. They did everything they were asked to do, and should not be penalized for not having as many opportunities as Big 10 and ACC schools get to play each other and pile up higher SOS numbers.

In my view, the P5 schools teams get enough major revenue advantages over all other schools that they should not also get rewarded for losing to each other, too.

We will see tonight...I see them at 5. Im pretty sure BC and Cuse could beat them and prob UNC also. We will see if im right in the tourney.

[Puck Swami]

Sorry, DU already beat BC 13-8 at Jacksonville this season, so I don't know how you morally seed BC ahead of Denver this season after the head-to-head win in Denver's favor. A Head-to-head win should be more important than an SOS ranking.

And a four-loss UNC team (who lost to #12 Notre Dame) is to be seeded ahead of an undefeated Denver team that beat the defending ACC Champ this season? I find that to be egregious...

Alas, I am biased. But the system sucks.

[Madlax59]

Sorry, DU already beat BC 13-8 at Jacksonville this season, so I don't know how you morally seed BC ahead of Denver this season after the head-to-head win in Denver's favor. A Head-to-head win should be more important than an SOS ranking.

And a four-loss UNC team (who lost to #12 Notre Dame) is to be seeded ahead of an undefeated Denver team that beat the defending ACC Champ this season? I find that to be egregious...

Alas, I am biased. But the system sucks.

As I said
#5
BC is a different team than when they played

[Puck Swami]

DU was badly screwed as a five seed. Egregiously. Even ESPN surprised.

By the way, Denver has improved since playing BC, too.

[Wheeler199]

Liza Bringing in a few transfers:

Olivia Penoyer (Yale) & Jane Earley (Middlebury) to Denver