(03-19-2019 10:12 AM)ken d Wrote:  

(03-19-2019 09:26 AM)The Cutter of Bish Wrote:  Houston was #4 NET...got them to the three line. UNC only shares the regular season ACC title, bows out in the semi-final, ends at #7, gets the top line. Even over two very good SEC schools. Wofford is in four territory, gets the 7-10 stew for it (as arbitrary a string of seeds as anything; the only gift is not getting the 8/9 kiss of death). Cincy, hate their NC-SOS all one wants, still should be better than a seven.

If you seed based on more of a composite, it starts to look a little more orderly, but, it still doesn’t expalin all of the lines. Wofford really got hosed when metrics moved to seeding, while Belmont, and especially St. John’s and Temple benefitted.

Not a fan of NET, though. And it looks like KenPom, too, given how you have reachable unreachables, has some issues with its math, like Texas and Penn State overlooked by teams more than 25 spots away in some cases. We got a new metric this year, but I think it only proves that it validates some decisions more than others.

I would say, though, that it may only be a matter of time until a .500 team gets an at-large. A death schedule with enough great wins...a year like this year, valuing strong wins over placement in a conference and the tiers...it may push through.

There are 36 at large entries. If you take the top 36 teams in the Massey Composite that did not have an autobid, only four of those did not make the cut. One of them, Texas, finished at 16-16 despite having five wins against Top 30 teams.

The other three were NC State, Clemson and TCU. Their Top 30 W-L records weren't so good. NCSU was 1-9, Clemson 2-8 and TCU 2-7.

Three of the teams outside the Massey Top 36 non-champions who did make the field had much better Top 30 records. Arizona State (#41) was 4-1, while Seton Hall (#37) and St John's (#50) were both 4-3. Temple (#43) is the only outlier at 1-4. For the most part, it would appear that the committee rewarded wins against better teams more than they penalized for "bad losses" or weak OOC schedules.

Why Temple got a golden ticket is a mystery to me, but they weren't so low in the Massey as to be indefensible. At least the three lowest ranked teams to make the field have to play their way in to the Round of 64.

It's hard to be too critical of the committee this year.

Probably didn't hurt them that it was their longtime HC's final year coaching. Fran is very well liked around the country. Having him coach his final collegiate game in the NCAAT is exactly the kind of story lines the committee looks for. Similar to OU making it last year so that the Trae Young kid could be in it.

Yeah I was completely fine with Temple getting in even if it was on the basis of a tiebreaker of lifetime achievement bid. Fran’s final year is a great storyline.

(03-19-2019 09:26 AM)The Cutter of Bish Wrote:  Houston was #4 NET...got them to the three line. UNC only shares the regular season ACC title, bows out in the semi-final, ends at #7, gets the top line. Even over two very good SEC schools. Wofford is in four territory, gets the 7-10 stew for it (as arbitrary a string of seeds as anything; the only gift is not getting the 8/9 kiss of death). Cincy, hate their NC-SOS all one wants, still should be better than a seven.

If you seed based on more of a composite, it starts to look a little more orderly, but, it still doesn’t expalin all of the lines. Wofford really got hosed when metrics moved to seeding, while Belmont, and especially St. John’s and Temple benefitted.

Not a fan of NET, though. And it looks like KenPom, too, given how you have reachable unreachables, has some issues with its math, like Texas and Penn State overlooked by teams more than 25 spots away in some cases. We got a new metric this year, but I think it only proves that it validates some decisions more than others.

I would say, though, that it may only be a matter of time until a .500 team gets an at-large. A death schedule with enough great wins...a year like this year, valuing strong wins over placement in a conference and the tiers...it may push through.

IU was pretty close at 17-15. UT was 16-16 and was from #5 to #8 out. If there losses had been more towards the beginning of the conference schedule, they would have moved up a little.

I just can't see giving a .500 team an at large. That just demonstrates the tourney is too large.

(03-19-2019 01:07 PM)IWokeUpLikeThis Wrote:  Yeah I was completely fine with Temple getting in even if it was on the basis of a tiebreaker of lifetime achievement bid. Fran’s final year is a great storyline.

Temple was a very solid 8-7 vs Quad 1 and 2a. Also 10-7 away from home didn't hurt at all whatsoever.

(03-19-2019 01:07 PM)bullet Wrote:  

(03-19-2019 09:26 AM)The Cutter of Bish Wrote:  Houston was #4 NET...got them to the three line. UNC only shares the regular season ACC title, bows out in the semi-final, ends at #7, gets the top line. Even over two very good SEC schools. Wofford is in four territory, gets the 7-10 stew for it (as arbitrary a string of seeds as anything; the only gift is not getting the 8/9 kiss of death). Cincy, hate their NC-SOS all one wants, still should be better than a seven.

If you seed based on more of a composite, it starts to look a little more orderly, but, it still doesn’t expalin all of the lines. Wofford really got hosed when metrics moved to seeding, while Belmont, and especially St. John’s and Temple benefitted.

Not a fan of NET, though. And it looks like KenPom, too, given how you have reachable unreachables, has some issues with its math, like Texas and Penn State overlooked by teams more than 25 spots away in some cases. We got a new metric this year, but I think it only proves that it validates some decisions more than others.

I would say, though, that it may only be a matter of time until a .500 team gets an at-large. A death schedule with enough great wins...a year like this year, valuing strong wins over placement in a conference and the tiers...it may push through.

IU was pretty close at 17-15. UT was 16-16 and was from #5 to #8 out. If there losses had been more towards the beginning of the conference schedule, they would have moved up a little.

I just can't see giving a .500 team an at large. That just demonstrates the tourney is too large.

Indiana was 3rd team out in a year where we had 4 bid theives.

Texas we really don't know as the NCAA committee doesn't seed the NIT. Just because a team is a 2 seed in the NIT doesn't automatically mean they were the 5th-8th teams out of the tourney.

I think the committee is pretty consistent in that all of your results matter, no matter when they happen. So your comment that Texas would have been closer had the losses happened earlier- sorry but no. 16 losses is still.... 16 losses....

(03-19-2019 01:04 PM)WhoseHouse? Wrote:  

(03-19-2019 10:12 AM)ken d Wrote:  

(03-19-2019 09:26 AM)The Cutter of Bish Wrote:  Houston was #4 NET...got them to the three line. UNC only shares the regular season ACC title, bows out in the semi-final, ends at #7, gets the top line. Even over two very good SEC schools. Wofford is in four territory, gets the 7-10 stew for it (as arbitrary a string of seeds as anything; the only gift is not getting the 8/9 kiss of death). Cincy, hate their NC-SOS all one wants, still should be better than a seven.

If you seed based on more of a composite, it starts to look a little more orderly, but, it still doesn’t expalin all of the lines. Wofford really got hosed when metrics moved to seeding, while Belmont, and especially St. John’s and Temple benefitted.

Not a fan of NET, though. And it looks like KenPom, too, given how you have reachable unreachables, has some issues with its math, like Texas and Penn State overlooked by teams more than 25 spots away in some cases. We got a new metric this year, but I think it only proves that it validates some decisions more than others.

I would say, though, that it may only be a matter of time until a .500 team gets an at-large. A death schedule with enough great wins...a year like this year, valuing strong wins over placement in a conference and the tiers...it may push through.

There are 36 at large entries. If you take the top 36 teams in the Massey Composite that did not have an autobid, only four of those did not make the cut. One of them, Texas, finished at 16-16 despite having five wins against Top 30 teams.

The other three were NC State, Clemson and TCU. Their Top 30 W-L records weren't so good. NCSU was 1-9, Clemson 2-8 and TCU 2-7.

Three of the teams outside the Massey Top 36 non-champions who did make the field had much better Top 30 records. Arizona State (#41) was 4-1, while Seton Hall (#37) and St John's (#50) were both 4-3. Temple (#43) is the only outlier at 1-4. For the most part, it would appear that the committee rewarded wins against better teams more than they penalized for "bad losses" or weak OOC schedules.

Why Temple got a golden ticket is a mystery to me, but they weren't so low in the Massey as to be indefensible. At least the three lowest ranked teams to make the field have to play their way in to the Round of 64.

It's hard to be too critical of the committee this year.

Probably didn't hurt them that it was their longtime HC's final year coaching. Fran is very well liked around the country. Having him coach his final collegiate game in the NCAAT is exactly the kind of story lines the committee looks for. Similar to OU making it last year so that the Trae Young kid could be in it.

The committee says it doesn't consider things like that, but it happens often enough to make you wonder. Maybe the committee members really believe it's true and are just in denial.

Congratulations to Belmont, the first bubble team with a win in this year's tourney.

One thing to keep in mind- for folks that want advanced metrics to not be used, and have it just be results based- games like last night with Temple don't do anything to help your cause at all.

So far NET in this tournament is 2-0 picking games.

(03-19-2019 01:14 PM)stever20 Wrote:  

(03-19-2019 01:07 PM)IWokeUpLikeThis Wrote:  Yeah I was completely fine with Temple getting in even if it was on the basis of a tiebreaker of lifetime achievement bid. Fran’s final year is a great storyline.

Temple was a very solid 8-7 vs Quad 1 and 2a. Also 10-7 away from home didn't hurt at all whatsoever.

Temple was beaten very solidly by Belmont.

(03-19-2019 01:14 PM)stever20 Wrote:  

(03-19-2019 01:07 PM)IWokeUpLikeThis Wrote:  Yeah I was completely fine with Temple getting in even if it was on the basis of a tiebreaker of lifetime achievement bid. Fran’s final year is a great storyline.

Temple was a very solid 8-7 vs Quad 1 and 2a. Also 10-7 away from home didn't hurt at all whatsoever.

Yeah, that shouldn’t get lost at all in this. Good wins, decent work done on the road. When your schedule has seven games against tournament teams, and a few more with teams just on the outside (Memphis and Davidson), with wins against each type, that won’t be overlooked.

For what is said about Temple or St. John’s bids over TCU, it probably doesn’t help TCU seeing Lipscomb right by them metrically, with Lippy having little hope for the at-large and the head to head over the frogs.

I think it's interesting looking at the bracket matrix ratings section that this year was by far the best year ever that the matrix had.

Also for those who don't like Lunardi and want to use Jerry Palm.....
Lunardi 5 year standings #55 of 133
Palm 5 year standings #85

Seeding was fine.

If you look at the seeds and the NET, and figure schools that move up or down 5 or fewer places are normal after evaluating their actual opponents, then you see NET did a good first cut of order.

Also remember that seeds 12-16 are reserved for the bottom 22 conference champions. (The 11 seeds are for the last four at-large to play in and the next two lowest at-large or top 10 conference champions)

I will start with the latter seeds, 12-16, or seeds 47-68. Basically they are almost in NET order, with these exceptions:

1) Bradley was moved up from 63 overall (16 seed) to 61 (15 seed), almost certainly because the committee valued the MVC above the SLC and Big South. They just felt that the MVC should not get the ignominy of a 16 seed, it's too much better an overall conference.

2) Saint Louis similarly was moved from 55 to 53 overall, or pushed up from a 14 seed to a 13 seed for winning the A10, a conference that often has an at-large bid. Same principle

Absolutely nothing anyone can complain about in the bottom 22 seeds. It's just a bit embarrassing that the P12 got a 13, A10 a 13 and MVC a 15 seed, while the MAC and SoCon made the top 10.

Now I'm going to break up the first 46 seeds or 11 rounds in groups. Starting with rounds 1-3, seeds 1-12

None of the first 12 schools moved more than 5 spots from their NET, which is the normal range.

School (NET, seed)
Duke (3, 1)
Virginia (1, 2)
North Carolina (7, 3)
Gonzaga (2, 4)
Tennessee (5, 5)
Michigan State (8, 6)
Kentucky (6, 7)
Michigan (9, 8)
Houston (4, 9)
Texas Tech (10, 10)
Louisiana State (14, 11)
Purdue (12, 12)

Houston was going to get dinged regardless for their American opponents not matching the SEC, ACC or B1G, and then they lost the AAC Championship game to Cincy. They were never seriously considered a top 5 or 6 school, NET inflated their strength. Nobody in their right mind thinks that team is a match for the 8 above them. The only other obvious comment is the committee really likes the top of the ACC. Tennessee probably needed to win the SEC Championship game to dislodge Gonzaga who inexplicably laid an egg in the WCC title game.

Again nothing to complain about in seeds 1-3. Houston "dropped" 2 seeds simply because they were the lowest 1 in NET and the highest 3 in overall seeding. But 2 and 3 seeds are really the same, they don't face the 1 seed until the regional finals, and facing a 14 or 15 in the first round is still a give me. 4 and 5 seeds get the more dangerous 12 and 13 seeds, best of the conference winners such as Oregon and Saint Louis, P12 and A10 winners with major conference level athletes, plus very solid mid-majors such as Murray State, NMSU and UCI sitting there.

Now where the committee did it's biggest adjustments, were seeds 4-11. This is the only group where teams moved up or slid down more than 5 places compared to NET, and in fact only a few of them did. I will only address those that did:

4 and 5 seeds:
Kansas (20, 13)
K State (24, 15)
Marquette (28, 17)

It's pretty obvious the committee liked the top of the Big 12 resumes, both jumped 2 seeds
Marquette is the one Big East school the committee showed love to, jumped 2 seeds also

6 and 7 seeds:
Buffalo (15, 23)
Wofford (13, 28)

Committee clearly didn't think much of the MAC and SoCon competition, Buffalo got a 2 seed drop, Wofford a huge 3 seed drop

8, 9 and 10 seeds:
Syracuse (42, 30)
Washington (45, 33)
Florida (31, 40)

Committee liked Syracuse's resume and their ACC record, pushed them up 2 seeds
Committee appears to have overlooked UWs stinkers against Cal and then Oregon in the P12 Championship game, a 2 seed jump
Committee did not like Florida's resume, dropped them 2 rounds; biggest hit of any major school

11 seeds:
St. Mary's (32, 44)

They probably would not have made the tournament had they not won the WCC title game, an unimpressive resume to say the least. 3 round drop. (UNCG, Alabama, TCU or Indiana lost a spot due them). But they were still ahead of NMSU and Murray State enough in NET that they remained in the top 11 seeds.

When you actually look the overall seeds, only 9 schools moved more than 5 spots.

The mistake a lot of you guys make is considering the 12 seed conference champions. Those are evaluated separately from the first 10 conference champions and the at-large. The system is designed to give seeds 1-5 first round opponents from the lower 22 conference champions (4 and 5 seeds get the most dangerous ones and if they survive they face each other). The 6 seeds get the play-ins. And 7-10 is the churn, where if you survive you get to face a 1 or 2 seed in the next round. If you keep these parameters in mind, then the selection makes a lot of sense, and NET was a very useful tool in getting a rough sorting.

It should also be noted that seeds 1-28 also had NET rankings 1-28, just different order within.

The committee got it right IMO.

For whatever it's worth...

(03-20-2019 04:48 PM)Wedge Wrote:  For whatever it's worth...

but what is he going to change? If he doesn't have the wins over Kentucky and Maryland, they likely aren't dancing right now.

(03-20-2019 04:48 PM)Wedge Wrote:  For whatever it's worth...

Seton Hall got a 10 seed (38 overall). One factor I noticed in power conferences was they paid attention to conference record. That helped Syracuse (10-9) move ahead of Clemson and NC State (9-9) for example, and also move Temple (13-5) ahead of Memphis (11-7) for slots. Seton Hall went 9-9, which was better than St. John's (47) 8-10, but worse than Marquette (17) 12-6 and Villanova (21) 13-5. No question that was the bigger factor than NET among the majors and upper Mid-Majors.

(03-20-2019 05:06 PM)Stugray2 Wrote:  

(03-20-2019 04:48 PM)Wedge Wrote:  For whatever it's worth...

Seton Hall got a 10 seed (38 overall). One factor I noticed in power conferences was they paid attention to conference record. That helped Syracuse (10-9) move ahead of Clemson and NC State (9-9) for example, and also move Temple (13-5) ahead of Memphis (11-7) for slots. Seton Hall went 9-9, which was better than St. John's (47) 8-10, but worse than Marquette (17) 12-6 and Villanova (21) 13-5. No question that was the bigger factor than NET among the majors and upper Mid-Majors.

so how could Marquette get a better seed than Nova then?

Why is Baylor a 9 seed going 10-8 while Iowa St is a 6 seed going 9-9?

Sorry but conference record is pretty much MEANINGLESS!

maybe St John's shouldn't have gotten into the tourney after all... Down 29-13 and it's looking like an absolute blowout.

stever read St John’s like a book.

So NET 4-0 so far in the tourney.....

(03-20-2019 05:00 PM)stever20 Wrote:  

(03-20-2019 04:48 PM)Wedge Wrote:  For whatever it's worth...

but what is he going to change? If he doesn't have the wins over Kentucky and Maryland, they likely aren't dancing right now.

The only things that will help in making scheduling decisions for next year are one-on-one conversations with every committee member plus a crystal ball. What has become clear (and we should have expected this) is that selection committee members are going to go with their gut feelings about bubble teams, and then basically participate in the horse trading that inevitably happens when they don't all agree.

There are no metrics anybody can or should use to make their scheduling decisions. Your best bet is to be good enough to avoid the bubble in the first place. After that, it's a crap shoot.