Obviously, this doesn’t impact JMU or any other CAA team this year, but how will the NET impact mid-majors who are seeking at-large bids?
It will be very interesting to see if the NCAA is all-in on the NET and uses it to justify and explain selections as well as seeding. They spent a lot of time researching and developing the algorithm to replace RPI and publicly announced that the NET would be the primary sorting tool for committee members. Will they honor it?
Ratings will shufffle a little over the next 24 hours based on conference tourney finals. As of today, these mid majors should be selected ahead of the power conference teams if the NCAA committee abides by the NET.
Furman 42
Belmont 47
Lipscomb 50
TCU 51
Indiana 53
Ohio St 55
Oregon 56
Minnesota 57
Arizona St 63
(03-16-2019 05:47 PM)Hart Foundation Wrote: [ -> ]Obviously, this doesn’t impact JMU or any other CAA team this year, but how will the NET impact mid-majors who are seeking at-large bids?
It will be very interesting to see if the NCAA is all-in on the NET and uses it to justify and explain selections as well as seeding. They spent a lot of time researching and developing the algorithm to replace RPI and publicly announced that the NET would be the primary sorting tool for committee members. Will they honor it?
Ratings will shufffle a little over the next 24 hours based on conference tourney finals. As of today, these mid majors should be selected ahead of the power conference teams if the NCAA committee abides by the NET.
Furman 42
Belmont 47
Lipscomb 50
TCU 51
Indiana 53
Ohio St 55
Oregon 56
Minnesota 57
Arizona St 63
We all know the committee will do what they want.
But, in case anyone wants to see that in writing:
NCAA explicitly says of NET:
“It is not a deciding factor.”
“It is not going to determine if a team is in or out of the bracket.”
“It is an organizational piece for the committee.”
https://www.ncaa.com/news/basketball-men...-they-mean
What I always find amusing is that you could have 2 P5 teams sandwiched around a mid major in all the polls and have their seedings be noticeably different. Mark my words, Wofford and Buffalo will be seeded 2 slots worse than the P5s on either side of them.
It’s ridiculous a team in a p5 can finish 8th out of 10 teams and be 7-12 in conference and still get in...Oklahoma.
And to think many bracketologists have 16-16 Texas still in the mix is even more ridiculous.
Losing doesn’t matter anymore, it’s all about how many of your 40 chances against quad 1 teams (read: other p5 teams) can you win. If it’s 3 or 4, you’re in, even if you lose the other 36 times.
I’m exagerrating the numbers of course, but you get the point.
Oregon stole a bid last night by winning the PAC 12 with 4 wins in 4 days.
The official NCAA NET explanation is below. They claim he NET will be the primary sorting tool for the committee and all team sheets will be sequenced as such. At the end of the article is throw out the disclaimer that subjective selection will still be a part of the process.
We shall see if anything has hanged or if it is the same ol same ol system rewarding teams with losing conference records.
Also, if you want to see exactly what the committee members see, you can download the team sheets and compare them side by side just like they do. It really has become an exercise of who can accurately analyze the full body of work with all data on a one-pager. I think what we will find is a continuing inability of smart committee members to evaluate qualitative info. I.e referencing a selected team as having 4 quad one wins without mentioning they only went 4-12 in those games winning only 25% of their chances. Mid majors only get a few chances by design so they will never have quantity but can have quality.
https://www.google.com/amp/s/www.ncaa.co...king%3famp
(03-17-2019 08:00 AM)Hart Foundation Wrote: [ -> ]Oregon stole a bid last night by winning the PAC 12 with 4 wins in 4 days.
The official NCAA NET explanation is below. They claim he NET will be the primary sorting tool for the committee and all team sheets will be sequenced as such. At the end of the article is throw out the disclaimer that subjective selection will still be a part of the process.
We shall see if anything has hanged or if it is the same ol same ol system rewarding teams with losing conference records.
Also, if you want to see exactly what the committee members see, you can download the team sheets and compare them side by side just like they do. It really has become an exercise of who can accurately analyze the full body of work with all data on a one-pager. I think what we will find is a continuing inability of smart committee members to evaluate qualitative info. I.e referencing a selected team as having 4 quad one wins without mentioning they only went 4-12 in those games winning only 25% of their chances. Mid majors only get a few chances by design so they will never have quantity but can have quality.
https://www.google.com/amp/s/www.ncaa.co...king%3famp
Yep. And on the last point, almost all the quad 1 games by mid majors are on the road, because p5s will not schedule a true road game against a mid major anymore.
Oklahoma can lose 5-6 in a row in conference, then beat a Kansas at home, and the reason for letting them in will be,
“Oklahoma beat Kansas, they have demonstrated an ability to win games in the ncaa tournament with that win, they belong in the tourney”
Here are some of my quick thoughts on this after reading the NCAA's explanation on this new system, looking at the Net rankings and looking at some schedules.
1. Some people, perhaps like Dukeman2, tout the strength of the Southern Conference this year. If you look more closely, that conference has 4 teams with really good records in Wofford, Furman, ETSU and UNCG, with all the rest having terrible records. Wofford played a number of highly ranked teams, but lost them all, many convincingly. I believe I saw an early win at Villanova by Furman I believe it was, which is commendable. In any event, I suspect these teams have benefited in the rankings from playing the name teams, but really aren't that great. Using the eye test the other day, I was not terribly impressed by the caliber of athlete I saw briefly in one of the Southern's marquee matchups.
2. The CAA, as usual, gets screwed. I actually think it's a good league and a team like Hofstra should get some consideration but doesn't. Compare VCU to Hofstra and tell me that VCU is so much better that it merits an at large while Hofstra with 27 wins isn't even in the conversation. They played, by the way, with VCU winning by 2 points in OT at VCU, I believe it was. VCU also lost to the College of Charleston and to ODU by quite a bit.
3. Why is ODU only 100 in the Net ranking when I know they're much better than that using the eye test? They have a really good record, have beaten VCU and won at Syracuse, yet are only a 14 seed. Meanwhile, many have Syracuse, not one of Boeheim's better teams, as a 9 seed.
4. Why, like DD says, is Texas even in consideration and ranked no. 38 in the Net ranking? They've only played 10 road games, whereas CAA teams play 15, have lost at home to VCU and Redford, while also losing 5 of their last 6 overall. On top of that, they're 8-10 in the Big 12. Shaka Smart could be on the way out after not winning an NCAA tournament game since he's been there, while their former coach, Rick Barnes, is leading TN to greatness.
5. Not sure this Net system is any better than RPI...in fact it may be worse. P5 schools all start the season with gaudy out of conference records because they play all their games at home to build up a strong rating and then play each other the rest of the way. The eye test still needs to be followed as does comparing teams outside of metrics that may be biased, qualitatively, as Hart says.
6. In what little I watched of the Pac 12 final, I was completely unimpressed. The supposed best team, Washington, got waxed in the final and scored very few points.
I think KenPom had St. Mary's (28-5) ranked 28th last year and they got snubbed. Thank goodness the Committee let in 19-15 Alabama and 18-13 Oklahoma instead.
It looks like Belmont got some love in the at large selection. However the committee went all the way to NET ranking 73 to select St Johns, 63 to select Arizona State, 57, to select Seton Hall, 56 to select Temple. That was strange.
Clemson and NC State in the low 30’s should be irate.
Why did the NCAA develop an official NET rating if they were going to go off the grid for the last selections?
Furman was ranked 41 and has a gripe with their road win over Villanova. I think their failures against other Socon mid majors in consideration doomed them.
Lost twice to Wofford and lost 2 of 3 to UNCG.
The NET ranking since it uses some scoring margin and efficiency margin (read uncapped scoring margin) is worlds better than the RPI for determining who actually the best teams are. It is a major improvement and I think part of the reason there is so little griping about selections this year. So to use NET to list/show which teams had the best records vs Quadrant 1/2/3/4 during the selection process makes a lot of sense as you are using a predictive type ranking to rank wins/accomplishments. That doesn't mean you just use the NET to guide selection by itself. That being said are we attempting to choose the "best" teams? If so you could probably do a mix of kenpom and Sagarin predictor and call it a day.
It seems like selection is always a mix of "best" and "deserving".
NC ST with its turnover causing style bullied its way against the 353rd (nations worst) OOC schedule to help inflate their numbers. They went from preseason kenpom 36th to 20th until conf play started up again and then regressed power rating wise towards their preseason projection.
If we are trying to choose the best (not most accomplished) teams throw Kenpom, sagarin and some oddsmakers in a room and just hash out the best teams, screw the resume's. NC St probably would be favored on a neutral vs St Johns, but John's was 5-7 vs Quad 1's and NC St was 3-6 with again the nations worse OOC. I can see how they got to selecting St John's as more deserving even if they aren't "better".
Not saying we deserve it any more than them, but how does Longwood get an invite to the CBI with a 15-17 overall ecord and 5-11 in the Big South?
CBI and CIT is usually about a teams willingness to play and also to pay for home games if it so comes down to that later in the tournament. Cal State Northridge has 20 losses and is in the CBI I believe. If you have a young team where everyone is coming back and are in the growth phase for your program it can make sense to continue practicing and working together as a team potentially building on the future. But it is a cash expense and for most schools it isn't something they are interested in.
Isabell, who played for Drexel after transferring from Missouri, is now on St. Louis. They squeaked into the tourney late yesterday.
(03-18-2019 09:59 AM)UofRfan Wrote: [ -> ]The NET ranking since it uses some scoring margin and efficiency margin (read uncapped scoring margin) is worlds better than the RPI for determining who actually the best teams are. It is a major improvement and I think part of the reason there is so little griping about selections this year. So to use NET to list/show which teams had the best records vs Quadrant 1/2/3/4 during the selection process makes a lot of sense as you are using a predictive type ranking to rank wins/accomplishments. That doesn't mean you just use the NET to guide selection by itself. That being said are we attempting to choose the "best" teams? If so you could probably do a mix of kenpom and Sagarin predictor and call it a day.
It seems like selection is always a mix of "best" and "deserving".
NC ST with its turnover causing style bullied its way against the 353rd (nations worst) OOC schedule to help inflate their numbers. They went from preseason kenpom 36th to 20th until conf play started up again and then regressed power rating wise towards their preseason projection.
If we are trying to choose the best (not most accomplished) teams throw Kenpom, sagarin and some oddsmakers in a room and just hash out the best teams, screw the resume's. NC St probably would be favored on a neutral vs St Johns, but John's was 5-7 vs Quad 1's and NC St was 3-6 with again the nations worse OOC. I can see how they got to selecting St John's as more deserving even if they aren't "better".
This is one of the criticisms I’ve seen of NET. If you absolutely pound low ranked teams, it apparently helps your rating. NC St did not look so great outside of that.
Maybe this will incentivize P5 schools to play JMU? Until we actually win a few that is. OTOH, they didn’t go dancing, so maybe it not.
To be fair, pounding bad teams has predictive value. Sagarin and Kenpom both use margin of victory to help rate teams. However, when determining who most deserves to be included in the NCAA field quality wins/losses seems like it is being leaned on more as of now. And to me that seems fine.
I never had a problem w/ including margin of victory in strength ratings. It has some merit. A good example would be our games against Bridgewater and EMU. Any half decent DI basketball team should easily pound their DIII opponents by 40 points. Our margin of victory was considerably smaller, 23 & 28 pts. To me, that was a big red flag for our upcoming season.
North Carolina State complained that they are 33rd in the NET and better than that in other rankings. Hofstra goes there in the NIT tonight. North Carolina State opened by beating three horrible teams by a combined score of 300-153. They're 97th in the RPI, which doesn't include margin of victory.
Post has an article on Net ranking that states that the algorithm has not been publicly disclosed, but incorporates offensive and defensive efficiency (like Kenpom basically) and margin of victory. NC State had the easiest out of conference schedule and pounded those teams but didn't fare well versus top half of ACC. State, Clemson and Texas all had Net rankings in the 30s and were not selected.
(03-18-2019 12:20 PM)Dukes84 Wrote: [ -> ]Isabell, who played for Drexel after transferring from Missouri, is now on St. Louis. They squeaked into the tourney late yesterday.
Isabell definitely made a good decision to transfer out of Drexel for his final year. He torched the CAA last year for 21ppg, 8 rpg, and 3 apg.
He transferred to St Louis who made the NCAA tourney on the strength of his A-10 tourney MVP performance.
Hmmm. Go dancing with an A-10 team or play with a crappy Drexel team.
Doesn’t always work out that way. Ryan Daly sat out at St Joe’s this year and Martelli was fired today.