Pool C

[deiscanton]

As pointed out by the people on D3Datacast and confirmed on the published schedules on the NCAA Statistics database that has been used over the past 2 weeks to create the official NCAA DIII Men's Basketball Summary Reports, the NCAA DIII regional advisory committees have been letting in some errors in the data entry of the games slip through that have skewed the official NCAA DIII NPI numbers for men's basketball.   The unofficial D3DataCast NPI numbers as entered through the Massey system have been the accurate NPI numbers for this season in men's basketball so far. 

Note-- the discrepancies in the NPI numbers have not been great enough so far so as to make the NPI rankings in the most crucial places of Top 8 seeding and at-large selections incorrect, but these errors have to be corrected or else the risk of getting an at-large selection wrong will increase with more data errors.

Here are the errors that we have to check to see if they are corrected in next Monday's NCAA DIII NPI Summary Report--

In Evaluation Region I, we have 2 GNAC (Great Northeast Conf) teams-- New England College and Dean, that played a non-conference game vs each other back in November, and the official conference game vs each other in January.  GNAC only plays a single round robin in men's conference play.

The NCAA DIII codes have listed under NE College for the games vs Dean for the past 2 summary reports--

Dean-- HL (C), AL (C)-- The AL code of (C) is not correct-- The away game is a non-conference game that was played back in November, and that should read AL (NC).

So the correct codes for NE College's opponent of Dean should read--

HL (C), AL (NC)

Similarly for Dean-- the codes for the opponent of New England College should read--

NE College-- AW (C), HW (NC)

For the games between UC Santa Cruz vs Salisbury and Christopher Newport, the NCAA has incorrectly coded those matchups as C2C conference games.  Officially, the C2C as listed in their men's basketball handbook-- a link to which I have posted in the Coast to Coast message Board-- has no regular season conference games before the C2C Tournament-- All matchups before the official conference tournament are to be considered as non-conference games.

So the codes of UC Santa Cruz counting Christopher Newport as AL (C) and Salisbury as AL (C) are not correct-- those games should both be coded as AL (NC).

Similarly, in Christopher Newport's schedule, the game vs UC Santa Cruz should be coded as HW (NC), and also in Salisbury's schedule, the game vs UC Santa Cruz should also be coded as HW (NC).

The Coast to Coast Conference has teams in what is still considered to be Regions V, VI, and X-- with the C2C RAC member serving in Evaluation Region VI.

Should those errors be corrected, and barring any other errors, the NPI numbers on the official NCAA DIII Men's Basketball Summary Report through games of Sunday, February 16, should match the D3 Datacast run numbers on the Monday morning, Feb. 17, 2025 run.  If there are still discrepancies next Monday, the data entries on the published schedules in the NCAA stats database will have to be checked again for errors. 

[Great Day to be a Royal]

Can anyone shed light on why Newport, Salsbury, and MW are not in a traditional conference? 

[Pat Coleman]

Short version: Schools generally left the CAC/C2C to get away from them. There aren't other D-III schools in the general area to replace Penn State-Harrisburg, St. Mary's (Md.), before them Wesley, Marymount and York (Pa.), before them Catholic and Goucher.

[Greek Tragedy]

https://d3datacast.com/npi/

The last Pool C bid has an overall rank of 42.

I included teams Pool C30 and above.

So, we don't have Regional Rankings anymore. We have the simple, yet super boring, NPI. Here is this morning's rankings via d3datacast. Not official, but pretty close.


Current (A) bids and Pool C position

AEC - 1 - Marymount (A) or Gwynedd-Mercy (31)
AMCC - 1 - Pitt-Bradford
ARC - 1 - Dubuque (A) or Nebraska Wes (23)
ASC - 1 - Hardin-Simmons (A) or UMHB (28)
C2C - 1 - CNU (A) or Salisbury (27)
CC - 2 or 3 - JHU (A), Gettysburg (14), Franklin & Marshall (21)
CNE - 1 - Endicott
CCIW - 2 - Carthage (A), IWU (6)
CCS - 1 - Belhaven
CUNYAC - 1 - John Jay
E8 - 1 - Brockport
GNAC - 1 - St. Joseph's (Maine)
HCAC - 1 - Anderson
LAND - 2 - Drew (A), Catholic (10)
LEC - 1 - Keene St
LL - 1 - RPI
MACC - 1 -York PA
MACF - 1 - Stevens
MASCAC - 1 - Westfield St
MIAA - 2 - Calvin (A), Trine (13)
MIAC - 2 - St. John's (A), Gustavus Adolphus (17)
MWC - 1 - Monmouth
NAC - 1 - Husson
NACC - 1 - WLC
NCAC - 1 - Denison (A) or Wooster (29)
NESCAC - 4 or 5 - Wesleyan (A), Trinity (4), Tufts (5), Hamilton (9), Williams (24)
NEWMAC - 1 - WPI
NJAC - 4 - Montclair St (A) Stockton (7), Ramapo (11), TCNJ (12)
NWC - 1 - Whitworth
OAC - 1 - John Carroll
ODAC - 4 - Hampden-Sydney (A) RMC (3), Virginia Wesleyan (19), Roanoke (20), Guilford (25)
PAC - 1 - Chatham
SAA - 1 - Berry
SCAC - 2 - St. Thomas (A), Trinity (22)
SCIAC - 2 - Redlands (A), Cal Lutheran (16)
SKY - 1 - Farmingdale St
SLIAC - 1 - Spalding
SUNYAC - 1 - Cortland
UAA - 4 - NYU (A), Emory (1), Wash U (8), Chicago (15), Brandeis (26), Carnegie Mellon (30)
UEC - 1 - Bryn Athyn
UMAC - 1 - North Central (MN)
USAC - 1 - Southern Virginia
WIAC - 2 or 3 - Platteville (A), La Crosse (2), Stevens Point (18),

[Great Day to be a Royal]

Short version: Schools generally left the CAC/C2C to get away from them. There aren't other D-III schools in the general area to replace Penn State-Harrisburg, St. Mary's (Md.), before them Wesley, Marymount and York (Pa.), before them Catholic and Goucher.

Wouldn't the ODAC seem like a geographical fit?

[Pat Coleman]

Short version: Schools generally left the CAC/C2C to get away from them. There aren't other D-III schools in the general area to replace Penn State-Harrisburg, St. Mary's (Md.), before them Wesley, Marymount and York (Pa.), before them Catholic and Goucher.

Wouldn't the ODAC seem like a geographical fit?

Sure.

But the ODAC has a bylaw limiting full members to private institutions. A geographical fit isn't the only consideration.

[ziggy]

Our updated Bubble Watch has been posted for games played through Saturday 2/15/25: https://d3datacast.com/npi/mbb-projections/

Interestingly, this morning's run of simulations produced just one team in our "bubble-out" category (which we've defined as receiving an at-large bid when needed in 25-50% of simulations). That puts Whitworth (39.9%) as the only team sitting between St. Thomas (51.3%) and Calvin (23.6%).

[bopol]

Our updated Bubble Watch has been posted for games played through Saturday 2/15/25: https://d3datacast.com/npi/mbb-projections/

Interestingly, this morning's run of simulations produced just one team in our "bubble-out" category (which we've defined as receiving an at-large bid when needed in 25-50% of simulations). That puts Whitworth (39.9%) as the only team sitting between St. Thomas (51.3%) and Calvin (23.6%).

I think in a system like this, things don't move that much by the end and there is actually a bizarre gap with only 2 teams in the 61s, but 15 in the 62s (mostly solid) and 10 in the 60s (mostly probably out).  The main drama this year will be the 'bid thieves'.

[ziggy]

Our updated Bubble Watch has been posted for games played through Saturday 2/15/25: https://d3datacast.com/npi/mbb-projections/

Interestingly, this morning's run of simulations produced just one team in our "bubble-out" category (which we've defined as receiving an at-large bid when needed in 25-50% of simulations). That puts Whitworth (39.9%) as the only team sitting between St. Thomas (51.3%) and Calvin (23.6%).

I think in a system like this, things don't move that much by the end and there is actually a bizarre gap with only 2 teams in the 61s, but 15 in the 62s (mostly solid) and 10 in the 60s (mostly probably out).  The main drama this year will be the 'bid thieves'.

I think that's right. Just a quick look at the top two groups (locks/near locks and solid), there are 15 bids "spoken for" without conference tournament upsets if you look at how many conferences are represented multiple times and take one away for the automatic qualifier.

That leaves six up for grabs before accounting for any bid-thieves, which happens to be exactly the number of teams in the "bubble-in" group.

So the final picture from here on out will be shaped by the number of bid-thieves, teams who are in good standing who might really stumble and hurt themselves and teams who are in less good standing who might still rack up some really good wins to help themselves.

[deiscanton]

The official NCAA DIII Men's Basketball NPI Summary Report for games through Feb. 16, 2025 has been released.   It is available at http://stats.ncaa.org/selection_rankings/nitty_gritties/42959

All of the mistakes pointed out over the past few weeks (C2C, GNAC) have been corrected, and the official NPI numbers match the D3Datacast numbers that were run this morning.

[ziggy]

The official NCAA DIII Men's Basketball NPI Summary Report for games through Feb. 16, 2025 has been released.   It is available at http://stats.ncaa.org/selection_rankings/nitty_gritties/42959

All of the mistakes pointed out over the past few weeks (C2C, GNAC) have been corrected, and the official NPI numbers match the D3Datacast numbers that were run this morning.

Not the full story but a good enough summary.

[deiscanton]

The official NCAA DIII Men's Basketball NPI Summary Report for games through Feb. 16, 2025 has been released.   It is available at http://stats.ncaa.org/selection_rankings/nitty_gritties/42959

All of the mistakes pointed out over the past few weeks (C2C, GNAC) have been corrected, and the official NPI numbers match the D3Datacast numbers that were run this morning.

Not the full story but a good enough summary.

I just did not want to repost the entire full story, Ziggy, but if I missed a few points in previous posts on the subject in forums such as Pool C and Coast to Coast Conference that should be brought up, feel free to do so.

The next big thing on the radar is how to code the conference tournaments.   In 41 of the 42 D3 men's basketball conferences that have these tournaments-- I suppose the results of the tournaments will use the home/away multiplier of 1.1 for road win/home loss, 1.0 for neutral court win/loss, and 0.9 for home win/road loss.   For the Coast to Coast Conference tournament, since there are no conference games in the sports that C2C members play in per league policy, that men's basketball tournament is being held at Salisbury.  Do C2C tournament results in games vs Salisbury played February 28 and/or March 1 (pre-determined host this year) use the 1.2 road win/home loss, 1.0 for neutral court win/loss, or 0.8 home win/road loss (non-conference multiplier), or are all of the C2C tournament games graded as neutral court games?

The DIII men's basketball committee is the only DIII sport committee using the home/away multiplier that decided to have a different home/away multiplier for non-conference vs conference games.   All other DIII sport committees that adopted a home/away multiplier for games decided to go with the 1.1/1.0/0.9 formula.  (DIII ice hockey being the best example.)

[ziggy]

You were correct with the data inaccuracies we made public but I was just meaning to refer to the fact that it was not an exhaustive list as we continued to comb through the data.

Conference tournament will enter the system as they are played and will be added to each team's resume consistent with any other conference game. In the case of the C2C, these are the only official conference games they play and will be weighted accordingly.

[deiscanton]

I did not mean to imply that new mistakes cannot come up.  If these mistakes happen, I will count on you, Ziggy, and D3Datacast to point them out. 

[deiscanton]

The NCAA DIII Men's Basketball Committee has released the official NPI summary report through games played on Monday, Feb. 17.    There were only 2 games played last night in the SCIAC, and those results are included in today's report.

Today's NPI summary report can be found at:

http://stats.ncaa.org/selection_rankings/nitty_gritties/42972

As far as I can tell, there are no notable mistakes this morning.