Pool C

[magicman]

hugenerd,
That was a great explanation. Plus k from me as well. If I remember correctly didn't you have to give a similar explanation last year around this time? 

[Hoosier Titan]

Another +k from me, hugenerd.  Great explanation!

[golden_dome]

Thanks.  It would be nice if the ASC adapted to the system by changing their format, but then again you never know if the NCAA will change the system again.  However, in any system if you are playing 80% (or more, non-d3 and out-of-region games dont count) of your region games in conference it is hard to get a handle how good a team is with respect the rest of the country.  With such a small sample size against non conference opponents (5 games or less), it is hard to think of a system where you would get a good idea of the relative strength of that team against the rest of the country.  Massey only has the ASC as having 67  games against out-of conference opponents (I dont know how many are non d3 or non region), which comes out to about 4.5 per team.  I think the system could use reform regardless.  However, I guess there is some added bonus to the schools, because I am sure they dont have to worry as much about filling a schedule and they probably have more consistent travel costs.

Hugenerd,
   I agree with a lot of your points, but there is very little a conference like the ASC can do to "change their format" to better fit the system. I noticed a couple posts about the current system being better, which it is for some conferences, but for many it is worse. This is just an opinion, but this system is still not much better, it just shifted around which teams receive benefit and which do not.

For example, in an area like the deep south there just aren't the # of DIII teams. The 20/21 conference games has less to do with travel costs and more to do with just playing DIII games. If the ASC lowers the number of conference games and travel costs are taken into consideration, then many teams will face local NAIA schools and only play 15-16 DIII games. A great deal of money is already spent on travel with teams so spread out, which makes it even harder to actually spend more money to find quality DIII opponents for those 4-5 nonconference games.

Conferences like the NESCAC have been further strengthened with the system due to the availability of weaker conferences to prey on and more emphasis on the strength of schedule. But to be honest I don't know how you could improve the system unless you factored in NAIA teams, which I'm not sure would be much better anyway. Teams playing in regions of the country where DIII opponents are plentiful will continue to benefit from the ranking system and the opportunity to play home sectionals in the NCAA Tourney.

Having said that, I do think there has been a lot of effort to make the process better, and you certainly can't fault teams or conferences for using whatever advantages they have to improve their situation. But the system isn't any better for regions with few DIII teams like the south and the west coast.

[Hugenerd]

Thanks.  It would be nice if the ASC adapted to the system by changing their format, but then again you never know if the NCAA will change the system again.  However, in any system if you are playing 80% (or more, non-d3 and out-of-region games dont count) of your region games in conference it is hard to get a handle how good a team is with respect the rest of the country.  With such a small sample size against non conference opponents (5 games or less), it is hard to think of a system where you would get a good idea of the relative strength of that team against the rest of the country.  Massey only has the ASC as having 67  games against out-of conference opponents (I dont know how many are non d3 or non region), which comes out to about 4.5 per team.  I think the system could use reform regardless.  However, I guess there is some added bonus to the schools, because I am sure they dont have to worry as much about filling a schedule and they probably have more consistent travel costs.

Hugenerd,
   I agree with a lot of your points, but there is very little a conference like the ASC can do to "change their format" to better fit the system. I noticed a couple posts about the current system being better, which it is for some conferences, but for many it is worse. This is just an opinion, but this system is still not much better, it just shifted around which teams receive benefit and which do not.

For example, in an area like the deep south there just aren't the # of DIII teams. The 20/21 conference games has less to do with travel costs and more to do with just playing DIII games. If the ASC lowers the number of conference games and travel costs are taken into consideration, then many teams will face local NAIA schools and only play 15-16 DIII games. A great deal of money is already spent on travel with teams so spread out, which makes it even harder to actually spend more money to find quality DIII opponents for those 4-5 nonconference games.

Conferences like the NESCAC have been further strengthened with the system due to the availability of weaker conferences to prey on and more emphasis on the strength of schedule. But to be honest I don't know how you could improve the system unless you factored in NAIA teams, which I'm not sure would be much better anyway. Teams playing in regions of the country where DIII opponents are plentiful will continue to benefit from the ranking system and the opportunity to play home sectionals in the NCAA Tourney.

Having said that, I do think there has been a lot of effort to make the process better, and you certainly can't fault teams or conferences for using whatever advantages they have to improve their situation. But the system isn't any better for regions with few DIII teams like the south and the west coast.

Agreed. In any system, it seems like you are going to have winners and losers.  I think the NCAA is trying to use as simple a formula as possible to solve the problem (hence the 1-15 point system earlier and now the OWP and OOWP), but the problem doesnt appear to be that simple.  It may be worthwhile to compile a more complicated statistic, such as the RPI or whatever Massey uses.  It just seems a bit unjust to me that the number of conference games you play can affect two primary NCAA criteria (not even the quality of your league or oppoents, but the number of conference games).

Anyway thanks for the positive feedback, and I did make a similar explanation last year.  I think I went a little more in depth last year with examples but I think everyone who wants to get it gets it by now.

The bigger problem with OWP and OOWP is that it not only clusters all the numbers together for conferences like the ASC, but all the numbers in general are pretty tightly grouped around 0.500. If you look at the numbers for the entire country, the spread is very small: the average for all OWP numbers is roughly 0.50 with a standard deviation of 0.054 and for OOWP the average is 0.50 with a standard deviation of 0.027.  This essentially means that you need to look at the third decimal place to distinguish between most teams and what does that mean really?  Does the NCAA really choose team A over team B if they have similar records and team A has an OWP of 0.530 and team B has an OWP of 0.527?  I think OWP and OOWP is really only effective if there is a large gap between two schools, like 0.25 or higher, but even then you need to be careful when comparing teams between regions and teams that are in dissimilar conferences.  OWP and OOWP seem like they could be useful when ranking teams within a region, but I would have a lot of reservations when using it to select Pool C teams when it comes down to 8 teams from 8 different regions.

[sac]

RPI -- isn't any better, it uses OWP and OOWP, ........what does that really tell you?

RPI isn't without controversy on the D1 level either, they've just been force fed it now for so long, they've come to accept it.

[Ryan Scott (Hoops Fan)]

RPI -- isn't any better, it uses OWP and OOWP, ........what does that really tell you?

RPI isn't without controversy on the D1 level either, they've just been force fed it now for so long, they've come to accept it.

They also select their at-large berths with national criteria and the RPI is not a determining factor, but an aid.

If the selection process wasn't so tied to OWP and OOWP, people might have less complaints.

[golden_dome]

The bigger problem with OWP and OOWP is that it not only clusters all the numbers together for conferences like the ASC, but all the numbers in general are pretty tightly grouped around 0.500. If you look at the numbers for the entire country, the spread is very small: the average for all OWP numbers is roughly 0.50 with a standard deviation of 0.054 and for OOWP the average is 0.50 with a standard deviation of 0.027.  This essentially means that you need to look at the third decimal place to distinguish between most teams and what does that mean really?  Does the NCAA really choose team A over team B if they have similar records and team A has an OWP of 0.530 and team B has an OWP of 0.527?  I think OWP and OOWP is really only effective if there is a large gap between two schools, like 0.25 or higher, but even then you need to be careful when comparing teams between regions and teams that are in dissimilar conferences.  OWP and OOWP seem like they could be useful when ranking teams within a region, but I would have a lot of reservations when using it to select Pool C teams when it comes down to 8 teams from 8 different regions.

I haven't given this a great deal of thought, but I wonder what the bracket would look like if each region had a specified number of pool c teams. At least way each region would be comparing apples to apples instead of having a west coast pool c team against a northeast region pool c team.

[Dave 'd-mac' McHugh]

Chris - that sounds like what the NCAA used to do! All bids were allocated to a certain number of teams from each region. That caused some MAJOR problems when teams that were obviously strong in say the Midwest were not taken while teams who were certainly less strong in the Mid-Atlantic were taken. That isn't a solution worth going back to!

[Dark Knight]

The main issue, for me, with OWP and OOWP, is that the metric relies too heavily on the number of conference games your league has.  You could be in the best conference, but if your schedule has 2/3 conference games, you are essentially weighting your OWP by 2/3*(0.5)+1/3(OWP of out of conference teams).

This points out a real problem -- but it may actually understate the problem.  Teams don't play themselves, so teams with records above .500 will have OWPs below .500 for conference games, and vice versa.

If a team goes 13-1 in a conference with 8 teams playing a double round robbin, that means the remaining teams will have 36 wins and 48 losses, or an OWP of .428 for conference games. A team going 3-11 would have an OWP of .548.

So above average teams get a ding to the OWP that depends on how far above average they are in a conference and how many games they play in-conference -- but not on how good the conference is, directly. Their OWP would be weighted by 2/3 * (something less than 0.5) + 1/3 (OWP of out-of-conference games).

I expect this could be addressed by adding some kind of quality-of-conference term...

[hickory_cornhusker]

The main issue, for me, with OWP and OOWP, is that the metric relies too heavily on the number of conference games your league has.  You could be in the best conference, but if your schedule has 2/3 conference games, you are essentially weighting your OWP by 2/3*(0.5)+1/3(OWP of out of conference teams).

This points out a real problem -- but it may actually understate the problem.  Teams don't play themselves, so teams with records above .500 will have OWPs below .500 for conference games, and vice versa.

If a team goes 13-1 in a conference with 8 teams playing a double round robbin, that means the remaining teams will have 36 wins and 48 losses, or an OWP of .428 for conference games. A team going 3-11 would have an OWP of .548.

So above average teams get a ding to the OWP that depends on how far above average they are in a conference and how many games they play in-conference -- but not on how good the conference is, directly. Their OWP would be weighted by 2/3 * (something less than 0.5) + 1/3 (OWP of out-of-conference games).

I expect this could be addressed by adding some kind of quality-of-conference term...

That problem could be solved by eliminating that game from your opponent's winning percentage. Say you beat Team A who is 12-5. Then that teams record these purposes would be 12-4. Then you wouldn't be hurting your own OWP by beating teams.

[KnightSlappy]

Dark Knight,  it is true that teams don't play themselves but they don't get hurt for beating teams either.  If a conference had 8 teams and none of them played any in-region-out-of-conference games, everyone would have an OWP of 0.500.  (You don't coun't a teams loss against yourself when figuring OWP).  After the first round of games, each team would have played 7 games and, for OWP purposes, would figure each opponent as playing 6 games (example: Calvin doesn't count Hope-Calvin games while figuring out Hope's contribution to their OWP so beating Hope doesn't hurt Calvin's OWP).

As hugenerd pointed out the value of OWP and OOWP as a metric increases as the amount of region and conference mixing increases.  The more in-breading that goes on (conferences like my example above) the more and more the OWP number trends toward 0.500 for all teams.

[Greek Tragedy]

WEEK 1	WEEK 2
ATLANTIC REGION	ATLANTIC REGION
Richard Stockton 19-2, 16-2	Richard Stockton 21-2, 18-2
Farmingdale State 16-2, 16-2	Farmingdale State 19-2, 19-2
William Paterson 15-4, 15-4	William Paterson 18-4, 18-4
Baruch 17-4, 15-3	Baruch 19-4, 17-3
Montclair State 15-5, 12-5	St. Joseph's (L.I.) 19-2, 18-2
EAST REGION	EAST REGION
Ithaca 17-1, 15-1	Ithaca 20-1, 18-1
St. Lawrence 13-4, 11-4	St. Lawrence 16-4, 14-4
Hamilton 12-5, 10-1	Hamilton 14-5, 12-1
Rochester 13-5, 13-5	Rochester 15-5, 15-5
Rochester Tech 14-4, 11-4	Rochester Tech 16-4, 12-4
GREAT LAKES REGION	GREAT LAKES REGION
Capital 17-2, 15-2	Capital 18-3, 16-3
Carnegie Mellon 14-4, 10-3	Carnegie Mellon 16-4, 12-3
John Carroll 14-4, 12-3	John Carroll 16-4, 14-3
Calvin 12-6, 7-1	Calvin 14-6, 9-1
Hope 14-5, 7-2	Wooster 15-6, 14-4
Ohio Northern 14-5, 12-4	Ohio Northern 15-6, 13-5
MIDDLE ATLANTIC REGION	MIDDLE ATLANTIC REGION
Franklin and Marshall 17-2, 16-2	Franklin and Marshall 19-2, 18-2
Gwynedd-Mercy 15-2, 14-1	DeSales 17-3, 16-3
DeSales 16-3, 15-3	St. Mary's (Md.) 18-4 15-2
Wesley 12-5, 10-1	Wesley 15-6, 13-2
St. Mary's (Md.) 15-4, 12-2	McDaniel 16-5, 14-4
McDaniel 14-5, 12-4	Gwynedd-Mercy 15-5, 14-4
Widener 14-5, 12-5	Widener 16-5, 14-5
Gettysburg 13-5, 13-5	Gettysburg 14-6, 14-6
MIDWEST REGION	MIDWEST REGION
Washington U. 17-1, 16-1	Washington U. 19-1, 18-1
Elmhurst 15-4, 15-4	Wheaton (Ill.) 18-3, 14-3
Wheaton (Ill.) 16-3, 12-3	Tranylvania 16-4, 13-2
St. Norbert 16-1, 15-1	Elmhurst 16-5, 16-5
Tranylvania 14-4, 11-2	St. Norbert 17-3, 16-3
Augustana 15-5, 14-5	Augustana 16-6, 15-6
North Central 13-6, 11-4	North Central 14-7, 12-5
Carroll 13-5, 13-5	Lawrence 14-6, 14-6
NORTHEAST REGION	NORTHEAST REGION
Worcester Polytech 16-3, 16-2	Middlebury 20-2, 18-1
Middlebury 18-2, 16-1	Mass-Dartmouth 19-2, 17-2
Mass-Dartmouth 17-2, 15-2	Elms 18-1, 17-1
Elms 17-1, 16-1	Worcester Polytech 16-4, 16-3
Salem State 15-4, 14-4	Rhode Island College 16-4, 16-4
MIT 14-5, 13-3	Bridgewater State 15-5, 14-3
Colby 13-5, 10-4	Salem State 17-4, 16-4
Rhode Island College 14-4, 14-4	Amherst 18-4, 16-4
Amherst 16-3, 14-3	MIT 16-6, 15-4
Bridgewater State 13-5, 12-3	Univ. of New England 18-3, 17-3
SOUTH REGION	SOUTH REGION
Trinity (Texas) 17-2, 14-2	Trinity (Texas) 19-2, 16-2
Texas Dallas 17-2, 15-2	Texas Dallas 18-3, 16-3
Guilford 15-4, 14-4	Guilford 17-4, 16-4
McMurry 13-5, 12-3	McMurry 15-6, 14-4
Randolph-Macon 14-5, 12-2	Randolph-Macon 16-5, 14-2
DePauw 15-5, 12-4	DePauw 17-5, 13-4
Centre 15-4, 12-4	Centre 17-4, 13-4
LeTourneau 13-6, 12-4	Mississippi College 15-5, 13-5
WEST REGION	WEST REGION
Stevens Point 17-2, 16-2	Stevens Point 19-2, 18-2
St. Thomas 19-0, 18-0	St. Thomas 21-0, 20-0
Platteville 18-2, 13-2	Platteville 19-2, 14-2
Puget Sound 17-2, 14-0	Puget Sound 19-2, 16-0
Whitewater 17-3, 16-3	Whitewater 18-3, 17-3
Buena Vista 18-2, 15-2	Buena Vista 20-2, 17-2
Whitworth 15-4, 13-4	Whitworth 17-4, 15-4
Bethel 15-4, 13-4	Bethel 16-5, 14-5

[Dark Knight]

You don't coun't a teams loss against yourself when figuring OWP

Aah, that would fix that problem. Thanks.

[golden_dome]

Chris - that sounds like what the NCAA used to do! All bids were allocated to a certain number of teams from each region. That caused some MAJOR problems when teams that were obviously strong in say the Midwest were not taken while teams who were certainly less strong in the Mid-Atlantic were taken. That isn't a solution worth going back to!

Dave,
  I agree with your opinion of the Midwest and I do remember some very good teams getting left out, but let me play devil's advocate just to give a dissenting opinion. There was an uproar every year about a CCIW or UAA team left out and that was a big reason why the switch was made to the new ranking process, and it has helped the midwest which I agree with. But I think a couple other outcomes have also resulted.  

1. Some regions are getting too many teams in who are not better teams, just beneficiaries of the many DIII area teams.
2. It has further decreased the chances of teams in the deep south or west coast to host sectionals, because they are more ostracized now than they have ever been.

I realize there are regions with much more DIII history than we have in the south, and the west coast as well, but I see great teams down here every year too and teams will have a hard time coming down here and winning in the NCAA Tourney. The Howard Payne women were finally able to get to Salem and win a NC last year when they hosted a sectional. The #3 UW-Whitewater men struggled at UMHB this christmas playing a relatively close game against a 14-7 LeTourneau team, then lost to a 14-7 Mary Hardin-Baylor team.  

I think the current system was shaped by a perception that may or may not be true. It doesn't bother me to give each region equal treatment because I know there is no way to accurately know who the top 30-40 DIII teams are anyway because of how regionalized Division III is every year.

[Dave 'd-mac' McHugh]

Chris - we would then have to roll back the definition of a regional game as well, then. Right now for a Mid-Atlantic Region team based in Maryland, any game in the Mid-Atlantic Region, as well as within 200 miles (i.e. in PA and VA), AND any game played in the NCAA Geographic Zone Maryland is located in (i.e. teams in Delaware, New Jersey, Connecticut, Vermont, New Hampshire, Mass, Rhode Island, and Maine) ALL count as regional games. That is very much NOT the region they are assigned to. If you went by NCAA geographic regions (four of them), you might be closer to something, but I think the map would have to be redrawn.

I promise you that weaker teams in the Mid-Atlantic will get into this tournament over more-deserving teams in other regions under the old plan. I know this because I am in the Mid-Atlantic Region and I don't think Gettysburg is better than the top half of the CCIW. But in your plan (and the former NCAA one), they would be in while maybe only two CCIW teams get in. Plus the fact, you would have to go to brackets that allow less flexiability. You can't say that you are picking the teams via each region and then move those teams out of that region's bracket. That wouldn't seem fair. The new bracket would force far MORE conference games early on and most regional teams knocking each other out of the tournament simply to get four teams from four different regions to the championship weekend.

Sure, the extreme west part of the West Region isn't going to get much travel help from the NCAA in the current format, but I don't think it will change in your format. If anything, more WIAC teams might make the tourney since they are in the West Region and tend to rank higher then the rest of the teams in the region. That would end up giving you the same problem you have now.

The NCAA is closer now to a better solution in picking teams, but they aren't on the actual solution. We at least have a bit more of a "national" region record then we did just a few years ago. However, they probably aren't going to go to every game counting anytime soon, because it would handicap too many teams that can't afford to fly around the country playing games to make them look better. They may be able to come up with a better formula. Heck, just a short time ago they only based things on an OWP... at least they have added the OOWP. I am sure more tinkering will happen in the future.