Great Lakes Region

[KnightSlappy]

I hope this illustrates why this is bad:

Playing Calvin on the road (16-1) and Finlandia at home (2-8) gives you a tougher OWP (.748 OWP) than

Playing Calvin on the road (16-1) and Alma at home (6-13) (.690 OWP).

Old method gives the first set a .663 OWP and the second a .707 OWP.

Or more extreme, replace Alma with Ohio Northern (9-11) and it's still not as high of an OWP as vs. Finlandia.

[ziggy]

I hope this illustrates why this is bad:

Playing Calvin on the road (16-1) and Finlandia at home (2-8) gives you a tougher OWP (.748 OWP) than

Playing Calvin on the road (16-1) and Alma at home (6-13) (.690 OWP).

Old method gives the first set a .663 OWP and the second a .707 OWP.

And the problem of the handful of DII provisionals will fade and we'll be left with that...

[wally_wabash]

Isn't the easier "fix" to this calculus to stop worrying about what is and what isn't regional?  Count 'em all the same and everybody winds up with the same number of games, plus or minus a neglible amount. 

As for schools like Calvin and Hope and others that either through geographic isolation or a "dead" period in the schedule when the rest of civilazation is playing conference games have to take on an increased number of out-of-division games, shouldn't this be precisely the reason why we have human beings on a selection committee; to be able to find the spots where the math doesn't necessarily paint a complete or representative picture and then adjust accordingly?  If the committee doesn't have that latitude, then we could just program an excel sheet to select and seed the teams based on the SOS math and RRO results and h2h's. 

[Dave 'd-mac' McHugh]

Well next year all Division III games will count as long as a team plays 70% of their games in region... so that is coming. And yes... there are humans who weigh the criteria that isn't just raw data driven and parse the numbers and decipher them accordingly... that is why none of the criteria has any weight over the other.

[ziggy]

Isn't the easier "fix" to this calculus to stop worrying about what is and what isn't regional?  Count 'em all the same and everybody winds up with the same number of games, plus or minus a neglible amount. 

As for schools like Calvin and Hope and others that either through geographic isolation or a "dead" period in the schedule when the rest of civilazation is playing conference games have to take on an increased number of out-of-division games, shouldn't this be precisely the reason why we have human beings on a selection committee; to be able to find the spots where the math doesn't necessarily paint a complete or representative picture and then adjust accordingly?  If the committee doesn't have that latitude, then we could just program an excel sheet to select and seed the teams based on the SOS math and RRO results and h2h's.

With an expected appearance in the MIAA tournament championship game, Calvin will have participated in 21 in-region games. They are hardly an example for why the calculation needed to change.

[Pat Coleman]

This "change" is really the calculation coming back to the way it used to be. It was just written wrong in the book for several years.

[wally_wabash]

ziggy- Somebody mentioned earlier that the MIAA runs into a stretch of the season where non-league D3 games are tough to come by because they don't start conference play until later than everybody else...that's why I mentioned them.  Certainly they do have more games this year than they've had in years prior, which is great to see.  In any case, I'm not picking on Calvin or Hope or the MIAA.  The regionality rules are ridiculous, especially for the MIAA. 

Dave- Why put a 70% in-region prerequisite for counting your games?  Shouldn't we want to count every game we possibly can? 

[KnightSlappy]

This "change" is really the calculation coming back to the way it used to be. It was just written wrong in the book for several years.

OWP has always meant an average of the opponents winning percentage for as long it's been a term. The D3 committee has maybe not always calculated it that way, but it's always the way it's explained when talking about OWP as a component of RPI.

[Pat Coleman]

That's odd, because our numbers matched the NCAA's (barring disputed data), and Patrick Abegg's numbers matched ours -- and that is the way we calculated it before the men's committee threw home/away into the calculation.

[Dave 'd-mac' McHugh]

Dave- Why put a 70% in-region prerequisite for counting your games?  Shouldn't we want to count every game we possibly can?

Because simply, school presidents, who run everything in Division III, still want a regional aspect of things in Division III. They have allowed more and more national thinking, but still want a regional focus. As for the number... that isn't based on just basketball, but rather all sports... so I am sure some careful math was done to find the right number that would work with all sports.

Remember, there once was a time that the regional rankings determined the exact number of teams from each region that would make the tournament. So if the East Region had 6 teams to be ranked and the Northeast had 13 ranked... the East would have those 6 teams in and the Northeast would have 13 whether the teams near the bottom deserved it or not. We have steadily moved away from that VERY regional focus to one that is certainly more national... but unless there is a major shift in thinking, the presidents will never allow the regional aspect of things to go away (thus why the regional realignment I have heard tons of talk about and even chatted on Hoopsville with those at the NCAA about has some major challenges - among other things presidents want their conferences to all be in the same region, except the UAA).

[KnightSlappy]

That's odd, because our numbers matched the NCAA's (barring disputed data), and Patrick Abegg's numbers matched ours -- and that is the way we calculated it before the men's committee threw home/away into the calculation.

Wikipedia being the definitive source on everything.

The OWP is calculated by taking the average of the WP's for each of the team's opponents with the requirement that all games against the team in question are removed from the calculation.

http://en.wikipedia.org/wiki/Ratings_Percentage_Index

[ziggy]

That's odd, because our numbers matched the NCAA's (barring disputed data), and Patrick Abegg's numbers matched ours -- and that is the way we calculated it before the men's committee threw home/away into the calculation.

Wikipedia being the definitive source on everything.

The OWP is calculated by taking the average of the WP's for each of the team's opponents with the requirement that all games against the team in question are removed from the calculation.

http://en.wikipedia.org/wiki/Ratings_Percentage_Index

You mean to tell me the NCAA might say one thing and mean another? Shocked, I am, shocked!

[KnightSlappy]

That's odd, because our numbers matched the NCAA's (barring disputed data), and Patrick Abegg's numbers matched ours -- and that is the way we calculated it before the men's committee threw home/away into the calculation.

Wikipedia being the definitive source on everything.

The OWP is calculated by taking the average of the WP's for each of the team's opponents with the requirement that all games against the team in question are removed from the calculation.

http://en.wikipedia.org/wiki/Ratings_Percentage_Index

You mean to tell me the NCAA might say one thing and mean another? Shocked, I am, shocked!

They still claim they're ranking 6.5% of the number of teams in each region too.

[sac]

They still claim they're ranking 6.5% of the number of teams in each region too.

It would be tough to crack the regional rankings of 3 teams.

[wally_wabash]

Dave- Why put a 70% in-region prerequisite for counting your games?  Shouldn't we want to count every game we possibly can?

Because simply, school presidents, who run everything in Division III, still want a regional aspect of things in Division III.

I've mentioned it before, but it bears repeating here...where is the evidence that once they take the shackles of regionality out of the equation that teams from the northeast or mid-atlantic are going to play all of their non-league games in Texas and California and not at any of the scores of schools that are within easy driving distance?  Where are all of the Division III presidents that are going to spend the dollars to ship their basketball team all over the country for single non-league games?  It just isn't going to happen.  Regional play is going to happen without anybody forcing it to. 

They have allowed more and more national thinking, but still want a regional focus. As for the number... that isn't based on just basketball, but rather all sports... so I am sure some careful math was done to find the right number that would work with all sports.

Oy.  You can't have a one size fits all set of rules for every NCAA championship.  Football is different than basketball which is different than soccer which is different than baseball which is different than swimming which is different than...there's nothing un-D3 about recognizing that each championship is unique and crafting championship selection criteria that best serve each championship.  Unless they're not willing to put in that effort, which is entirely possible.  And sad.