Author Topic: Final Reg'l Poll and OWP and OOWP  (Read 5448 times)

coachmilburn

  • Guest
Final Reg'l Poll and OWP and OOWP
« on: May 11, 2009, 06:51:28 pm »
Why does not the NCAA release that information ?  Makes you think they are hiding something.  The OWP and OOWP should be released along with the regional rankings, is that asking to much?

Offline Pat Coleman

  • D3sports.com Guru
  • Administrator
  • All-American
  • *****
  • Posts: 36806
  • Karma: +4443/-2184
  • Check the front page or FAQs before you ask.
    • View Profile
    • D3sports.com
Re: Final Reg'l Poll and OWP and OOWP
« Reply #1 on: May 11, 2009, 07:05:56 pm »
Division III schools tried to legislate that OWP and OOWP be open for everyone to see. Currently, only committee members can see it -- coaches cannot even see their own stats.

The NCAA main office really pressured people into withdrawing their legislation, claiming it would cost too much money to make it available to everyone.

It does seem they have something to hide. We have seen in basketball that the NCAA's data is often inaccurate. Perhaps they don't want people to know exactly how imperfect the data is that they're using to make selections and seedings.
Publisher. Questions? Check our FAQ for D3f, D3h.
Let's discuss (sports) in a positive way, sometimes kidding each other with no disrespect.

Offline d3baseballnut

  • Starter
  • ***
  • Posts: 661
  • Karma: +35/-27
    • View Profile
Re: Final Reg'l Poll and OWP and OOWP
« Reply #2 on: May 11, 2009, 07:10:19 pm »
Division III schools tried to legislate that OWP and OOWP be open for everyone to see. Currently, only committee members can see it -- coaches cannot even see their own stats.

The NCAA main office really pressured people into withdrawing their legislation, claiming it would cost too much money to make it available to everyone.

It does seem they have something to hide. We have seen in basketball that the NCAA's data is often inaccurate. Perhaps they don't want people to know exactly how imperfect the data is that they're using to make selections and seedings.

OWP is inaccurate and is factoring way too much into decision-making.

It is one thing for it to be a secondary criteria, but it was not functioning like that for this tournament. It was primary. There is no other way St. Norbert gets in. Its absurd. OWP is stupid. You can play a great team with their #4 pitcher and on that day that team is not really as good as their OWP shows.

WHat i dont understand is howBeloit was going to need a pool A, while USN got a pool C. Beloit would not have gotten a pool C. But they played 90% the same teams, and Beloit had a better record. Plus they were 4-2 vs. USN. If you are telling me that the committee put them in the tournament bc they scheduled doubleheaders with the WIAC, i think that is dumb. Head to head should matter more. THe only reason we use the other criteria is bc we dont have head to head. BUT WE HAVE IT....Beloit is better.
« Last Edit: May 11, 2009, 07:14:32 pm by d3baseballnut »

Offline CrashDavisD3

  • All-American
  • ******
  • Posts: 2494
  • Karma: +289/-78
  • If winning isn't everything,why do they keep score
    • View Profile
    • Crash Davis in Action
Re: Final Reg'l Poll and OWP and OOWP
« Reply #3 on: May 11, 2009, 07:19:39 pm »
I agree that a data should be open and transparent from the NCAA in their selection process. Many times it is illogical the choices that are made.

Cost issue is really somthing to hide behind since the use of electronic communications drives down the cost for providing information,
This... is a simple game. You throw the ball. You hit the ball. You catch the ball.  "There are three types of baseball players: those who make things happen, those who watch it happen, and those who wonder what happened."
Crash Davis Bio - http://www.baseballamerica.com/today/minors/crash0908.html

Offline RSSmith

  • Starter
  • ***
  • Posts: 657
  • Karma: +30/-5
    • View Profile
Re: Final Reg'l Poll and OWP and OOWP
« Reply #4 on: May 11, 2009, 07:29:26 pm »
Just a wild guess on my part that OOWP numbers don't do a lot of discriminating.  By the time you get one generation out, everybody is playing the same pool of opponents.  It is a lot like inbreeding.  With everybody doing the same everybody else, everybody starts to look alike.
A walk is never as good as a hit!

golden_dome

  • Guest
Re: Final Reg'l Poll and OWP and OOWP
« Reply #5 on: May 11, 2009, 08:38:15 pm »
Division III schools tried to legislate that OWP and OOWP be open for everyone to see. Currently, only committee members can see it -- coaches cannot even see their own stats.

The NCAA main office really pressured people into withdrawing their legislation, claiming it would cost too much money to make it available to everyone.

It does seem they have something to hide. We have seen in basketball that the NCAA's data is often inaccurate. Perhaps they don't want people to know exactly how imperfect the data is that they're using to make selections and seedings.

I don't know all the specifics or what it would take on the NCAA's part to make those numbers public, maybe it's valid. But it's just very hard to believe the cost factor. The system is already set up for the committee members to log in and get their information for the conference calls, so any school could log in if they had the information.

Just to add to the discussion about the importance of OWP and OOWP. In basketball, Patrick Abegg produces numbers that have been very accurate the last few years based on a standard RPI model.

Basically, I think his formula weights 25% for winning percentage, 50% for OWP and 25% for OOWP.

Basically it would break down like this.

A team with a .600 OWP and OOWP with a 15-10 record would produce the same RPI number as a 20-5 team with a .500 OWP and OOWP. I'll go back and check but I'm pretty sure that is what I came up with when looking over his numbers this year.

Offline Mr. Ypsi

  • All-American
  • ******
  • Posts: 21629
  • Karma: +3224/-3196
    • View Profile
Re: Final Reg'l Poll and OWP and OOWP
« Reply #6 on: May 11, 2009, 08:52:09 pm »
Just a wild guess on my part that OOWP numbers don't do a lot of discriminating.  By the time you get one generation out, everybody is playing the same pool of opponents.  It is a lot like inbreeding.  With everybody doing the same everybody else, everybody starts to look alike.

In isolated regions, that may be true.  Otherwise, we've found OWP and OOWP to be quite illuminating on the basketball boards.  Sometimes OOWP reveals OWP to be based on beating a bunch of mediocre teams who beat up on really bad teams.

It would be great if baseball could start getting the transparency of numbers that basketball has.  There's more games, the unpredictiblility of rainouts, etc., but perhaps in a couple of years? ;)

Offline Ralph Turner

  • Hall of Fame
  • All-American
  • ********
  • Posts: 27872
  • Karma: +1738/-376
  • Hall of Famer
    • View Profile
Re: Final Reg'l Poll and OWP and OOWP
« Reply #7 on: May 11, 2009, 08:55:03 pm »
Here are the numbers of which Chris Brooks and Mr Ypsi speak.

Patrick Abegg  (R.I.P.) ran some numbers for me using his calculations last spring.

Patrick compiled the OWP  and OOWP for D3hoops.com.

I looked at teams that might have been left on the table when the last Pool C bid (#18) was given.

I wanted to see what the impact of changing 1-2 games would be in the rank order of OWP OOWP in his table.  Most of us believe that these are good numbers.

THE BACKGROUND:

McMurry Regionally ranked in the South Region and winner of the ASC West.  Lost in the semifinals of an 8-team  tourney.  (ASC is a 15 team conference.)  The ASC is very isolated such that OOWP's hover around .470-.530  In the Regular season, McMurry lost to North Central from the Midwest Region and the CCIW.  This was an in-region game on a neutral floor by 3 points, a representative game.

North Central is in the very competitive CCIW.  They and Augustana are in a conference with very good OWP and OOWP's.  They contend for multiple at large bids often.   These teams were considered neck and neck in the Pool C considerations.

McDaniel is team from the Centennial Conference in the Mid-Atlantic Region.  The OWP's and OOWP's can be very good in the conferences.  There are numerous conferences from which to choose in-region foes.

Patrick changes one and two outcomes on these teams to show what that impact is.

Please look at it for the perspective of what the committee is dealing with.

http://www.d3boards.com/index.php?topic=4232.2384


Here is Patrick's final table for Pool C teams in the 2009 season.

http://www.d3boards.com/index.php?topic=4232.2361

« Last Edit: May 11, 2009, 08:58:19 pm by Ralph Turner »

OshDude

  • Guest
Re: Final Reg'l Poll and OWP and OOWP
« Reply #8 on: May 12, 2009, 02:27:58 am »
Division III schools tried to legislate that OWP and OOWP be open for everyone to see. Currently, only committee members can see it -- coaches cannot even see their own stats.

The NCAA main office really pressured people into withdrawing their legislation, claiming it would cost too much money to make it available to everyone.

It does seem they have something to hide. We have seen in basketball that the NCAA's data is often inaccurate. Perhaps they don't want people to know exactly how imperfect the data is that they're using to make selections and seedings.

OWP is inaccurate and is factoring way too much into decision-making.

It is one thing for it to be a secondary criteria, but it was not functioning like that for this tournament. It was primary. There is no other way St. Norbert gets in. Its absurd. OWP is stupid. You can play a great team with their #4 pitcher and on that day that team is not really as good as their OWP shows.

WHat i dont understand is howBeloit was going to need a pool A, while USN got a pool C. Beloit would not have gotten a pool C. But they played 90% the same teams, and Beloit had a better record. Plus they were 4-2 vs. USN. If you are telling me that the committee put them in the tournament bc they scheduled doubleheaders with the WIAC, i think that is dumb. Head to head should matter more. THe only reason we use the other criteria is bc we dont have head to head. BUT WE HAVE IT....Beloit is better.
SNC's OWP was .543. Beloit's was .453. That's substantial.

Offline kune14

  • Junior Varsity
  • *
  • Posts: 2
  • Karma: +0/-0
    • View Profile
Re: Final Reg'l Poll and OWP and OOWP
« Reply #9 on: May 12, 2009, 11:48:46 am »
Part of the money issue that the NCAA folks said prevents them from making the numbers public is more about the IT side of things then anything else.  Not being an IT person I do not know how this all weighs out, but the word is that they would need to substantially increase "things" in order to enable one and all to access the information without crashing or jeopardiizing the system.  This was the word that got out there, but folks also feel that there would also be a bunch of issues that would need to be dealt with if everything was publicly available. 

Personally I think it might reduce some of the already rampant scuttlebutt as it would raise understanding since many coaches do not even understand the process.  As someone who has served on these committees before I can tell you the criteria has never been prioritized on the committees I was on such that W/L weighed more than OWP which weighed more than OOWP.  All the factors were looked at together and often ranking decisions were made based on one team "beating" another team in 2 of 3 categories; or 3 of 5 categories or whatever. 

The system, while not fail-proof of course, was worked on by committee members, in other words NOT just folks who work at the NCAA office, but fellow coaches, ADs, etc; for several years and tweaked and tweaked in an effort to make it as fair as possible.  The bottom line of course for those in conferences with an AQ, is take care of business and get your AQ and you'll be in for sure!

Offline d3baseballnut

  • Starter
  • ***
  • Posts: 661
  • Karma: +35/-27
    • View Profile
Re: Final Reg'l Poll and OWP and OOWP
« Reply #10 on: May 12, 2009, 12:17:54 pm »
Division III schools tried to legislate that OWP and OOWP be open for everyone to see. Currently, only committee members can see it -- coaches cannot even see their own stats.

The NCAA main office really pressured people into withdrawing their legislation, claiming it would cost too much money to make it available to everyone.

It does seem they have something to hide. We have seen in basketball that the NCAA's data is often inaccurate. Perhaps they don't want people to know exactly how imperfect the data is that they're using to make selections and seedings.

OWP is inaccurate and is factoring way too much into decision-making.

It is one thing for it to be a secondary criteria, but it was not functioning like that for this tournament. It was primary. There is no other way St. Norbert gets in. Its absurd. OWP is stupid. You can play a great team with their #4 pitcher and on that day that team is not really as good as their OWP shows.

WHat i dont understand is howBeloit was going to need a pool A, while USN got a pool C. Beloit would not have gotten a pool C. But they played 90% the same teams, and Beloit had a better record. Plus they were 4-2 vs. USN. If you are telling me that the committee put them in the tournament bc they scheduled doubleheaders with the WIAC, i think that is dumb. Head to head should matter more. THe only reason we use the other criteria is bc we dont have head to head. BUT WE HAVE IT....Beloit is better.
SNC's OWP was .543. Beloit's was .453. That's substantial.

What was miss college's? I bet it was higher than St. Norberts...

Offline Jim Dixon

  • D3baseball.com Guru
  • Global Moderator
  • All-Region
  • *****
  • Posts: 1730
  • Karma: +137/-3
  • Managing Editor, D3baseball.com
    • View Profile
Re: Final Reg'l Poll and OWP and OOWP
« Reply #11 on: May 12, 2009, 12:36:44 pm »
Just a wild guess on my part that OOWP numbers don't do a lot of discriminating.  By the time you get one generation out, everybody is playing the same pool of opponents.  It is a lot like inbreeding.  With everybody doing the same everybody else, everybody starts to look alike.

In isolated regions, that may be true.  Otherwise, we've found OWP and OOWP to be quite illuminating on the basketball boards.  Sometimes OOWP reveals OWP to be based on beating a bunch of mediocre teams who beat up on really bad teams.

It would be great if baseball could start getting the transparency of numbers that basketball has.  There's more games, the unpredictiblility of rainouts, etc., but perhaps in a couple of years? ;)


Actually if conference reported scores to PrestoSports I expect that these numbers could be generated.  The number of conferences that use PrestoSports is about 60%, maybe a little more.  The work to add those 35-40% is too much for D3baseball.com to take on (35% of 400 teams with 30 games/team is over 4,000 games and a lot of work).

Offline kune14

  • Junior Varsity
  • *
  • Posts: 2
  • Karma: +0/-0
    • View Profile
Re: Final Reg'l Poll and OWP and OOWP
« Reply #12 on: May 12, 2009, 01:05:16 pm »
Actually I think your guesstimate of 60% of conferences using presto is a bit high and eventhough the conf office might use them that doesn't mean every school within that conf uses them.

Offline Pat Coleman

  • D3sports.com Guru
  • Administrator
  • All-American
  • *****
  • Posts: 36806
  • Karma: +4443/-2184
  • Check the front page or FAQs before you ask.
    • View Profile
    • D3sports.com
Re: Final Reg'l Poll and OWP and OOWP
« Reply #13 on: May 12, 2009, 01:18:41 pm »
Part of the money issue that the NCAA folks said prevents them from making the numbers public is more about the IT side of things then anything else.  Not being an IT person I do not know how this all weighs out, but the word is that they would need to substantially increase "things" in order to enable one and all to access the information without crashing or jeopardiizing the system. 

Yeah. This is a ploy. They can give me the contract and I'll bring itin for a quarter of the cost. :)
Publisher. Questions? Check our FAQ for D3f, D3h.
Let's discuss (sports) in a positive way, sometimes kidding each other with no disrespect.

Offline Pat Coleman

  • D3sports.com Guru
  • Administrator
  • All-American
  • *****
  • Posts: 36806
  • Karma: +4443/-2184
  • Check the front page or FAQs before you ask.
    • View Profile
    • D3sports.com
Re: Final Reg'l Poll and OWP and OOWP
« Reply #14 on: May 12, 2009, 01:21:33 pm »
Actually I think your guesstimate of 60% of conferences using presto is a bit high and eventhough the conf office might use them that doesn't mean every school within that conf uses them.

When you take into account the non-conference games played by leagues outside of Presto we get about 60% of the baseball schedule into the database.

If a conference is using Presto -- and 20 of them are so far -- that means the entire schedule and results are supposed to be in the system, for every school. Otherwise the standings on the conference's site aren't very complete. :)
Publisher. Questions? Check our FAQ for D3f, D3h.
Let's discuss (sports) in a positive way, sometimes kidding each other with no disrespect.