Forum Search:
RSD No Spam
rec.sport.disc without the spam


Home » RSD » RSD Posts » South Regionals...
South Regionals... [message #18731] Mon, 13 April 2009 18:56 Go to next message
Muff
Messages: 9
Registered: October 2008
Junior Member
Lets look at seeding:

The following was proposed by Michael Schulz of LSU on the bama
section blog (http://bamasecs.wordpress.com/) and looks to be pretty
good

1. Texas (H-H Win over Kansas)
2. Kansas (Best Regular Season/RRI)
3. UNT (2nd in TX Section)
4. Texas State (3rd in TX + H-H over Arkansas)
5. Arkansas (2nd in OZ + H-H over A&M)
6. Texas A&M (4th in TX + H-H over LSU)
7. LSU (1st in BAMA + H-H W&L to Arkansas)
8. Wash U (3rd in OZ)
9. Truman (4th in OZ)
10. Missourri (5th in OZ)
11. Ole Miss - These guys could be 9th but Truman’s RRI is way better
but got 4th in the section so Truman has to be behind Wash U.
Missourri and Ole Miss could go either way but to avoid rematches.
12. Rice - (5th TX)
13. Harding - (6th OZ) - Rice and Harding could switch.
14. Miss St. - (3rd BAMA) Harding has a better record against common
opponents and better RRI.
15. Houston - (6th TX)
16. UTA - (7th TX)

This gives us:

Texas
UTA

Wash U
Truman

Arkansas
Rice

Tx St
Harding

UNT
Miss St

Tx AM
Ole Miss

LSU
Missouri

Kansas
Houston

This leaves only 2 inter-sectional matchups Tx-UTa and Wash U-Truman
in the first round which is pretty good considering 13 teams come from
2 sections.
Re: South Regionals... [message #18739 is a reply to message #18731] Mon, 13 April 2009 19:21 Go to previous messageGo to next message
aclement
Messages: 62
Registered: October 2008
Member
On Apr 13, 8:56 pm, Muff <john.as...@gmail.com> wrote:
> Lets look at seeding:
>
> The following was proposed by Michael Schulz of LSU on the bama
> section blog (http://bamasecs.wordpress.com/) and looks to be pretty
> good
>
> 1. Texas (H-H Win over Kansas)
> 2. Kansas (Best Regular Season/RRI)
> 3. UNT (2nd in TX Section)
> 4. Texas State (3rd in TX + H-H over Arkansas)
> 5. Arkansas (2nd in OZ + H-H over A&M)
> 6. Texas A&M (4th in TX + H-H over LSU)
> 7. LSU (1st in BAMA + H-H W&L to Arkansas)
> 8. Wash U (3rd in OZ)
> 9. Truman (4th in OZ)
> 10. Missourri (5th in OZ)
> 11. Ole Miss - These guys could be 9th but Truman’s RRI is way better
> but got 4th in the section so Truman has to be behind Wash U.
> Missourri and Ole Miss could go either way but to avoid rematches.
> 12. Rice - (5th TX)
> 13. Harding - (6th OZ) - Rice and Harding could switch.
> 14. Miss St. - (3rd BAMA) Harding has a better record against common
> opponents and better RRI.
> 15. Houston - (6th TX)
> 16. UTA - (7th TX)
>
> This gives us:
>
> Texas
> UTA
>
> Wash U
> Truman
>
> Arkansas
> Rice
>
> Tx St
> Harding
>
> UNT
> Miss St
>
> Tx AM
> Ole Miss
>
> LSU
> Missouri
>
> Kansas
> Houston
>
> This leaves only 2 inter-sectional matchups Tx-UTa and Wash U-Truman
> in the first round which is pretty good considering 13 teams come from
> 2 sections.

You started it.

Before you get seeds from me, you get a statement about what seedings
are good for:

You can either view seedings as predictive of the results of the
upcoming tournament or reflective of the results of the previous
played games. Given the UPA requirement that sectional (regional)
finish must be maintained, seeding at regionals (nationals) must be
reflective rather than predictive.

Before you get seeds, you get to read how I got the seedings -- for
me, the process is more important than the outcome (gee, have I said
that before?). The following explanation captures the key steps and
tries to explain the rationale for subtle parts of the process.

(1) consider the highest remaining team from each section
(2) compare these teams to eachother in a mini tournament (similar to
what is done on the UPA SRT tool in the team grid)
the points for consideration are, in order:
(a) head to head record
(b) record against common opponents at regionals that have not
already been seeded
(c) record against common opponents not at regionals
(d) RRI, last years results, flip a coin -- at this point
there really isnt much useful information. Get teams to play more
games!
(3) after playing out this mini tournament, see if any team has a
unique best record. If that team exists, they get the next seed. If
that team does not exist, then there are two possiblities
(a) if every matchup was decided with the same criteria (i.e.
head to head) then consider the next criteria (common opponents) to
make the decision
(b) if matchups were decided using different criteria, then
nullify all of the lowest criteria used to make decisions (i.e. 1
matchup with head to head, 2 with rri, then the rri matchups become
draws)
(4) repeat this process until all teams are seeded.

There are a couple of important clarifications when applying this
algorithm
(1) When comparing common opponents, compare the winning percentage
against each individual team (i.e. aTm goes 2-1 against Davidson while
LSU goes 1-0 against Davidson, then LSU would 'win' that matchup based
on that common opponent)

(2) Placing higher than another team at sectionals negates all
previous victories/losses (i.e. if texas went 0-3 against aTm this
year, then because texas placed higher than atm at sectionals, that
record would be interpreted as 1-0).

(3) do not consider common opponents that have already been seeded.
Basically, using Arkansas's victory over Texas as an argument to seed
Arkansas ahead of aTm is also an argument to seed arkansas above
Texas. If texas has already been seeded, then we the argument is no
longer compelling (and we have already decided that Arkansas is below
texas, so the whole transitivity relationship begins to fall apart).

(4) Consider common opponents at regionals separately from common
opponents not at regionals for 2 reasons. First, it gives us another
layer of 'tie breaks'. Second, a common opponent at the tournament
will be directly impacted by the seedings so results with that
opponent should weigh more heavily than results against opponents not
at the tournament.

And now you get the seedings, with the short explanation for why each
seed was selected.

(1) Texas
o Texas over Kansas head to head
o Texas over LSU based on common opponents 2-1 (GA Tech, atm in
favor of texas, Arkansas in favor of LSU)
- note this is tied 1-1 for teams at regionals, and 1-0 for
teams not at regionals
(2) Kansas
o Kansas over UNT based on common opponents UCSD, arkansas
- again, arkansas at regionals is enough to stop consideration.
UCSD is just gravy
o Kansas over LSU based on common opponents OK State, arkansas,
notredame, michigan state
- see the comment above
(3) Arkansas
o arkansas over UNT based on head to head
o arkansas over LSU based on common opponents Missouri state and
atm
- atm is at regionals, so missouri state is just extra
(4) UNT
o LSU over UNT based on common opponent Davidson
- not at regionals
o Washington over LSU based on common opponents 3-1 (Notre
Dame, Michigan St, Missouri state in favor of wash, davidson in favor
of LSU).
- no teams at regionals in the common opponent list
o UNT over washington based on common opponent Tx State.
- tx state at regionals, carries more weight than other options

This is the first tricky one. LSU and UNT have a common opponent
arkansas, but Arkansas has already been seeded so should be ignored.
LSU over UNT is based on team not at regionals, Washington over LSU is
based on teams not at regionals. UNT over washington is based on a
team at regionals. So the UNT edge remains at the 'top' and we break
one of the latter two. End result, UNT as the 4 seed.

(5) tx state
o Washington over LSU based on common opponents (3-1 as above)
o Tx State over washington based on head to head
o tx state over lsu based on common opponents atm at regionals,
common opponents Ohio St, OK St not at regionals

(6) washington
o washington over lsu as above
o washington over atm on common opponents 2-1 (OK, Missouri in
favor of washington, davidson for atm)
- again, Missouri at regionals is enough at 1-0. the other two
are a wash
o atm over lsu head to head

(7) truman state
o truman over lsu based on common opponents emory, missour state,
ok state
- none at regionals
o truman over atm based on common opponent OK state, missouri
- missouri at regionals stops discussion
o atm over lsu head to head

(8) atm
o atm over lsu head to head
o missouri over atm head to head
o lsu over missouri head to head
====
o atm over lsu on rri (common opponents are identical records)
o atm over missouri on common opponent grinnel
- not at regionals
o missouri over lsu on common opponents ok state, missouri state
- not at regionals

This one is also tricky given the chain of head to heads. at the end
of the day, atm has claims over the other two teams on weaker
conditions so they get the seed.

(9) lsu
o lsu over missouri head to head
o lsu over rice common opponents 2-1 (new mexico, davidson for lsu,
ok state for rice)
- none at regionals
o missouri over rice on RRI

(10) missouri
o missouri over rice on rri
o missouri over mississippi on rri
o mississippi over rice on rri

(11) mississippi
o mississippi over rice on rri
o rice over harding on rri
o mississipppi over harding head to head

(12) rice
o rice over harding on rri
o harding over missippi state on rri
o rice over miss. state on rri

(13) ms state
o ms state over harding on rri
o harding over houston on rri

(14) harding
o harding over houston on rri

(15) houston

(16) UT-Arlington

For the TL;DR folks:

1) Texas
2) Kansas
3) Arkansas
4) UNT
5) tx state
6) washington
7) truman
8) atm
9) lsu
10) missouri
11) mississippi
12) rice
13) ms state
14) harding
15) houston
16) ut-arlington

These seeds give an unfortunate number of sectionals rematches and put
the texas and ozarks sections on opposite sides of the bracket
(assuming a 16 team bracket). In order to resolve those issues, I
advocate swapping the 3/4 seeds, the 7/8 seeds, and the 12/13 seeds.
This leaves only one inter-sectional matchup between two teams from
the section that has 7 teams at regionals (texas - UT Arlington).


ps. Mike, is this a good start on reaching 100 by Tuesday?
Re: South Regionals... [message #18743 is a reply to message #18739] Mon, 13 April 2009 19:52 Go to previous messageGo to next message
ajax.xaja
Messages: 40
Registered: October 2008
Member
Allen, what are you doing posting a concise method for seeding? We
will never get to 100 posts using logic.

http://www.upa.org/scores/scores.cgi?div=127&page=3& tourn=6418

-ajax
Re: South Regionals... [message #18745 is a reply to message #18743] Mon, 13 April 2009 20:12 Go to previous messageGo to next message
Torre
Messages: 226
Registered: September 2008
Senior Member
obviously it should've been:

1. Auburn
2-16. everyone else.
Re: South Regionals... [message #18748 is a reply to message #18745] Mon, 13 April 2009 20:24 Go to previous messageGo to next message
Mike
Messages: 23
Registered: September 2008
Junior Member
As long as Missouri is their first game again.
Re: South Regionals... [message #18751 is a reply to message #18743] Mon, 13 April 2009 20:34 Go to previous messageGo to next message
aclement
Messages: 62
Registered: October 2008
Member
On Apr 13, 9:52 pm, ajax.x...@gmail.com wrote:
> Allen, what are you doing posting a concise method for seeding?  We
> will never get to 100 posts using logic.
>
> http://www.upa.org/scores/scores.cgi?div=127&page=3& tourn=6418
>
> -ajax

Damnit Ajax. You caught me!
Re: South Regionals... [message #18753 is a reply to message #18739] Mon, 13 April 2009 21:03 Go to previous messageGo to next message
Alex DW
Messages: 21
Registered: April 2009
Junior Member
I'm glad that you were able to legitimize placing the 1st seed from
the Bama section in the bottom half of the tournament. At least that
should be uncontested from here on out.

-DW
Re: South Regionals... [message #18754 is a reply to message #18751] Mon, 13 April 2009 21:29 Go to previous messageGo to next message
Honna
Messages: 5
Registered: April 2009
Junior Member
On Apr 13, 10:34 pm, Allen <aclem...@gmail.com> wrote:
> On Apr 13, 9:52 pm, ajax.x...@gmail.com wrote:
>
> > Allen, what are you doing posting a concise method for seeding?  We
> > will never get to 100 posts using logic.
>
> > http://www.upa.org/scores/scores.cgi?div=127&page=3& tourn=6418
>
> > -ajax
>
> Damnit Ajax.  You caught me!

Bravo C-bag. Now go finish your degree.
Re: South Regionals... [message #18755 is a reply to message #18753] Mon, 13 April 2009 21:33 Go to previous messageGo to next message
Zeke
Messages: 20
Registered: November 2008
Junior Member
On Apr 13, 11:03 pm, Alex DW <kadrl...@gmail.com> wrote:
> I'm glad that you were able to legitimize placing the 1st seed from
> the Bama section in the bottom half of the tournament.  At least that
> should be uncontested from here on out.
>
> -DW

Totally agreed. Next year I say we petition to remove the 'Bama
section from Regionals. They are just hogging up 3 whole spots. Now
what about that 17th seed.

- Zeke
Re: South Regionals... [message #18763 is a reply to message #18739] Mon, 13 April 2009 23:45 Go to previous messageGo to next message
Muff
Messages: 9
Registered: October 2008
Junior Member
On Apr 13, 9:21 pm, Allen <aclem...@gmail.com> wrote:
> On Apr 13, 8:56 pm, Muff <john.as...@gmail.com> wrote:
>
>
>
> > Lets look at seeding:
>
> > The following was proposed by Michael Schulz of LSU on the bama
> > section blog (http://bamasecs.wordpress.com/) and looks to be pretty
> > good
>
> > 1. Texas (H-H Win over Kansas)
> > 2. Kansas (Best Regular Season/RRI)
> > 3. UNT (2nd in TX Section)
> > 4. Texas State (3rd in TX + H-H over Arkansas)
> > 5. Arkansas (2nd in OZ + H-H over A&M)
> > 6. Texas A&M (4th in TX + H-H over LSU)
> > 7. LSU (1st in BAMA + H-H W&L to Arkansas)
> > 8. Wash U (3rd in OZ)
> > 9. Truman (4th in OZ)
> > 10. Missourri (5th in OZ)
> > 11. Ole Miss - These guys could be 9th but Truman’s RRI is way better
> > but got 4th in the section so Truman has to be behind Wash U.
> > Missourri and Ole Miss could go either way but to avoid rematches.
> > 12. Rice - (5th TX)
> > 13. Harding - (6th OZ) - Rice and Harding could switch.
> > 14. Miss St. - (3rd BAMA) Harding has a better record against common
> > opponents and better RRI.
> > 15. Houston - (6th TX)
> > 16. UTA - (7th TX)
>
> > This gives us:
>
> > Texas
> > UTA
>
> > Wash U
> > Truman
>
> > Arkansas
> > Rice
>
> > Tx St
> > Harding
>
> > UNT
> > Miss St
>
> > Tx AM
> > Ole Miss
>
> > LSU
> > Missouri
>
> > Kansas
> > Houston
>
> > This leaves only 2 inter-sectional matchups Tx-UTa and Wash U-Truman
> > in the first round which is pretty good considering 13 teams come from
> > 2 sections.
>
> You started it.
>
> Before you get seeds from me, you get a statement about what seedings
> are good for:
>
> You can either view seedings as predictive of the results of the
> upcoming tournament or reflective of the results of the previous
> played games.  Given the UPA requirement that sectional (regional)
> finish must be maintained, seeding at regionals (nationals) must be
> reflective rather than predictive.
>
> Before you get seeds, you get to read how I got the seedings -- for
> me, the process is more important than the outcome (gee, have I said
> that before?).  The following explanation captures the key steps and
> tries to explain the rationale for subtle parts of the process.
>
> (1)  consider the highest remaining team from each section
> (2)  compare these teams to eachother in a mini tournament (similar to
> what is done on the UPA SRT tool in the team grid)
>        the points for consideration are, in order:
>        (a)  head to head record
>        (b)  record against common opponents at regionals that have not
> already been seeded
>        (c)  record against common opponents not at regionals
>        (d)  RRI, last years results, flip a coin -- at this point
> there really isnt much useful information.  Get teams to play more
> games!
> (3)  after playing out this mini tournament, see if any team has a
> unique best record.  If that team exists, they get the next seed.  If
> that team does not exist, then there are two possiblities
>        (a)  if every matchup was decided with the same criteria (i.e.
> head to head) then consider the next criteria (common opponents) to
> make the decision
>        (b)  if matchups were decided using different criteria, then
> nullify all of the lowest criteria used to make decisions (i.e. 1
> matchup with head to head, 2 with rri, then the rri matchups become
> draws)
> (4)  repeat this process until all teams are seeded.
>
> There are a couple of important clarifications when applying this
> algorithm
> (1)  When comparing common opponents, compare the winning percentage
> against each individual team (i.e. aTm goes 2-1 against Davidson while
> LSU goes 1-0 against Davidson, then LSU would 'win' that matchup based
> on that common opponent)
>
> (2)  Placing higher than another team at sectionals negates all
> previous victories/losses (i.e. if texas went 0-3 against aTm this
> year, then because texas placed higher than atm at sectionals, that
> record would be interpreted as 1-0).
>
> (3)  do not consider common opponents that have already been seeded.
> Basically, using Arkansas's victory over Texas as an argument to seed
> Arkansas ahead of aTm is also an argument to seed arkansas above
> Texas.  If texas has already been seeded, then we the argument is no
> longer compelling (and we have already decided that Arkansas is below
> texas, so the whole transitivity relationship begins to fall apart).
>
> (4)  Consider common opponents at regionals separately from common
> opponents not at regionals for 2 reasons.  First, it gives us another
> layer of 'tie breaks'. Second, a common opponent at the tournament
> will be directly impacted by the seedings so results with that
> opponent should weigh more heavily than results against opponents not
> at the tournament.
>
> And now you get the seedings, with the short explanation for why each
> seed was selected.
>
> (1)  Texas
>      o  Texas over Kansas head to head
>      o   Texas over LSU based on common opponents 2-1 (GA Tech, atm in
> favor of texas, Arkansas in favor of LSU)
>       -  note this is tied 1-1 for teams at regionals, and 1-0 for
> teams not at regionals
> (2)  Kansas
>      o  Kansas over UNT based on common opponents UCSD, arkansas
>      -  again, arkansas at regionals is enough to stop consideration.
> UCSD is just gravy
>      o Kansas over LSU based on common opponents OK State, arkansas,
> notredame, michigan state
>      - see the comment above
> (3) Arkansas
>      o arkansas over UNT based on head to head
>      o  arkansas over LSU based on common opponents Missouri state and
> atm
>      - atm is at regionals, so missouri state is just extra
> (4)  UNT
>      o  LSU over UNT based on common opponent Davidson
>     - not at regionals
>      o  Washington over LSU based on common opponents 3-1  (Notre
> Dame, Michigan St, Missouri state in favor of wash, davidson in favor
> of LSU).
>     - no teams at regionals in the common opponent list
>     o   UNT over washington based on common opponent Tx State.
>     - tx state at regionals, carries more weight than other options
>
>     This is the first tricky one.  LSU and UNT have a common opponent
> arkansas, but Arkansas has already been seeded so should be ignored.
> LSU over UNT is based on team not at regionals, Washington over LSU is
> based on teams not at regionals.  UNT over washington is based on a
> team at regionals.  So the UNT edge remains at the 'top' and we break
> one of the latter two.  End result, UNT as the 4 seed.
>
> (5)  tx state
>     o Washington over LSU based on common opponents (3-1 as above)
>     o  Tx State over washington based on head to head
>     o  tx state over lsu based on common opponents atm at regionals,
> common opponents Ohio St, OK St not at regionals
>
> (6)  washington
>     o washington over lsu as above
>     o washington over atm on common opponents 2-1 (OK, Missouri in
> favor of washington, davidson for atm)
>     - again, Missouri at regionals is enough at 1-0.  the other two
> are a wash
>     o atm over lsu head to head
>
> (7)  truman state
>    o truman over lsu based on common opponents emory, missour state,
> ok state
>     - none at regionals
>    o truman over atm based on common opponent OK state, missouri
>     - missouri at regionals stops discussion
>    o  atm over lsu head to head
>
> (8) atm
>    o atm over lsu head to head
>    o  missouri over atm head to head
>    o lsu over missouri head to head
>  ====
>    o atm over lsu on rri  (common opponents are identical records)
>    o atm over missouri on common opponent grinnel
>     - not at regionals
>    o missouri over lsu on common opponents ok state, missouri state
>     - not at regionals
>
> This one is also tricky given the chain of head to heads.  at the end
> of the day, atm has claims over the other two teams on weaker
> conditions so they get the seed.
>
> (9)  lsu
>    o lsu over missouri head to head
>    o lsu over rice common opponents 2-1 (new mexico, davidson for lsu,
> ok state for rice)
>      - none at regionals
>     o missouri over rice on RRI
>
> (10)  missouri
>    o missouri over rice on rri
>    o missouri over mississippi on rri
>    o  mississippi over rice on rri
>
> (11)  mississippi
>    o mississippi over rice on rri
>    o  rice over harding on rri
>    o mississipppi over harding head to head
>
> (12)  rice
>    o rice over harding on rri
>    o harding over missippi state on rri
>    o rice over miss. state on rri
>
> (13)  ms state
>    o ms state over harding on rri
>    o harding over houston on rri
>
> (14) harding
>    o harding over houston on rri
>
> (15)  houston
>
> (16)  UT-Arlington
>
> For the TL;DR folks:
>
> 1) Texas
> 2) Kansas
> 3) Arkansas
> 4) UNT
> 5) tx state
> 6) washington
> 7) truman
> 8) atm
> 9) lsu
> 10) missouri
> 11) mississippi
> 12) rice
> 13) ms state
> 14)  harding
> 15) houston
> 16) ut-arlington
>
> These seeds give an unfortunate number of sectionals rematches and put
> the texas and ozarks sections on opposite sides of the bracket
> (assuming a 16 team bracket). In order to resolve those issues, I
> advocate swapping the 3/4 seeds, the 7/8 seeds, and the 12/13 seeds.
> This leaves only one inter-sectional matchup between two teams from
> the section that has 7 teams at regionals (texas - UT Arlington).
>
> ps. Mike, is this a good start on reaching 100 by Tuesday?

So you make this long post with a thorough (though still arbitrary and
based on which factors you deem to be more important than others)
analysis of seedings, and then you conclude it by saying basically -
this doesn't work so switch these 6 teams around. If you just move
teams around to fit into the part of the framework not accounted for
in your system, then what is the point of the system? Despite the
hard work put into the system you provide, since it is incomplete, are
you not in the end just placing teams as you see fit? Your omission
of one part of the whole (avoiding sectionals rematches) necessarily
changes all of the results of your system. For instance, if you were
to omit some other part of the system (sectional wins trump all other
meetings or ignoring already seeded teams) the results would be very
different.


ps. I am doing my part for "100 or Bust 2009"
Re: South Regionals... [message #18769 is a reply to message #18763] Tue, 14 April 2009 01:01 Go to previous messageGo to next message
vila
Messages: 1
Registered: April 2009
Junior Member
On Apr 14, 2:45 pm, john.as...@gmail.com wrote:
> On Apr 13, 9:21 pm, Allen <aclem...@gmail.com> wrote:
>
>
>
> > On Apr 13, 8:56 pm, Muff <john.as...@gmail.com> wrote:
>
> > > Lets look at seeding:
>
> > > The following was proposed by Michael Schulz of LSU on the bama
> > > section blog (http://bamasecs.wordpress.com/) and looks to be pretty
> > > good
>
> > > 1. Texas (H-H Win over Kansas)
> > > 2. Kansas (Best Regular Season/RRI)
> > > 3. UNT (2nd in TX Section)
> > > 4. Texas State (3rd in TX + H-H over Arkansas)
> > > 5. Arkansas (2nd in OZ + H-H over A&M)
> > > 6. Texas A&M (4th in TX + H-H over LSU)
> > > 7. LSU (1st in BAMA + H-H W&L to Arkansas)
> > > 8. Wash U (3rd in OZ)
> > > 9. Truman (4th in OZ)
> > > 10. Missourri (5th in OZ)
> > > 11. Ole Miss - These guys could be 9th but Truman’s RRI is way better
> > > but got 4th in the section so Truman has to be behind Wash U.
> > > Missourri and Ole Miss could go either way but to avoid rematches.
> > > 12. Rice - (5th TX)
> > > 13. Harding - (6th OZ) - Rice and Harding could switch.
> > > 14. Miss St. - (3rd BAMA) Harding has a better record against common
> > > opponents and better RRI.
> > > 15. Houston - (6th TX)
> > > 16. UTA - (7th TX)
>
> > > This gives us:
>
> > > Texas
> > > UTA
>
> > > Wash U
> > > Truman
>
> > > Arkansas
> > > Rice
>
> > > Tx St
> > > Harding
>
> > > UNT
> > > Miss St
>
> > > Tx AM
> > > Ole Miss
>
> > > LSU
> > > Missouri
>
> > > Kansas
> > > Houston
>
> > > This leaves only 2 inter-sectional matchups Tx-UTa and Wash U-Truman
> > > in the first round which is pretty good considering 13 teams come from
> > > 2 sections.
>
> > You started it.
>
> > Before you get seeds from me, you get a statement about what seedings
> > are good for:
>
> > You can either view seedings as predictive of the results of the
> > upcoming tournament or reflective of the results of the previous
> > played games.  Given the UPA requirement that sectional (regional)
> > finish must be maintained, seeding at regionals (nationals) must be
> > reflective rather than predictive.
>
> > Before you get seeds, you get to read how I got the seedings -- for
> > me, the process is more important than the outcome (gee, have I said
> > that before?).  The following explanation captures the key steps and
> > tries to explain the rationale for subtle parts of the process.
>
> > (1)  consider the highest remaining team from each section
> > (2)  compare these teams to eachother in a mini tournament (similar to
> > what is done on the UPA SRT tool in the team grid)
> >        the points for consideration are, in order:
> >        (a)  head to head record
> >        (b)  record against common opponents at regionals that have not
> > already been seeded
> >        (c)  record against common opponents not at regionals
> >        (d)  RRI, last years results, flip a coin -- at this point
> > there really isnt much useful information.  Get teams to play more
> > games!
> > (3)  after playing out this mini tournament, see if any team has a
> > unique best record.  If that team exists, they get the next seed.  If
> > that team does not exist, then there are two possiblities
> >        (a)  if every matchup was decided with the same criteria (i.e.
> > head to head) then consider the next criteria (common opponents) to
> > make the decision
> >        (b)  if matchups were decided using different criteria, then
> > nullify all of the lowest criteria used to make decisions (i.e. 1
> > matchup with head to head, 2 with rri, then the rri matchups become
> > draws)
> > (4)  repeat this process until all teams are seeded.
>
> > There are a couple of important clarifications when applying this
> > algorithm
> > (1)  When comparing common opponents, compare the winning percentage
> > against each individual team (i.e. aTm goes 2-1 against Davidson while
> > LSU goes 1-0 against Davidson, then LSU would 'win' that matchup based
> > on that common opponent)
>
> > (2)  Placing higher than another team at sectionals negates all
> > previous victories/losses (i.e. if texas went 0-3 against aTm this
> > year, then because texas placed higher than atm at sectionals, that
> > record would be interpreted as 1-0).
>
> > (3)  do not consider common opponents that have already been seeded.
> > Basically, using Arkansas's victory over Texas as an argument to seed
> > Arkansas ahead of aTm is also an argument to seed arkansas above
> > Texas.  If texas has already been seeded, then we the argument is no
> > longer compelling (and we have already decided that Arkansas is below
> > texas, so the whole transitivity relationship begins to fall apart).
>
> > (4)  Consider common opponents at regionals separately from common
> > opponents not at regionals for 2 reasons.  First, it gives us another
> > layer of 'tie breaks'. Second, a common opponent at the tournament
> > will be directly impacted by the seedings so results with that
> > opponent should weigh more heavily than results against opponents not
> > at the tournament.
>
> > And now you get the seedings, with the short explanation for why each
> > seed was selected.
>
> > (1)  Texas
> >      o  Texas over Kansas head to head
> >      o   Texas over LSU based on common opponents 2-1 (GA Tech, atm in
> > favor of texas, Arkansas in favor of LSU)
> >       -  note this is tied 1-1 for teams at regionals, and 1-0 for
> > teams not at regionals
> > (2)  Kansas
> >      o  Kansas over UNT based on common opponents UCSD, arkansas
> >      -  again, arkansas at regionals is enough to stop consideration.
> > UCSD is just gravy
> >      o Kansas over LSU based on common opponents OK State, arkansas,
> > notredame, michigan state
> >      - see the comment above
> > (3) Arkansas
> >      o arkansas over UNT based on head to head
> >      o  arkansas over LSU based on common opponents Missouri state and
> > atm
> >      - atm is at regionals, so missouri state is just extra
> > (4)  UNT
> >      o  LSU over UNT based on common opponent Davidson
> >     - not at regionals
> >      o  Washington over LSU based on common opponents 3-1  (Notre
> > Dame, Michigan St, Missouri state in favor of wash, davidson in favor
> > of LSU).
> >     - no teams at regionals in the common opponent list
> >     o   UNT over washington based on common opponent Tx State.
> >     - tx state at regionals, carries more weight than other options
>
> >     This is the first tricky one.  LSU and UNT have a common opponent
> > arkansas, but Arkansas has already been seeded so should be ignored.
> > LSU over UNT is based on team not at regionals, Washington over LSU is
> > based on teams not at regionals.  UNT over washington is based on a
> > team at regionals.  So the UNT edge remains at the 'top' and we break
> > one of the latter two.  End result, UNT as the 4 seed.
>
> > (5)  tx state
> >     o Washington over LSU based on common opponents (3-1 as above)
> >     o  Tx State over washington based on head to head
> >     o  tx state over lsu based on common opponents atm at regionals,
> > common opponents Ohio St, OK St not at regionals
>
> > (6)  washington
> >     o washington over lsu as above
> >     o washington over atm on common opponents 2-1 (OK, Missouri in
> > favor of washington, davidson for atm)
> >     - again, Missouri at regionals is enough at 1-0.  the other two
> > are a wash
> >     o atm over lsu head to head
>
> > (7)  truman state
> >    o truman over lsu based on common opponents emory, missour state,
> > ok state
> >     - none at regionals
> >    o truman over atm based on common opponent OK state, missouri
> >     - missouri at regionals stops discussion
> >    o  atm over lsu head to head
>
> > (8) atm
> >    o atm over lsu head to head
> >    o  missouri over atm head to head
> >    o lsu over missouri head to head
> >  ====
> >    o atm over lsu on rri  (common opponents are identical records)
> >    o atm over missouri on common opponent grinnel
> >     - not at regionals
> >    o missouri over lsu on common opponents ok state, missouri state
> >     - not at regionals
>
> > This one is also tricky given the chain of head to heads.  at the end
> > of the day, atm has claims over the other two teams on weaker
> > conditions so they get the seed.
>
> > (9)  lsu
> >    o lsu over missouri head to head
> >    o lsu over rice common opponents 2-1 (new mexico, davidson for lsu,
> > ok state for rice)
> >      - none at regionals
> >     o missouri over rice on RRI
>
> > (10)  missouri
> >    o missouri over rice on rri
> >    o missouri over mississippi on rri
> >    o  mississippi over rice on rri
>
> > (11)  mississippi
> >    o mississippi over rice on rri
> >    o  rice over harding on rri
> >    o mississipppi over harding head to head
>
> > (12)  rice
> >    o rice over harding on rri
> >    o harding over missippi state on rri
> >    o rice over miss. state on rri
>
> > (13)  ms state
> >    o ms state over harding on rri
> >    o harding over houston on rri
>
> > (14) harding
> >    o harding over houston on rri
>
> > (15)  houston
>
> > (16)  UT-Arlington
>
> > For the TL;DR folks:
>
> > 1) Texas
> > 2) Kansas
> > 3) Arkansas
> > 4) UNT
> > 5) tx state
> > 6) washington
> > 7) truman
> > 8) atm
> > 9) lsu
> > 10) missouri
> > 11) mississippi
> > 12) rice
> > 13) ms state
> > 14)  harding
> > 15) houston
> > 16) ut-arlington
>
> > These seeds give an unfortunate number of sectionals rematches and put
> > the texas and ozarks sections on opposite sides of the bracket
> > (assuming a 16 team bracket). In order to resolve those issues, I
> > advocate swapping the 3/4 seeds, the 7/8 seeds, and the 12/13 seeds.
> > This leaves only one inter-sectional matchup between two teams from
> > the section that has 7 teams at regionals (texas - UT Arlington).
>
> > ps. Mike, is this a good start on reaching 100 by Tuesday?
>
> So you make this long post with a thorough (though still arbitrary and
> based on which factors you deem to be more important than others)
> analysis of seedings, and then you conclude it by saying basically -
> this doesn't work so switch these 6 teams around.  If you just move
> teams around to fit into the part of the framework not accounted for
> in your system, then what is the point of the system?  Despite the
> hard ...
>
> read more »

http://www.stylelead.com
http://www.zhuzhiyan.com
Re: South Regionals... [message #18785 is a reply to message #18739] Tue, 14 April 2009 07:45 Go to previous messageGo to next message
Flo
Messages: 46
Registered: October 2008
Member
>
> Before you get seeds, you get to read how I got the seedings -- for
> me, the process is more important than the outcome (gee, have I said
> that before?).  The following explanation captures the key steps and
> tries to explain the rationale for subtle parts of the process.
>
> (1)  consider the highest remaining team from each section

Allen,
this is your most important tool to make your algorithm manageable.
But this is the place where your system is flawed. If an upset happens
at sectionals you punish the upsettee severely (give them a seed very
far below the seed they would have gotten without the upset) and
reward the upsetter only barely (give them a seed just slightly above
the seed they would have gotten without the upset).

Punishing the upsettee harshly may seem ok, but this also punishes the
first round opponent of that team by having to play a much better team
than earned. Yes, our regional formats are fairly robust against
unbalanced (in the sense of predictive seeding), but if you argue that
way you may as well seed alphabetically or randomly.

The problem at hand is that the UPA seeding tries to do two things:
1. Seed teams according to their accomplishments all year
2. Keep the order decided at sectionals to mimick a big tournament
consisting of sectionals/regionals/nationals

These two things don't work together if severe upsets happen. You
could get 1 if you dropped the requirement of seeding by placement of
previous stages and used the results from previous stages merely as
most recent (and thus important) full strength comparison between two
teams.

Or you could completely go with the second philosophy, where teams in
the region are completely seeded before sectionals, and if upsets
happen the two teams merely switch seeds in the next stage.

The UPA currently wants a mixture of both... and I like Tarr's
philosophy on this: Seed all teams according to philosophy 1 ("true
seeds"). Then, if this violates the sectionals-regionals rule due to
an upset at sectionals, seed both these teams as close as possible to
the average of their combined strength (look at their combined record,
normalized for number of games played per team and treat them as one
team) in the order inherited from sectionals.

Flo.
Re: South Regionals... [message #18793 is a reply to message #18785] Tue, 14 April 2009 08:36 Go to previous messageGo to next message
aclement
Messages: 62
Registered: October 2008
Member
On Apr 14, 9:45 am, Flo <Flo.Pfen...@googlemail.com> wrote:
> > Before you get seeds, you get to read how I got the seedings -- for
> > me, the process is more important than the outcome (gee, have I said
> > that before?).  The following explanation captures the key steps and
> > tries to explain the rationale for subtle parts of the process.
>
> > (1)  consider the highest remaining team from each section
>
> Allen,
> this is your most important tool to make your algorithm manageable.
> But this is the place where your system is flawed. If an upset happens
> at sectionals you punish the upsettee severely (give them a seed very
> far below the seed they would have gotten without the upset) and
> reward the upsetter only barely (give them a seed just slightly above
> the seed they would have gotten without the upset).

My take on the seeding instructions from the UPA is that they are
underspecified and trying to do two contradictory things at once,
which you have identified. Given the strict "Obey sectional finish"
requirement, it seems to me that the UPA is at least, implicitly,
stating that seeding teams according to the accomplishments of the
year is the correct thing to do and that "predictive seeding" is not
the right answer. Consequently, I do think that a mechanical
mechanism for seeding that can be applied by any diligent person is
appropriate. Which mechanism should we use? I find that to be an
interesting, and open, question.

I think it is important and reasonable to consider the next team from
each section in isolation rather than in the context of the lower
finishing teams because of the very real possiblity (Delaware* from
the Colonial section this year) of a good team being "upset" and not
qualifying for regionals. Does the fact that Delaware was expected to
be the 3 seed at ME regionals justify pushing the teams that upset
Delaware up into the high seeds of regionals even though Delaware is
not at that tournament? To me, the answer is no. I don't think the
seed for the top seed from the colonial section should not change
dramatically based on how many teams from the section qualify for
regionals.

>
> Punishing the upsettee harshly may seem ok, but this also punishes the
> first round opponent of that team by having to play a much better team
> than earned. Yes, our regional formats are fairly robust against
> unbalanced (in the sense of predictive seeding), but if you argue that
> way you may as well seed alphabetically or randomly.

Honestly, I would not object to alphabetic or random seeding,
especially for the pool play to bracket formats.

>
> The problem at hand is that the UPA seeding tries to do two things:
> 1. Seed teams according to their accomplishments all year
> 2. Keep the order decided at sectionals to mimick a big tournament
> consisting of sectionals/regionals/nationals
>
> These two things don't work together if severe upsets happen. You
> could get 1 if you dropped the requirement of seeding by placement of
> previous stages and used the results from previous stages merely as
> most recent (and thus important) full strength comparison between two
> teams.
>

I disagree with the assessment that the most recent game is the most
important comparison between two teams. How do you compare a game at
stanford with perfect weather to a game at fool's fest with 35 mph
sustained winds in which every game is decided by the flip?
(hypothetical). Fool's fest is more recent, but the reseults are
clearly skewed by weather. If you discard the 35mph winds, what about
30? 20? at what point are the wins 'acceptable' to consider?

> Or you could completely go with the second philosophy, where teams in
> the region are completely seeded before sectionals, and if upsets
> happen the two teams merely switch seeds in the next stage.
>
> The UPA currently wants a mixture of both... and I like Tarr's
> philosophy on this: Seed all teams according to philosophy 1 ("true
> seeds"). Then, if this violates the sectionals-regionals rule due to
> an upset at sectionals, seed both these teams as close as possible to
> the average of their combined strength (look at their combined record,
> normalized for number of games played per team and treat them as one
> team) in the order inherited from sectionals.

How do you seed teams according to philosophy 1 ("true seeds"
reflective of their results from this year)? My candidate mechanism
for doing that seeding would be to go back to my initial description
and replace "next team from each section" with "all candidate teams
for the next seed". When seeding regionals, I currently define the
candidate team from each section to be the highest available finisher
from that section; other definitions are viable. Do you have another
candidate that is better?

allen

* I am basing all discussion of Delaware on the fact that they were
mentioned in the C1 team list at the end of the year. If their actual
play did not match that preseason evaluation, then please treat them
as a strawman example.
Re: South Regionals... [message #18795 is a reply to message #18763] Tue, 14 April 2009 08:44 Go to previous messageGo to next message
aclement
Messages: 62
Registered: October 2008
Member
On Apr 14, 1:45 am, john.as...@gmail.com wrote:

> So you make this long post with a thorough (though still arbitrary and
> based on which factors you deem to be more important than others)
> analysis of seedings, and then you conclude it by saying basically -
> this doesn't work so switch these 6 teams around.  If you just move
> teams around to fit into the part of the framework not accounted for
> in your system, then what is the point of the system?  Despite the
> hard ...
>

(1) I'm more interested in the seeding mechanism than the results of
the seeding. I freely acknowledge that the factors I choose to
consider are somewhat arbitrary. After all, there are many factors
(total games played, points scored, point differential, number of blue
eyed eskimos) that I am not considering. What factors do you think I
should have used? which factors should I have not used? Should the
factors be weighted differently? These are all interesting questions
and I would love to hear your answers.

(2) If it were completely up to me, I would not modify the seedings

(3) It is a legitimate concern that the bracket could end up as a
replay of sectionals. this is undesirable from a competition and
fairness perspective

(4) Given the possibility to address (3) with minor changes to the
initial seeds, I think the change is worth considering and is less
offensive than other alternatives that I have seen.
Re: South Regionals... [message #18805 is a reply to message #18795] Tue, 14 April 2009 09:28 Go to previous messageGo to next message
mikehouston37
Messages: 76
Registered: October 2008
Member
In talking about THIS year's regionals with regards to Allen's formula
(which I've come to love), do we have any Sectional upsets that we're
worried about? Doesn't look like it to me. Top 4 in the Ozarks went
to seed from Sectionals. UNT leaped Tx St at Sectionals but had a
pretty comparable season with slightly better common opp, so no
problem there. LSU jumped Mississippi at Sectionals, but we all know
the Bama section is moot. So where's the upsets that make Allen's
formula an abomination? I see none.

-MH
Re: South Regionals... [message #18824 is a reply to message #18805] Tue, 14 April 2009 11:15 Go to previous messageGo to next message
pingy
Messages: 7
Registered: April 2009
Junior Member
On Apr 14, 11:28 am, MH <mikehousto...@gmail.com> wrote:
> In talking about THIS year's regionals with regards to Allen's formula
> (which I've come to love), do we have any Sectional upsets that we're
> worried about?  Doesn't look like it to me.  Top 4 in the Ozarks went
> to seed from Sectionals.  UNT leaped Tx St at Sectionals but had a
> pretty comparable season with slightly better common opp, so no
> problem there.  LSU jumped Mississippi at Sectionals, but we all know
> the Bama section is moot.  So where's the upsets that make Allen's
> formula an abomination?  I see none.
>
> -MH

the bama section is moot? thats an angry statement. sure none of the
bama section teams are C1 worthy, but i'd pick vandy or auburn to be
in regionals over UT-Arlington or Houston. dont underestimate the bama
section just because its small and doesn't get much national
attention. and to seed the winner of the section as 9th out of 16?
thats pretty bold! lsu doesn't deserve first or second, or even up to
the 4th. the fifth seed is a long stretch, but surely they're above
the 9th seed, in the top half at least. i implore the people to
reconsider the rankings.
Re: South Regionals... [message #18830 is a reply to message #18824] Tue, 14 April 2009 11:57 Go to previous messageGo to next message
paul.adkisson
Messages: 35
Registered: October 2008
Member
On Apr 14, 1:15 pm, pingy <littletigr...@gmail.com> wrote:
> On Apr 14, 11:28 am, MH <mikehousto...@gmail.com> wrote:
>
> > In talking about THIS year's regionals with regards to Allen's formula
> > (which I've come to love), do we have any Sectional upsets that we're
> > worried about?  Doesn't look like it to me.  Top 4 in the Ozarks went
> > to seed from Sectionals.  UNT leaped Tx St at Sectionals but had a
> > pretty comparable season with slightly better common opp, so no
> > problem there.  LSU jumped Mississippi at Sectionals, but we all know
> > the Bama section is moot.  So where's the upsets that make Allen's
> > formula an abomination?  I see none.
>
> > -MH
>
> the bama section is moot? thats an angry statement. sure none of the
> bama section teams are C1 worthy, but i'd pick vandy or auburn to be
> in regionals over UT-Arlington or Houston. dont underestimate the bama
> section just because its small and doesn't get much national
> attention. and to seed the winner of the section as 9th out of 16?
> thats pretty bold! lsu doesn't deserve first or second, or even up to
> the 4th. the fifth seed is a long stretch, but surely they're above
> the 9th seed, in the top half at least. i implore the people to
> reconsider the rankings.

hahahaha.
Re: South Regionals... [message #18839 is a reply to message #18830] Tue, 14 April 2009 12:30 Go to previous messageGo to next message
ncwohl31
Messages: 43
Registered: November 2008
Member
Think about the most popular tournament in the country (the NCAA Men's
Basketball Tournament). You can earn a bid into this tournament by
winning your conference tournament, no matter how your season went
(Syracuse did this a couple years back in 2006). Sectionals gives
teams a chance to qualify for Regionals, thats it. Big upsets should
be taken into account, but the whole season resume takes precedence.
In general, Regionals should be a separate tournament with the sole
purpose of giving teams a chance to qualify for Nationals. The whole
season resume should be taken into account (including Sectionals
games). Just because a 'Bama team won their section, doesn't mean they
are a top team necessarily, it just means they are a top team in their
section only. Look at their season schedules/results as a whole and
compare that against other teams in their region. Futhermore, since
this is a separate tournament than Sectionals, seedings should not be
"re-arranged" because these teams are from the same section or
whatever. Allen's seedings (before suggested rearranging) seem fair.
Re: South Regionals... [message #18841 is a reply to message #18824] Tue, 14 April 2009 12:44 Go to previous messageGo to next message
Alex DW
Messages: 21
Registered: April 2009
Junior Member
On Apr 14, 11:15 am, pingy <littletigr...@gmail.com> wrote:
> On Apr 14, 11:28 am, MH <mikehousto...@gmail.com> wrote:
>
> > In talking about THIS year's regionals with regards to Allen's formula
> > (which I've come to love), do we have any Sectional upsets that we're
> > worried about?  Doesn't look like it to me.  Top 4 in the Ozarks went
> > to seed from Sectionals.  UNT leaped Tx St at Sectionals but had a
> > pretty comparable season with slightly better common opp, so no
> > problem there.  LSU jumped Mississippi at Sectionals, but we all know
> > the Bama section is moot.  So where's the upsets that make Allen's
> > formula an abomination?  I see none.
>
> > -MH
>
> the bama section is moot? thats an angry statement. sure none of the
> bama section teams are C1 worthy, but i'd pick vandy or auburn to be
> in regionals over UT-Arlington or Houston. dont underestimate the bama
> section just because its small and doesn't get much national
> attention. and to seed the winner of the section as 9th out of 16?
> thats pretty bold! lsu doesn't deserve first or second, or even up to
> the 4th. the fifth seed is a long stretch, but surely they're above
> the 9th seed, in the top half at least. i implore the people to
> reconsider the rankings.

Be careful what you say about Houston. I hear that a certain skinny
fellow with sticky fingers playing with them this year.

-DW
Re: South Regionals... [message #18843 is a reply to message #18841] Tue, 14 April 2009 12:55 Go to previous messageGo to next message
Caleb
Messages: 52
Registered: September 2008
Member
> Be careful what you say about Houston.  I hear that a certain skinny
> fellow with sticky fingers playing with them this year.
>
> -DW

That fast kid with glasses who sometimes whines about calls?

-CS
Re: South Regionals... [message #18847 is a reply to message #18841] Tue, 14 April 2009 13:23 Go to previous messageGo to next message
pingy
Messages: 7
Registered: April 2009
Junior Member
On Apr 14, 2:44 pm, Alex DW <kadrl...@gmail.com> wrote:
> On Apr 14, 11:15 am, pingy <littletigr...@gmail.com> wrote:
>
>
>
> > On Apr 14, 11:28 am, MH <mikehousto...@gmail.com> wrote:
>
> > > In talking about THIS year's regionals with regards to Allen's formula
> > > (which I've come to love), do we have any Sectional upsets that we're
> > > worried about?  Doesn't look like it to me.  Top 4 in the Ozarks went
> > > to seed from Sectionals.  UNT leaped Tx St at Sectionals but had a
> > > pretty comparable season with slightly better common opp, so no
> > > problem there.  LSU jumped Mississippi at Sectionals, but we all know
> > > the Bama section is moot.  So where's the upsets that make Allen's
> > > formula an abomination?  I see none.
>
> > > -MH
>
> > the bama section is moot? thats an angry statement. sure none of the
> > bama section teams are C1 worthy, but i'd pick vandy or auburn to be
> > in regionals over UT-Arlington or Houston. dont underestimate the bama
> > section just because its small and doesn't get much national
> > attention. and to seed the winner of the section as 9th out of 16?
> > thats pretty bold! lsu doesn't deserve first or second, or even up to
> > the 4th. the fifth seed is a long stretch, but surely they're above
> > the 9th seed, in the top half at least. i implore the people to
> > reconsider the rankings.
>
> Be careful what you say about Houston.  I hear that a certain skinny
> fellow with sticky fingers playing with them this year.
>
> -DW

well i haven't seen houston play, so i'm not sure about their level of
play, i was just going by upa.
Re: South Regionals... [message #18852 is a reply to message #18841] Tue, 14 April 2009 13:53 Go to previous messageGo to next message
patrick.eberle
Messages: 12
Registered: October 2008
Junior Member
On Apr 14, 2:44 pm, Alex DW <kadrl...@gmail.com> wrote:
> On Apr 14, 11:15 am, pingy <littletigr...@gmail.com> wrote:
>
>
>
>
>
> > On Apr 14, 11:28 am, MH <mikehousto...@gmail.com> wrote:
>
> > > In talking about THIS year's regionals with regards to Allen's formula
> > > (which I've come to love), do we have any Sectional upsets that we're
> > > worried about?  Doesn't look like it to me.  Top 4 in the Ozarks went
> > > to seed from Sectionals.  UNT leaped Tx St at Sectionals but had a
> > > pretty comparable season with slightly better common opp, so no
> > > problem there.  LSU jumped Mississippi at Sectionals, but we all know
> > > the Bama section is moot.  So where's the upsets that make Allen's
> > > formula an abomination?  I see none.
>
> > > -MH
>
> > the bama section is moot? thats an angry statement. sure none of the
> > bama section teams are C1 worthy, but i'd pick vandy or auburn to be
> > in regionals over UT-Arlington or Houston. dont underestimate the bama
> > section just because its small and doesn't get much national
> > attention. and to seed the winner of the section as 9th out of 16?
> > thats pretty bold! lsu doesn't deserve first or second, or even up to
> > the 4th. the fifth seed is a long stretch, but surely they're above
> > the 9th seed, in the top half at least. i implore the people to
> > reconsider the rankings.
>
> Be careful what you say about Houston.  I hear that a certain skinny
> fellow with sticky fingers playing with them this year.
>
> -DW- Hide quoted text -
>
> - Show quoted text -

I watched UH v ATM and that dude was throwing pretty well in the
wind. However his sticky fingers must have changes since leaving Wash
U because he must have had 7 or 8 drops on dump passes. Pretty tough
to watch. Houston was surprisingly not bad.
Re: South Regionals... [message #18853 is a reply to message #18852] Tue, 14 April 2009 13:59 Go to previous messageGo to next message
Caleb
Messages: 52
Registered: September 2008
Member
Is it true that they are a first year program? Or did I hear this
wrong?
Re: South Regionals... [message #18860 is a reply to message #18852] Tue, 14 April 2009 14:05 Go to previous messageGo to next message
pingy
Messages: 7
Registered: April 2009
Junior Member
> I watched UH v ATM and that dude was throwing pretty well in the
> wind.  However his sticky fingers must have changes since leaving Wash
> U because he must have had 7 or 8 drops on dump passes.  Pretty tough
> to watch.  Houston was surprisingly not bad.

oops, maybe i shouldn't rag on them then. either way they're in
regionals so they must be a decent team at the least.
Re: South Regionals... [message #18861 is a reply to message #18853] Tue, 14 April 2009 14:09 Go to previous messageGo to next message
aclement
Messages: 62
Registered: October 2008
Member
On Apr 14, 3:59 pm, Captain Crunch <CalebSta...@gmail.com> wrote:
> Is it true that they are a first year program?  Or did I hear this
> wrong?

they've been in and out of sectionals the past 5 years or so.
Sometimes tehy have a team, some times they dont.

allen
Re: South Regionals... [message #18867 is a reply to message #18805] Tue, 14 April 2009 14:23 Go to previous messageGo to next message
Michael McNeil
Messages: 36
Registered: October 2008
Member
On Apr 14, 12:28 pm, MH <mikehousto...@gmail.com> wrote:
> In talking about THIS year's regionals with regards to Allen's formula
> (which I've come to love), do we have any Sectional upsets that we're
> worried about?  Doesn't look like it to me.  Top 4 in the Ozarks went
> to seed from Sectionals.  UNT leaped Tx St at Sectionals but had a
> pretty comparable season with slightly better common opp, so no
> problem there.  LSU jumped Mississippi at Sectionals, but we all know
> the Bama section is moot.  So where's the upsets that make Allen's
> formula an abomination?  I see none.
>
> -MH

Wow Mike, way to stick up for your former Region.

Leave the Bama Section alone!

http://bamasecs.wordpress.com/
Re: South Regionals... [message #18900 is a reply to message #18867] Tue, 14 April 2009 18:19 Go to previous messageGo to next message
mikehouston37
Messages: 76
Registered: October 2008
Member
Calm down Bama Section just jokes. It's always funner to make fun of
the kid who likes to defend himself than the one who takes it.
Anyways, show up and win your games, that's all that matters.

-MH
Re: South Regionals... [message #18902 is a reply to message #18867] Tue, 14 April 2009 18:32 Go to previous messageGo to next message
McB
Messages: 65
Registered: September 2008
Member
On Apr 14, 5:23 pm, Michael McNeil <rmichael.mcn...@gmail.com> wrote:
> On Apr 14, 12:28 pm, MH <mikehousto...@gmail.com> wrote:
>
> > In talking about THIS year's regionals with regards to Allen's formula
> > (which I've come to love), do we have any Sectional upsets that we're
> > worried about?  Doesn't look like it to me.  Top 4 in the Ozarks went
> > to seed from Sectionals.  UNT leaped Tx St at Sectionals but had a
> > pretty comparable season with slightly better common opp, so no
> > problem there.  LSU jumped Mississippi at Sectionals, but we all know
> > the Bama section is moot.  So where's the upsets that make Allen's
> > formula an abomination?  I see none.
>
> > -MH
>
> Wow Mike, way to stick up for your former Region.
>
> Leave the Bama Section alone!
>
> http://bamasecs.wordpress.com/

Some historical Regionals perspective:

2008: Bama sectionals teams enter at 6, 10, 13, 15 (average 11) and
leave at 9T, 11T, 11T, 13T (average 11)
2007: 3 of the 4 Bama sectionals teams broke seed in pool play and LSU
and Auburn finished above seed (I have no idea about Vandy and
Mississippi, as score reporter is incomplete and the 17 team regionals
throws off traditional snake seeding.) - LSU finishes top 3 (breaking
seed by at least 2)
2006: Bama section enters at 5, 9, 14 (average 9.333) and leave at 5,
7, 9 (average 7) - LSU finishes top 5
2005: Bama section enters 8, 9 and leave 5T, 9T (LSU not in attendance
due to DQ after sectionals.)
2004: Enter 5, 6, 11 and leave 2, 5T, 9T - LSU finishes top 2
(breaking seed by 3)

That's as far back as score reporter was cooperative. So, every year
that LSU has been to Regionals, they have held or broken seed,
finishing top 5. The section as a whole tends to break seed. The
bottom Bama teams are generally better than the bottom teams from
other sections and LSU is generally (read: always) better than over
half the field.

I have no idea how things should be seeded this year, as I have played
against practically 0 of these teams, but the Bama section in general
and LSU in particular traditionally get underseeded.

-McB
Re: South Regionals... [message #18909 is a reply to message #18902] Tue, 14 April 2009 19:32 Go to previous messageGo to next message
Caleb
Messages: 52
Registered: September 2008
Member
2008:

Ozark Section
Coming in seeded (3,4,5,8,11,12) average = 7.16
Placement (3,4,5T,7,8,14) average = 6.83

Texas Section
Coming in seeded (1,2,7,9,14,16) average = 8.16
Placement (1,2,5T,9T,15,16) average = 8

Looks like every section broke seed!!!

Those ties naturally bring down the average on the final results...

The other years are probably more legit, but that is still a bad way
to look at it.

There are also other factors that affect final placement, like last
year our team (Truman State) only took 9 guys down to Regionals
because it was an 14 hour drive and ended up playing savage.
Initially seeding it would have never factored in something like that.
Re: South Regionals... [message #18923 is a reply to message #18902] Tue, 14 April 2009 22:01 Go to previous messageGo to next message
dude
Messages: 6
Registered: September 2008
Junior Member
On Apr 14, 8:32 pm, McB <Christopher.M.McBr...@gmail.com> wrote:
> On Apr 14, 5:23 pm, Michael McNeil <rmichael.mcn...@gmail.com> wrote:
>
>
>
> > On Apr 14, 12:28 pm, MH <mikehousto...@gmail.com> wrote:
>
> > > In talking about THIS year's regionals with regards to Allen's formula
> > > (which I've come to love), do we have any Sectional upsets that we're
> > > worried about?  Doesn't look like it to me.  Top 4 in the Ozarks went
> > > to seed from Sectionals.  UNT leaped Tx St at Sectionals but had a
> > > pretty comparable season with slightly better common opp, so no
> > > problem there.  LSU jumped Mississippi at Sectionals, but we all know
> > > the Bama section is moot.  So where's the upsets that make Allen's
> > > formula an abomination?  I see none.
>
> > > -MH
>
> > Wow Mike, way to stick up for your former Region.
>
> > Leave the Bama Section alone!
>
> >http://bamasecs.wordpress.com/
>
> Some historical Regionals perspective:
>
> 2008: Bama sectionals teams enter at 6, 10, 13, 15 (average 11) and
> leave at 9T, 11T, 11T, 13T (average 11)
> 2007: 3 of the 4 Bama sectionals teams broke seed in pool play and LSU
> and Auburn finished above seed (I have no idea about Vandy and
> Mississippi, as score reporter is incomplete and the 17 team regionals
> throws off traditional snake seeding.) - LSU finishes top 3 (breaking
> seed by at least 2)
> 2006: Bama section enters at 5, 9, 14 (average 9.333) and leave at 5,
> 7, 9 (average 7) - LSU finishes top 5
> 2005: Bama section enters 8, 9 and leave 5T, 9T (LSU not in attendance
> due to DQ after sectionals.)
> 2004: Enter 5, 6, 11 and leave 2, 5T, 9T - LSU finishes top 2
> (breaking seed by 3)
>
> That's as far back as score reporter was cooperative. So, every year
> that LSU has been to Regionals, they have held or broken seed,
> finishing top 5. The section as a whole tends to break seed. The
> bottom Bama teams are generally better than the bottom teams from
> other sections and LSU is generally (read: always) better than over
> half the field.
>
> I have no idea how things should be seeded this year, as I have played
> against practically 0 of these teams, but the Bama section in general
> and LSU in particular traditionally get underseeded.
>
> -McB

Wow!!! Tied for 9th=9th!
LSU dominated last years regionals!
Re: South Regionals... [message #18924 is a reply to message #18909] Tue, 14 April 2009 22:04 Go to previous messageGo to next message
clrydin
Messages: 30
Registered: October 2008
Member
unfortunately I don't have an amazing statistical breakdown for every
position, but for the sake of getting this forum back on track... my
opinion:

1) Texas (win over kansas)
2) Kansas
3) Arkansas (win over North Texas)
4) North Texas (superior season results than the next options Wash U
and LSU)
5) Texas State (win over Wash U, superior results to LSU)
6) Texas A&M (win over LSU, mixed results with Wash U but higher RRI)
7) Wash U
8) LSU
9) Truman
10) Mizzou
11) Ole Miss
12) Rice
13) Harding
14) Miss St
15) Houston
16) UTA

much of the bottom seeds based solely on RRI.

My opinion seems to be close to that of the original post, with the
main exception being Arkansas ahead of both North Texas and Texas
State due to arkansas' head to head over NTexas.

If you disagree with any positions, please give reason to support your
disagreement.
Re: South Regionals... [message #18925 is a reply to message #18924] Tue, 14 April 2009 22:27 Go to previous messageGo to next message
aclement
Messages: 62
Registered: October 2008
Member
On Apr 15, 12:04 am, Clrydin34 <clry...@hotmail.com> wrote:
> unfortunately I don't have an amazing statistical breakdown for every
> position, but for the sake of getting this forum back on track... my
> opinion:
>
> 1) Texas (win over kansas)
> 2) Kansas
> 3) Arkansas (win over North Texas)
> 4) North Texas (superior season results than the next options Wash U
> and LSU)
> 5) Texas State (win over Wash U, superior results to LSU)
> 6) Texas A&M (win over LSU, mixed results with Wash U but higher RRI)
> 7) Wash U
> 8) LSU
> 9) Truman
> 10) Mizzou
> 11) Ole Miss
> 12) Rice
> 13) Harding
> 14) Miss St
> 15) Houston
> 16) UTA
>
> much of the bottom seeds based solely on RRI.
>
> My opinion seems to be close to that of the original post, with the
> main exception being Arkansas ahead of both North Texas and Texas
> State due to arkansas' head to head over NTexas.
>
> If you disagree with any positions, please give reason to support your
> disagreement.

Your rationality isnt wanted here. The whole point that matters is
that we're dissing the Bama section and they never get respect!!!
It's all because the man wants to put the bama section in its place!
Re: South Regionals... [message #19001 is a reply to message #18923] Wed, 15 April 2009 10:52 Go to previous messageGo to next message
kdoe
Messages: 33
Registered: September 2008
Member
>
> Wow!!! Tied for 9th=9th!
> LSU dominated last years regionals!

LSU didn't make Regionals last year.

Last year all the Bama teams were eliminated from contention after the
3rd round.
Re: South Regionals... [message #19008 is a reply to message #19001] Wed, 15 April 2009 11:31 Go to previous messageGo to next message
Daag Alemayehu
Messages: 249
Registered: September 2008
Senior Member
kdoe13@gmail.com wrote:
>> Wow!!! Tied for 9th=9th!
>> LSU dominated last years regionals!
>
> LSU didn't make Regionals last year.
>
> Last year all the Bama teams were eliminated from contention after the
> 3rd round.
>

Bam. Someone just got Doeged.
Re: South Regionals... [message #19048 is a reply to message #19001] Wed, 15 April 2009 15:41 Go to previous messageGo to next message
dude
Messages: 6
Registered: September 2008
Junior Member
On Apr 15, 12:52 pm, kdo...@gmail.com wrote:
> > Wow!!! Tied for 9th=9th!
> > LSU dominated last years regionals!
>
> LSU didn't make Regionals last year.
>
> Last year all the Bama teams were eliminated from contention after the
> 3rd round.

ORLY! Way to pickup on the sarcasm
Re: South Regionals... [message #19363 is a reply to message #18867] Fri, 17 April 2009 09:36 Go to previous messageGo to next message
zontals-new
Messages: 17
Registered: December 2008
Junior Member
What happened South Region? You used to be fun.
Re: South Regionals... [message #19490 is a reply to message #18867] Fri, 17 April 2009 09:36 Go to previous messageGo to next message
zontals-new
Messages: 17
Registered: December 2008
Junior Member
What happened South Region? You used to be fun.
Re: South Regionals... [message #19883 is a reply to message #19363] Mon, 20 April 2009 17:20 Go to previous messageGo to next message
Leefus
Messages: 10
Registered: April 2009
Junior Member
On Apr 17, 12:36 pm, zontals-new <horrorzont...@gmail.com> wrote:
> What happenedSouthRegion? You used to be fun.

yea, this thread needs a Mad Dog comment
Re: South Regionals... [message #20139 is a reply to message #19363] Tue, 21 April 2009 22:18 Go to previous messageGo to next message
clrydin
Messages: 30
Registered: October 2008
Member
any word on seedings?
Re: South Regionals... [message #20201 is a reply to message #20139] Wed, 22 April 2009 08:23 Go to previous messageGo to previous message
The Dick Formerly Kno
Messages: 189
Registered: September 2008
Senior Member
http://upa.org/scores/tourn.cgi?div=127&id=5921
Previous Topic:College All-Star Alumni vs Team USA Showcase game at Nationals
Next Topic:D-III Nationals - Top 25 (4-24-09)
Goto Forum:
  


Current Time: Sat Mar 28 22:41:45 PDT 2020
.:: Contact :: Home ::.

Powered by: FUDforum 3.0.0RC2.
Copyright ©2001-2009 FUDforum Bulletin Board Software