Winthrop

"Edward M. Kennedy" <doidy.RemoveThis@wox.com> wrote

> > is demonstrating why I'm uncomfortable with even a 20 point lead.
>
> IFYPFY.

I got that one in in the nick of time.

--Tedward
 >> Stay informed about: Winthrop 

On Mar 16, 4:00 pm, "Fred Garvin, Male Prostitute"
<nos....RemoveThis@whitehouse.gov> wrote:
> In messagenews:1174074780.731898.118690@e65g2000hsc.googlegroups.com,
> James Gibson sprach forth the following:
>
> > somebody like Winthrop who played
> > mostly bad competition all year.
>
> Their only losses were to TAMU, UNC, Maryland and Wisconsin. OK that last
> one looks like a Bad Loss (TM) today, but hell - give me another team not
> favored to make it past the weekend with Maryland being their "worst loss".


They definitely only lost to good teams. Beyond those 4, they only
had 3 more losable games - @Mississippi State, @Old Dominion,
@Missouri State. Hard to judge them on that. I think just about any
bubble team sould have been able to go undefeated against the rest of
their schedule. With that sort of record, anywhere between 8th and
50th seems reasonable.
 >> Stay informed about: Winthrop 

This message is not archived
 >> Stay informed about: Winthrop 

"James Gibson" <james.m.gibson.DeleteThis@gmail.com> writes:

>is demonstrating why I'm uncomfortable with just using predictive
>ratings to set the tournament seedings. Across the predictive
>ratings, the ranked in the 50s and 60s. Among ELO-style ratings,
>they ranked in the single digits and teens. I think predictive
>ratings probably work well for the major conferences who have a lot
>of links between them, but are difficult for somebody like Winthrop
>who played mostly bad competition all year. Of course, that doesn't
>mean that the ELO-style ratings are necessarily better. Both sets
>would have linkage problems.

Interestingly, SAGARINELO is two games ahead of both SAGARIN and
SAGARINPREDICTOR in terms of picking winners, but is still behind both
of them in RMS error.

--
Milt Epstein
mepstein.DeleteThis@uiuc.edu
 >> Stay informed about: Winthrop 

James Gibson <james.m.gibson DeleteThis @gmail.com> wrote:

> They definitely only lost to good teams. Beyond those 4, they only
> had 3 more losable games - @Mississippi State, @Old Dominion,
> @Missouri State. Hard to judge them on that. I think just about any
> bubble team sould have been able to go undefeated against the rest of
> their schedule. With that sort of record, anywhere between 8th and
> 50th seems reasonable.

Note, the fact that they beat Notre Dame doesn't mean they really should
have been ranked in the teens. I do think Winthrop was a team that you
couldn't really know how good they were, the kind of team that often
ends up in the 12-13 spots. (Usually this archetype has even fewer
losses but only one or no wins in the bubble range or better). One
thing you did know from their 0-4 record against very good teams -- they
probably didn't belong in the top 10.

But their 11 seed could be quite reasonable and they still could beat
ND. I'd guess the predictors didn't underrank them by as much as ELO
overranked them.


Michael

--
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing on usenet and in e-mail?
 >> Stay informed about: Winthrop 

"Michael Sullivan" <michael.DeleteThis@bcect.com> wrote

> > They definitely only lost to good teams. Beyond those 4, they only
> > had 3 more losable games - @Mississippi State, @Old Dominion,
> > @Missouri State. Hard to judge them on that. I think just about any
> > bubble team sould have been able to go undefeated against the rest of
> > their schedule. With that sort of record, anywhere between 8th and
> > 50th seems reasonable.
>
> Note, the fact that they beat Notre Dame doesn't mean they really should
> have been ranked in the teens. I do think Winthrop was a team that you
> couldn't really know how good they were, the kind of team that often
> ends up in the 12-13 spots. (Usually this archetype has even fewer
> losses but only one or no wins in the bubble range or better). One
> thing you did know from their 0-4 record against very good teams -- they
> probably didn't belong in the top 10.

Even if they don't want to use an MOV based rating system,
what bugs me is that Winthrop gets no props for playing
*those* four teams pretty close, especially considering how
it fits in with the rest of their results.

I think anyone with a basketball clue would have seeded
them higher than 11 if they didn't need to follow whatever
it is the committee follows. The reduction in mid-majors
seems downright fraudulent, doubly so with the rise of the
Mighty Valley Conference.

--Tedward
 >> Stay informed about: Winthrop 

This message is not archived
 >> Stay informed about: Winthrop 

On Mar 16, 3:53 pm, "James Gibson" <james.m.gib....DeleteThis@gmail.com> wrote:
> is demonstrating why I'm uncomfortable with just using predictive
> ratings to set the tournament seedings. Across the predictive
> ratings, the ranked in the 50s and 60s. Among ELO-style ratings, they
> ranked in the single digits and teens. I think predictive ratings
> probably work well for the major conferences who have a lot of links
> between them, but are difficult for somebody like Winthrop who played
> mostly bad competition all year. Of course, that doesn't mean that
> the ELO-style ratings are necessarily better. Both sets would have
> linkage problems.

You might be right about Wintrop being hard to rank. But I am not
sure its a linkage problem. Might be that margin of victory (MOV) is
less predictive against all those weak teams. After about MOV>15
there is not much motivation for going higher.

Before the tourney they did not beat any ranked teams. They did give
Wisconsin a run for their money, but they did not challenge the others.
 >> Stay informed about: Winthrop 

On Mar 22, 8:32 am, "stat monkey" <tadams....RemoveThis@yahoo.com> wrote:
> On Mar 16, 3:53 pm, "James Gibson" <james.m.gib....RemoveThis@gmail.com> wrote:
>
> > is demonstrating why I'm uncomfortable with just using predictive
> > ratings to set the tournament seedings. Across the predictive
> > ratings, the ranked in the 50s and 60s. Among ELO-style ratings, they
> > ranked in the single digits and teens. I think predictive ratings
> > probably work well for the major conferences who have a lot of links
> > between them, but are difficult for somebody like Winthrop who played
> > mostly bad competition all year. Of course, that doesn't mean that
> > the ELO-style ratings are necessarily better. Both sets would have
> > linkage problems.
>
> You might be right about Wintrop being hard to rank. But I am not
> sure its a linkage problem. Might be that margin of victory (MOV) is
> less predictive against all those weak teams. After about MOV>15
> there is not much motivation for going higher.
>
> Before the tourney they did not beat any ranked teams. They did give
> Wisconsin a run for their money, but they did not challenge the others.

One thing I am sure of, Wintrop's coach is due for a job upgrade.
 >> Stay informed about: Winthrop 

On Mar 22, 8:32 am, "stat monkey" <tadams... DeleteThis @yahoo.com> wrote:
> On Mar 16, 3:53 pm, "James Gibson" <james.m.gib... DeleteThis @gmail.com> wrote:
>
> > is demonstrating why I'm uncomfortable with just using predictive
> > ratings to set the tournament seedings. Across the predictive
> > ratings, the ranked in the 50s and 60s. Among ELO-style ratings, they
> > ranked in the single digits and teens. I think predictive ratings
> > probably work well for the major conferences who have a lot of links
> > between them, but are difficult for somebody like Winthrop who played
> > mostly bad competition all year. Of course, that doesn't mean that
> > the ELO-style ratings are necessarily better. Both sets would have
> > linkage problems.
>
> You might be right about Wintrop being hard to rank. But I am not
> sure its a linkage problem. Might be that margin of victory (MOV) is
> less predictive against all those weak teams. After about MOV>15
> there is not much motivation for going higher.
>
> Before the tourney they did not beat any ranked teams. They did give
> Wisconsin a run for their money, but they did not challenge the others.

They didn't beat any ranked teams. But they won at Missouri State and
at Mississippi State, two teams that had good predictive ratings (40th
and 38th according to the LRMC). Winthrop was only 68th in LRMC, and
I suspect that somewhat narrow wins over their own conference was the
thing that brought them down there. However, Winthrop actually fairs
slightly worse (71st) in LRMC capped, while they move up to 41st in
LRMC(0).
 >> Stay informed about: Winthrop 

On Mar 22, 8:43 am, "James Gibson" <james.m.gib....DeleteThis@gmail.com> wrote:
> On Mar 22, 8:32 am, "stat monkey" <tadams....DeleteThis@yahoo.com> wrote:
>
>
>
>
>
> > On Mar 16, 3:53 pm, "James Gibson" <james.m.gib....DeleteThis@gmail.com> wrote:
>
> > > is demonstrating why I'm uncomfortable with just using predictive
> > > ratings to set the tournament seedings. Across the predictive
> > > ratings, the ranked in the 50s and 60s. Among ELO-style ratings, they
> > > ranked in the single digits and teens. I think predictive ratings
> > > probably work well for the major conferences who have a lot of links
> > > between them, but are difficult for somebody like Winthrop who played
> > > mostly bad competition all year. Of course, that doesn't mean that
> > > the ELO-style ratings are necessarily better. Both sets would have
> > > linkage problems.
>
> > You might be right about Wintrop being hard to rank. But I am not
> > sure its a linkage problem. Might be that margin of victory (MOV) is
> > less predictive against all those weak teams. After about MOV>15
> > there is not much motivation for going higher.
>
> > Before the tourney they did not beat any ranked teams. They did give
> > Wisconsin a run for their money, but they did not challenge the others.
>
> They didn't beat any ranked teams. But they won at Missouri State and
> at Mississippi State, two teams that had good predictive ratings (40th
> and 38th according to the LRMC). Winthrop was only 68th in LRMC, and
> I suspect that somewhat narrow wins over their own conference was the
> thing that brought them down there. However, Winthrop actually fairs
> slightly worse (71st) in LRMC capped, while they move up to 41st in
> LRMC(0).- Hide quoted text -
>
> - Show quoted text -

They are 14 in Sagarin ELO right now. Guess they were pretty high
back
on March 11 to compared with LRMC(0).

I wonder if LRMC(0) is any better than Sagarin ELO?
 >> Stay informed about: Winthrop 

"stat monkey" <tadamsmar.RemoveThis@yahoo.com> wrote

> > is demonstrating why I'm uncomfortable with just using predictive
> > ratings to set the tournament seedings. Across the predictive
> > ratings, the ranked in the 50s and 60s. Among ELO-style ratings, they
> > ranked in the single digits and teens. I think predictive ratings
> > probably work well for the major conferences who have a lot of links
> > between them, but are difficult for somebody like Winthrop who played
> > mostly bad competition all year. Of course, that doesn't mean that
> > the ELO-style ratings are necessarily better. Both sets would have
> > linkage problems.
>
> You might be right about Wintrop being hard to rank. But I am not
> sure its a linkage problem. Might be that margin of victory (MOV) is
> less predictive against all those weak teams. After about MOV>15
> there is not much motivation for going higher.
>
> Before the tourney they did not beat any ranked teams. They did give
> Wisconsin a run for their money, but they did not challenge the others.

They were leading UNC till midway through the
second half. Are you trolling?

--Tedward
 >> Stay informed about: Winthrop 

On Mar 22, 6:02 pm, "Edward M. Kennedy" <d....TakeThisOut@wox.com> wrote:
> "stat monkey" <tadams....TakeThisOut@yahoo.com> wrote
>
>
>
>
>
> > > is demonstrating why I'm uncomfortable with just using predictive
> > > ratings to set the tournament seedings. Across the predictive
> > > ratings, the ranked in the 50s and 60s. Among ELO-style ratings, they
> > > ranked in the single digits and teens. I think predictive ratings
> > > probably work well for the major conferences who have a lot of links
> > > between them, but are difficult for somebody like Winthrop who played
> > > mostly bad competition all year. Of course, that doesn't mean that
> > > the ELO-style ratings are necessarily better. Both sets would have
> > > linkage problems.
>
> > You might be right about Wintrop being hard to rank. But I am not
> > sure its a linkage problem. Might be that margin of victory (MOV) is
> > less predictive against all those weak teams. After about MOV>15
> > there is not much motivation for going higher.
>
> > Before the tourney they did not beat any ranked teams. They did give
> > Wisconsin a run for their money, but they did not challenge the others.
>
> They were leading UNC till midway through the
> second half. Are you trolling?
>
> --Tedward- Hide quoted text -
>
> - Show quoted text -

You're right, 7 point loss, not bad. That was suppose to be a warm-up
game.
 >> Stay informed about: Winthrop