Subj:	Selection of 12 teams
Melanie writes:
>Mike, I understand what you are saying, and for the most part agree, but:
>1) who would decide who the best 12 teams were? 2) the idea is used in every
>sport, but there is always someone who is going to disagree with the
>selections.  3) what assurance would we have that the person(s) that select
>won't be biased?
 
I don't believe that the current system is that flawed that it needs to be
completely revamped.  Since a representative from each conference sits on
the selection committee, all teams are in some way represented (I'll get to
Independents in a minute).  And I believe that all the members of the
committee are above reproach when it comes to allegations of members ganging
up for or against certain teams or conferences.  Remember, when either the
selection or seeding of a team represented by one of the committee members
is being discussed, that member must leave the room and the discussion is
held among the three remaining members.
 
Melanie, I'm sure one of the reasons you're particularly interested in
this subject is that UAA and the other Independents have had no direct
representation on the selection committee, and I agree this is a problem.
If it were up to me, I would add a fifth member to the committee who would
represent the Independents.  After this season there will only be five, but
I still feel they should get representation.  Yet, as I said before, I think
the automatic bid was a quick and easy solution to a problem that resulted
in some deserving teams not being awarded bids to the tourney in favor of
Independents that were clearly not among the nation's 12 best teams.  In
the four years that the automatic bid has been in place, I think at least
two of those years the automatic bid resulted in an injustice.
 
I realize that prior to the automatic bid, Independents claimed they
were being slighted in favor of conference teams because all four committee
members are from conferences.  But prior to 1988, numbers show that no
Independents performed well enough to merit consideration for a bid.  It is
only in recent years that many Independents have begun playing schedules
that allow them to be considered among the better teams in the country,
should they also perform well in those games.  That was a major problem
for Independents in the past.
 
This brings us to another continued improvement that I think is necessary.
I think even most casual fans would agree that the process and resulting
selection/seedings for the 1991 tournament were among the smoothest and
fairest in history.  Not uncoincidentally, this is likely due to the fact
that the committee had at its disposal the most complete and unbiased set
of statistics used for comparing the teams that it has ever had.  This is
made possible by the retaining of certain politically unbiased individuals
to compile these statistics and provide them to the committee.  Of course,
it is then up to the committee to use these statistics in a fair manner,
but I believe the more facts that are available to the committee, the
easier their job is.  This needs to continue.  These statistics include
Independents as well as conference teams, too, so Independents are now given
as equal a chance as any other team when the statistics are given to the
committee.
 
In the past, many of the statistics the committee had were actually provided
to them by the individual teams.  Of course, they only provided statistics
that made their cause look impressive.  One team that shall go nameless
provided all sorts of ridiculous numbers such as their home record, etc.
(not Northern Michigan), but most of these numbers were not even part of
the selection criteria.  I feel confident now that while the committee
may still receive faxes or mailings containing this type of information,
it is at least getting a set of unbiased statistics that can accurately
show a team's performance against all of the selection criteria.
 
I don't believe any restrictions should be placed on the committee, as I
said before, so that they can more accurately select the 12 teams that
deserve most to play for the national championship.  Unfortunately we do
have some restrictions, such as tourney winners getting a bid and a
requirement to select at least two teams from each conference.  But among
the choices the committee does have, they should be provided with as much
unbiased information as possible pertaining to the established selection
criteria, and in recent years there have been great strides made towards
insuring that this will continue to take place.
 
>For example, I subscribe to Volleyball Monthly and there is always a list at
>the end of the magazine with the list of the top 20 high school teams.  After
>reading a couple months of the Letter to the Editor section, there are at the
>minimum of two letters a month from an upset person because such and such a
>team wasn't selected.  They didn't get anything from being in the list, so
>don't you think that someone would be even more upset if they were 'robbed'
>from the chance at the national title?
 
On the letters you talk about, I would bet that most if not all come from
people who are upset that *their* team is not included.  I am always wary
of complaints of this sort because these people are likely not to be as
unbiased.  Also, these silly high school polls that appear in places such
as USA Today are really meaningless because there is almost no way to
compare teams from different states who never play each other and have no
common opponents.  I put no stock in them.  But in selecting teams to
play in the NCAA hockey tournament, we can look at things like strength
of schedule, record against common opponents, etc.  Bill has already
summarized the extensive criteria that were used to select Cornell over
SLU in 1991, for example.
 
In addition, it has become very clear to me that when people are upset that
their team wasn't selected or seeded high enough, they invariably quote
only the statistics that support their team and ignore the others.  The
truth is that in many situations where two teams are being compared, some
of the statistics support one team and some support the other.  The job of
the committee is to look at *all* factors and make a decision based upon
all of them.  In 1990 we discussed on HOCKEY-L (then "college_hockey") a
letter written to The Hockey News by a Michigan fan who was upset that the
Wolverines weren't given a bid to the tourney, and he quoted only the facts
that supported him.  The problem was that there were many other factors that
didn't support him, and he never mentioned these.  But if you just look at
his letter, you come away tempted to agree with him.
 
I suggest that people curious about how the selection was done over the past
two years take a look at the list archives from March 1990 and 1991.  We
discussed all of the questions people had as to why certain teams were
selected and others weren't, and there's a lot of information there that
may help a somewhat enigmatic process become more clear.  I think it really
is a lot fairer than many people might think, and it is fairer now than
it has been in the past due in large part to the committee establishing
criteria and having a full set of statistics that show how the teams stack
up.
 
Here's an excerpt from mail Bill Fenwick sent to the list back in March
detailing an interview with Cornell AD Laing Kennedy, one of the members
of the committee, in which Kennedy discusses the selection criteria.
 
]The selection criteria, according to Kennedy, are win-loss percentage in
]Division I competition and "strength of schedule".  If there is no clear
]winner in these two areas, the committee then considers head-to-head
]results.  There is also a fourth criteria, which is record against "teams
]under consideration".  Kennedy said Cornell got the #6 seed over St. Law-
]rence due to the Big Red's 2-1 record against the Saints, as Cornell's
]winning percentage was better than St. Lawrence's (0.638 to 0.614) while the
]Saints had the stronger schedule (in addition, Cornell's record against
]teams under consideration was 4-3-1 while St. Lawrence went 2-9).
 
 
- mike