Loading ...
Sorry, an error occurred while loading the content.

Re: [hockeydisk] 2001-2002 disk

Expand Messages
  • D. Atkinson
    ... Yep. I think Jeff and I are up for at least one more season of this. ... If we can get an earlier start this year (next 3 weeks) on the ratings, and have
    Message 1 of 25 , Mar 24, 2002
    • 0 Attachment
      Daniel Robitaille wrote:

      > Today I was scanning the yahoogroups archives and came up
      > with these release dates for previous disks:
      >
      > July 10/1999
      > August 13/2000
      > Sept 15/2001
      >
      > See the trend? :)
      >
      > I know it's still very early, but here are my 2 questions:
      >
      > *) is this group planning to release a disk like in previous years?
      >

      Yep. I think Jeff and I are up for at least one more season of this.

      > *) and if yes, in light of previous years experience, is the later
      > date a more realistic expectation for the end users than the early
      > one?

      If we can get an earlier start this year (next 3 weeks) on the ratings, and
      have
      the raters respond somewhat quicker, I think that Aug 1 would be a good
      target date. One of the problems in the past is that Jeff and I were super
      busy with our real jobs. My main problem in the past was that I was
      on business travel 2 weeks of every month. Since 9/11, however, I have
      been on only one trip, and I am not quite as busy as past years. I will
      have more time early on to dedicate to getting the disk out. However, good
      use of my time depends on the raters being quick about things. I have
      no idea how Jeff's schedule is stacking up this year, and his programming
      work up front is critical.
      Let's start organizing to hit the ground running the day the NHL season ends
      (so we have a complete list) and shoot for August 1.

      Dave
    • D. Atkinson
      ... The problem with using alot of raters is that the fringe players tend to get averaged out. Alot of raters won t give a 1 rating for anyone except the
      Message 2 of 25 , Mar 24, 2002
      • 0 Attachment
        Eymond Toupin wrote:

        > Hi,
        >
        > I don't contribute very often to the discussion
        > but I like to follow what is happening with
        > APBA users. I participate in two APBA leagues
        > with twelve teams each. We really enjoy
        > the friendly competition and rely on the numbers
        > generated. Although we may not always agree with
        > all the ratings generated, we respect them
        > because of the efforts of so many impartial
        > contributors. I agree with the comments of D.G.
        > In both the leagues I participate in, players
        > with less than 30 games played are not eligible
        > for play. In addition the roster size is
        > limited to 35 and 40 players per team, which is
        > about half? of the number of players in the NHL in
        > a given season. Therefore, for us the ratings
        > of marginal players is pretty irrelevant.
        >
        > I know what you are going to say, the performance
        > of the important players will be affected if the
        > ratings of fringe players is less than accurate.
        > For the most part however, fringe players (half the
        > NHL these days!) get 2's and 3's for most
        > of the "important" skills like defensive rating,
        > passing and manoeuvering and the odd four and five
        > for physical play and intimidation. What's more,
        > they don't score.
        >
        > Anyhow, I don't want to give the impression I am
        > complaining. Like most owners of APBA team I
        > am anxious to see the "numbers". I remember someone's
        > comments last year suggesting that "good owners"
        > shouldn't need the numbers to make trade and
        > drafting decisions. However there are always good and
        > bad surprises with respect to the "numbers" (depending
        > on your perspective).
        >

        The problem with using alot of raters is that the fringe players
        tend to get averaged out. Alot of raters won't give a 1 rating
        for anyone except the absolute worst 3 or 4 players. That shouldn't
        be the case. The ratings should follow a bell curve, with 3's dominating,
        and things rolling off from there. There should be as roughly as many
        5's as 1's. Also, the ratings should be a snapshot of that season, meaning
        a very bad defensive player should get a 1, not a 2 based on "there have been
        worse defensive players in NHL history". So, my advice to all those
        raters out there is to spread some 1's around where they are deserved.
        If you give out 10 DEF 5 ratings, and go back and look and see that
        you only gave out 1 DEF 1 ratings, make another run through and
        add a few ones. And please, don't heap all the bad DEF ratings on
        the goons. Granted, these guys aren't defensive whizzes (usually due
        to bad skating), but nearly all of them try hard at D, and often get
        things done on sheer size and strength, as opposed to the floaters
        skating away from their zone looking for the breakaway.
      • D. Atkinson
        ... I think there is still demand, based on the messages in this group and some emails I ve received. Thus, we ll do it again for at least another year. ...
        Message 3 of 25 , Mar 24, 2002
        • 0 Attachment
          kenneth danila wrote:

          > I also hope there's still a demand. As a commish of one APBA league and a
          > member of 2 others, I'd be sad to see the production stop.

          I think there is still demand, based on the messages in this group and some
          emails I've received. Thus, we'll do it again for at least another year.

          > Since I believe it's never too early to start thinking about these things,
          > I'd like to get a few questions out there.
          >
          > - What have been the major sticking points to an easy release in the past?
          >

          Two items come to mind....the busy schedules of Jeff and myself and
          raters taking too long to turn in their reviews. From my end, I should
          be less busy than past years, and thus more available to keep the
          disk moving along. And, if we get an early review start and everyone
          does their ratings fairly quickly, that should fix that issue.

          > - What can members of the community do other than submit ratings? (i haven't
          > done this in the past, as i don't feel i see enough games to give an
          > accurate set of ratings.)

          Too many cooks can spoil the broth. I think the community can aim
          to do the best it can on the reviews, as that is probably 80% of
          the disk. The last 20% is testing, tweaking, testing, tweaking, ad nauseum.

          > - What can be done as far as tasking/project management to help everyone
          > working on the disk out? (the way i see it, this is a major task with tons
          > of man-hours, and maybe some project management software/collaboration
          > software and a "project manager" would help everyone out.)

          Jeff and I have done this for I think 3 years now, and I don't see any issues
          in doing it again nearly the same way. If anyone has any great new ideas,
          let us know (Derrick Underwood had a wonderful pass rating analysis
          that he did last year that we'll use to tweak this year if Derrick is up for
          it). But in the end, other than a few adjustments like Derrick's or Bruce
          Carriker on the faceoffs, the disk quality is mainly due to the quality of
          the reviews. I implore the community to rate only what you know. The
          problem with rating players you don't know is that you tend to gravitate
          toward the middle, i.e. 2's and 3's. For example, let's take a hypothetical
          rookie who is rated by 4 reviewers. Two of the reviewers see him play
          every night, and definitely know he is horrid defensively, and give him 1's.
          However, 2 other reviewers guess, one gives a 2 and one gives a 3. Well,
          in this case, the player ends up with a 2, even though he doesn't deserve it.
          This happens more than you might think....we have examples of some
          players getting a 4 in a rating from one reviewer and a 1 from another. This
          kind of discrepancy tells me that a reviewer was guessing. One of the flags
          that I have been using is reviewer standard deviation.....I can bring it up
          easily in Access (where I query the review database) and it tells me if
          there has been too wide of a spread in the reviews. I will often seek extra
          opinions in these cases. However, the earlier we get started, the better we
          can deal with these issues. Last year, the yelling for the disk was getting loud

          enough that we almost ignored some of these problems. We need the
          time to work through it all.

          > The disk is geat and seems to keep getting better year after year. I'd hate
          > to see it stop now.

          Thanks for the kind words. We hope to have even a better disk this year.

          Dave
        • D. Atkinson
          D. G. wrote: Hi Jeff et al. Now that I m on a roll, and dealing with goaltenders, I m not sure running sims necessarily helps to give back-up goaltenders an
          Message 4 of 25 , Mar 24, 2002
          • 0 Attachment
            "D. G." wrote:
            Hi Jeff et al.

            Now that I'm on a roll, and dealing with goaltenders, I'm not sure running
            sims necessarily helps to give back-up goaltenders an accurate rating. For
            instance, in our draft league this year, one of our teams has both Belfour
            and Turco. Belfour was the starter for Dallas last year and Turco the backup
            but Turco is the starter for the team in our league because his save
            percentage is .925 vs. Belfour's .905 (and his hidden rating is better).
            Gee, was Dallas that dumb to play the wrong goalie for most of the year?
            Well no! Teams usually know who the best goalie is and they play the better
            goalie. When they play their back-up to give their starter a rest, do they
            start him against the Colorado's and Detroit's of the league? No, of course
            not. They give them the starts against the Tampa Bay's and Nashville's of
            the league. I could have a save percentage of .925 against those teams! (Not
            really, but I think you know where I'm going with this!) When the sims are
            run, I'm assuming the back-up goalies get an even distribution of good teams
            and bad teams, whereas in real life they almost always play against the bad
            teams. If Turco had a .900 save percentage card but a manual coach played
            him only against poor offensive teams, his save percentage potentially would
            come close to .925, and Belfour would still have the better card and would
            not be sitting on the bench in a draft league in favour of Turco. It would
            seem to me some subjectivity should be used in rating the goalies so back-up
            goalies are not better than the starters in front of them even if their
            actual stats are better than the starter's.
             

            Jeff and I have discussed this issue in detail, and for the past 2 years have been
            looking back through which teams the backups played against. If I remember
            correctly (Jeff can correct me) we only had 1 or 2 goalies in the past 2 years
            which needed adjustments due to most games being played against bad teams.
                        Dave

             
          • Derrick Underwood
            I thought I should give an update of where my work on the disk is... NHL official stats: NHL.com has changed some of their formatting on their web site (they
            Message 5 of 25 , Mar 24, 2002
            • 0 Attachment
              I thought I should give an update of where my work on the disk is...

              NHL official stats: NHL.com has changed some of their formatting on their
              web site (they seem to do it every month or so), so I need to adjust my
              application that harvests their stats at the end of the season. I should
              have that completed within a week or so of the close of the NHL regular
              season, assuming they don't do something to shut me out.

              Pass ratings analysis: I had a goal last year of figuring out how we could
              reduce the subjectivity of the disk using actual NHL stats. To make a long
              story short, I thought forward's pass ratings should be very directly
              related to assists per minute played - and have felt that too many good
              players were getting 3 pass, and too many weak players were getting 3's. I
              know lot's of guys will say that there are guys that get assists from shots,
              but I will deal with that in a bit. My approach was simple - check to see
              if assists directly correlated with the pass ratings, and if so, show how we
              can implement it. I analyzed APBA disks from the past 6 or so seasons.
              There was a huge correlation. And, interestingly, it increased dramatically
              when the disk team took over. I think last year, we just saw 8 or 9 funny
              ratings. So even though there is bound to be dissenting opinions, the
              rater's perceptions were that a high assist guy was a good passer. I have
              built an analysis that set's up thresholds that maintains the distribution
              of each pass ratings. When the final stats come out, I will re-create the
              thresholds based on the new distribution of assists. When I get time, I
              will see if there is a way of highlighting any high SOG guys to give us a
              check to make sure this doesn't become a problem. I have been pretty busy
              at work, so my latest analysis is behind schedule... But I hope to have
              some time in the next month to get that figured out. I will work on the
              same thing for defensemen, but I suspect assists are not as strongly related
              to their pass ratings, and probably shouldn't be.

              Derrick
            • Jeff Kraus
              Dave, et. al., Meant to reply to some of the threads out there, but mid-March to late April is my busiest time of the year in my real job so I m a little
              Message 6 of 25 , Mar 24, 2002
              • 0 Attachment

                Dave, et. al.,
                Meant to reply to some of the threads out there, but mid-March to late April is my busiest time of the year in my "real" job so I'm a little behind on my e-mail.  I think there were three things I wanted to respond to...

                > I implore the community to rate only what you know.

                I'll second that!  I limit myself to rating the one team I get to see more than 10 times a year (the Caps), and even then I'll disqualify myself from rating some of the guys I didn't see much if my ratings are out of line with guys who have established themselves as league-wide "experts".

                > If I remember correctly (Jeff can correct me) we only had 1
                > or 2 goalies in the past 2 years which needed adjustments due
                > to most games being played against bad teams.

                Yeah, two seasons ago I was thinking the same thing, that back-ups get skewed ratings because they always play against crummy teams.  I spent hours going through box scores and separating their performance against the top third of the league, the middle third, and the bottom third, doing the best I could to figure their save percentage against each group.  I think I did it for any goalie who played 20 games or less. 

                I was stunned to find out that while many coaches did, in fact, sit their star against Nashville and Tampa Bay, an equal number of coaches actually sacrificed their back-up to lost causes -- games at Detroit, Colorado, Philly, etc.  And it wasn't just the bad teams.  I think Robby Ftorek actually sat Brodeur in such games.  Anyway, it turned out guys like Turco and Billington actually had *better* save percentages against the top teams.  I still think there's a bias there -- maybe they played a defense-heavy scheme when the back-up was in -- but you can only do so much guessing before you make a farce of the ratings.  I think we wound  up tweaking 3 back-ups down a little, and that was an awful lot of time to spend on such an inconsequential result that I didn't bother with it last year.

                If someone would like to pick up that ball and run with it, feel free.  We can let you know the best way to go about it.



                Get more from the Web. FREE MSN Explorer download : http://explorer.msn.com
              • D. Atkinson
                Jeff Kraus wrote: Dave, et. al., Meant to reply to some of the threads out there, but mid-March to late April is my busiest time of the year in my real job
                Message 7 of 25 , Mar 24, 2002
                • 0 Attachment
                  Jeff Kraus wrote:
                   

                  Dave, et. al.,
                  Meant to reply to some of the threads out there, but mid-March to late April is my busiest time of the year in my "real" job so I'm a little behind on my e-mail.  I think there were three things I wanted to respond to...

                  > I implore the community to rate only what you know.

                  I'll second that!  I limit myself to rating the one team I get to see more than 10 times a year (the Caps), and even then I'll disqualify myself from rating some of the guys I didn't see much if my ratings are out of line with guys who have established themselves as league-wide "experts".

                  > If I remember correctly (Jeff can correct me) we only had 1
                  > or 2 goalies in the past 2 years which needed adjustments due
                  > to most games being played against bad teams.

                  Yeah, two seasons ago I was thinking the same thing, that back-ups get skewed ratings because they always play against crummy teams.  I spent hours going through box scores and separating their performance against the top third of the league, the middle third, and the bottom third, doing the best I could to figure their save percentage against each group.  I think I did it for any goalie who played 20 games or less.

                  I was stunned to find out that while many coaches did, in fact, sit their star against Nashville and Tampa Bay, an equal number of coaches actually sacrificed their back-up to lost causes -- games at Detroit, Colorado, Philly, etc.  And it wasn't just the bad teams.  I think Robby Ftorek actually sat Brodeur in such games.  Anyway, it turned out guys like Turco and Billington actually had *better* save percentages against the top teams.  I still think there's a bias there -- maybe they played a defense-heavy scheme when the back-up was in -- but you can only do so much guessing before you make a farce of the ratings.  I think we wound  up tweaking 3 back-ups down a little, and that was an awful lot of time to spend on such an inconsequential result that I didn't bother with it last year.

                  If someone would like to pick up that ball and run with it, feel free.  We can let you know the best way to go about it.

                  And what was the third thing? :)
                   
                • Jeff Kraus
                  Oh yeah, the third thing. Hang on, this is one of my passions so it takes some explaining. I lost the original note, but somebody was asking about the
                  Message 8 of 25 , Mar 24, 2002
                  • 0 Attachment

                    Oh yeah, the third thing.  Hang on, this is one of my passions so it takes some explaining.  <g>

                    I lost the original note, but somebody was asking about the ratings tweaks to match real-life performance and the danger of skewing league stats by giving players "ratings they don't deserve" (I think those were the words, sorry if I misquoted).

                    What we're trying to do by tweaking is to give the players ratings they *do* deserve.  If we run 100 simulated seasons and, say a scorer, is consistently taking 30% more shots than he did in real life, then something in his teammates ratings is influencing the player in question. 

                    My favorite example is always Kip Miller.  Here's a guy who had a rather meaningless career, who suddenly became Jaromir Jagr's left wing and just happened to score 20 goals that year.  How much of those 20 goals were because Miller suddenly found a scoring touch, and how many were because he had the league's best playmaker feeding him pucks?  Well, here's the thing: the APBA game engine only sees a player's ratings *isolated from their normal environment*.  In other words, to the game engine, this guy was surrounded by 3's in every category.  Put him on an APBA line with, say, Chris Gratton and Bill Muckalt - two very ordinary playmakers - and he'll score 20 goals with his default ratings.  Put him with Jagr in APBA, and he's liable to score 30, because Jagr's 5 or 6 PAS rating will lift his scoring totals.  It shouldn't.  It was Jagr's 5 or 6 PAS that *got* him his real life scoring ratings.  Putting him with lesser players should actually *drop* his totals to maybe 12-15 goals.

                    Here's a real life APBA example.  In the OTHL's first season, based on the 1993-94 disk, we started with actual NHL teams.  The guy whose team was based on the Rangers had Messier and Graves on his first line.  Teamed with Messier's 5 PAS, Graves scored an unbelievable 92 goals.  Obviously, that should never have happened.  Graves's scoring ratings were based on the fact he played with Messier in real life.  That was his normal environment.  He should not have the added benefit of having his ability further boosted by Messier's APBA ratings.  The same thing happend the following year, when Sergei Fedorov boosted Ray Sheppard to 70+ goals.  I think you'd have to agree that the ratings tweaks we've done on the recent disks have kept these embarassing abnormalities to a minimum. 

                    I think we can still do a little better on the goalies, but they are much tougher to normalize because their secondary ratings (PAS, SKT, SPD) seem to do funky things to their performance.  In fact, I wonder if we'd be better off giving all goalies straight 3's and see how they turn out during testing.

                    Now, I *do* have to admit that by tweaking to adjust a player to his normal environement, we're assuming the APBA engine is using the player the same way he was used in real life.  That's a tough assumption, but it's all we can do.  Like I said, we've seen a lot more realistic results by tweaking than by not.

                    In a truly *perfect* world, and one day I'll actually sit down and write a 10-pager on this subject, the tweaks cover not a single season but a sliding average of N seasons, where N varies depending on the player.  

                    Say we have a Player X, who has established the ability to score 15-20 goals per season, and suddenly one year he busts one year and scores 40.  Just for fun, let's call him "Bob Kudelski".  Clearly, that 40-goal season is an aberration, the upper limit of his ability.  But when you isolate it to ONE season of APBA, that 40 isn't his upper limit, it's his norm.  Run him through 100 APBA seasons and he'll have his share of 40's, but he'll also have a bunch of 30-goal seasons, a bunch of 50-goal seasons, and surely he'll have a few 20's and 60's.  Bob Kudelski - 60 goals?  Insane!  But it'll happen, just like Adam Graves's 92. 

                    In my world, Kudelski would still have the ratings of a guy who scored about 20.  Maybe there's only a 10% chance he'll score 40, and a 10% chance he only scores 5.  But most of the time he'll score 15-25.  A 40-goal season for him would be a fluke, just as it was in real life.

                    Jeff



                    Get more from the Web. FREE MSN Explorer download : http://explorer.msn.com
                  • D. Atkinson
                    ... I m glad Jeff took the time to write this....it spares me writing essentially the same thing in response to some of the recent posts. I thoroughly stand by
                    Message 9 of 25 , Mar 24, 2002
                    • 0 Attachment
                      Jeff Kraus wrote:

                      >
                      >
                      > Oh yeah, the third thing. Hang on, this is one of my passions so it takes
                      > some explaining. <g>
                      >
                      > I lost the original note, but somebody was asking about the ratings tweaks
                      > to match real-life performance and the danger of skewing league stats by
                      > giving players "ratings they don't deserve" (I think those were the words,
                      > sorry if I misquoted).
                      >
                      > What we're trying to do by tweaking is to give the players ratings they *do*
                      > deserve. If we run 100 simulated seasons and, say a scorer, is consistently
                      > taking 30% more shots than he did in real life, then something in his
                      > teammates ratings is influencing the player in question.
                      >
                      > My favorite example is always Kip Miller. Here's a guy who had a rather
                      > meaningless career, who suddenly became Jaromir Jagr's left wing and just
                      > happened to score 20 goals that year. How much of those 20 goals were
                      > because Miller suddenly found a scoring touch, and how many were because he
                      > had the league's best playmaker feeding him pucks? Well, here's the thing:
                      > the APBA game engine only sees a player's ratings *isolated from their
                      > normal environment*. In other words, to the game engine, this guy was
                      > surrounded by 3's in every category. Put him on an APBA line with, say,
                      > Chris Gratton and Bill Muckalt - two very ordinary playmakers - and he'll
                      > score 20 goals with his default ratings. Put him with Jagr in APBA, and
                      > he's liable to score 30, because Jagr's 5 or 6 PAS rating will lift his
                      > scoring totals. It shouldn't. It was Jagr's 5 or 6 PAS that *got* him his
                      > real life scoring ratings. Putting him with lesser players should actually
                      > *drop* his totals to maybe 12-15 goals.
                      >
                      > Here's a real life APBA example. In the OTHL's first season, based on the
                      > 1993-94 disk, we started with actual NHL teams. The guy whose team was
                      > based on the Rangers had Messier and Graves on his first line. Teamed with
                      > Messier's 5 PAS, Graves scored an unbelievable 92 goals. Obviously, that
                      > should never have happened. Graves's scoring ratings were based on the fact
                      > he played with Messier in real life. That was his normal environment. He
                      > should not have the added benefit of having his ability further boosted by
                      > Messier's APBA ratings. The same thing happend the following year, when
                      > Sergei Fedorov boosted Ray Sheppard to 70+ goals. I think you'd have to
                      > agree that the ratings tweaks we've done on the recent disks have kept these
                      > embarassing abnormalities to a minimum.
                      >
                      > I think we can still do a little better on the goalies, but they are much
                      > tougher to normalize because their secondary ratings (PAS, SKT, SPD) seem to
                      > do funky things to their performance. In fact, I wonder if we'd be better
                      > off giving all goalies straight 3's and see how they turn out during
                      > testing.
                      >
                      > Now, I *do* have to admit that by tweaking to adjust a player to his normal
                      > environement, we're assuming the APBA engine is using the player the same
                      > way he was used in real life. That's a tough assumption, but it's all we
                      > can do. Like I said, we've seen a lot more realistic results by tweaking
                      > than by not.
                      >
                      > In a truly *perfect* world, and one day I'll actually sit down and write a
                      > 10-pager on this subject, the tweaks cover not a single season but a sliding
                      > average of N seasons, where N varies depending on the player.
                      >
                      > Say we have a Player X, who has established the ability to score 15-20 goals
                      > per season, and suddenly one year he busts one year and scores 40. Just for
                      > fun, let's call him "Bob Kudelski". Clearly, that 40-goal season is an
                      > aberration, the upper limit of his ability. But when you isolate it to ONE
                      > season of APBA, that 40 isn't his upper limit, it's his norm. Run him
                      > through 100 APBA seasons and he'll have his share of 40's, but he'll also
                      > have a bunch of 30-goal seasons, a bunch of 50-goal seasons, and surely
                      > he'll have a few 20's and 60's. Bob Kudelski - 60 goals? Insane! But
                      > it'll happen, just like Adam Graves's 92.
                      >
                      > In my world, Kudelski would still have the ratings of a guy who scored about
                      > 20. Maybe there's only a 10% chance he'll score 40, and a 10% chance he
                      > only scores 5. But most of the time he'll score 15-25. A 40-goal season
                      > for him would be a fluke, just as it was in real life.
                      >
                      > Jeff

                      I'm glad Jeff took the time to write this....it spares me writing essentially
                      the same
                      thing in response to some of the recent posts.
                      I thoroughly stand by tweaking to NHL performance levels. That's the only
                      real data we have anyway. We can argue all we want about how this isn't the
                      way it should be for draft leagues, but without throwing the stats away and
                      going totally objective, it is the only handle we have on player performance.
                      If someone wants to develop a progressive system which covers 5 years of
                      past performance tied to linemates, be my guest. I think our approach is the
                      best
                      we can do given the information we have.

                      Dave
                    • Hank Stalica
                      Additional Raters... Has anyone considered using newsgroups as a source of additional raters? Just post an appropriate template to each newsgroup and see what
                      Message 10 of 25 , Mar 24, 2002
                      • 0 Attachment
                        Additional Raters...

                        Has anyone considered using newsgroups as a source of additional raters? Just post an appropriate template to each newsgroup and see what happens?


                        ________________________________________________________________
                        There's only one basic human right, the right to do as you damn well please.
                        And with it comes the only basic human duty, the duty to take the consequences.
                      • D. Atkinson
                        ... The homer factor in the newsgroups is unbelievably high. It is truly scary how deluded many of those people are concerning THEIR team. Always a Cup
                        Message 11 of 25 , Mar 25, 2002
                        • 0 Attachment
                          Hank Stalica wrote:

                          > Additional Raters...
                          >
                          > Has anyone considered using newsgroups as a source of additional raters? Just post an appropriate template to each newsgroup and see what happens.

                          The homer factor in the newsgroups is unbelievably high. It is truly scary how
                          deluded many of those people are concerning THEIR team. Always a Cup
                          contender and half their team are Hart trophy candidates.
                        • Ed Hebscher
                          Speaking of Newsgroups, I am looking for some help. I belong to a rotisserie hockey pool (Toronto based) and we are looking to add 2-3 new owners for next
                          Message 12 of 25 , Mar 25, 2002
                          • 0 Attachment
                            Speaking of Newsgroups, I am looking for some help.
                            I belong to a rotisserie hockey pool (Toronto based) and we are looking to
                            add 2-3 new owners for next season. Is anyone nearby and interested or does
                            anyone know a good newsgroup to post to?

                            Thanks,

                            Ed

                            -----Original Message-----
                            From: D. Atkinson [mailto:bigdog@...]
                            Sent: March 25, 2002 8:58 AM
                            To: hockeydisk@yahoogroups.com
                            Subject: Re: [hockeydisk] 2001-2002 disk


                            Hank Stalica wrote:

                            > Additional Raters...
                            >
                            > Has anyone considered using newsgroups as a source of additional raters?
                            Just post an appropriate template to each newsgroup and see what happens.

                            The homer factor in the newsgroups is unbelievably high. It is truly scary
                            how
                            deluded many of those people are concerning THEIR team. Always a Cup
                            contender and half their team are Hart trophy candidates.







                            Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
                          • D. Atkinson
                            ... alt.sports.hockey.fantasy
                            Message 13 of 25 , Mar 25, 2002
                            • 0 Attachment
                              Ed Hebscher wrote:

                              > Speaking of Newsgroups, I am looking for some help.
                              > I belong to a rotisserie hockey pool (Toronto based) and we are looking to
                              > add 2-3 new owners for next season. Is anyone nearby and interested or does
                              > anyone know a good newsgroup to post to?
                              >
                              > Thanks,
                              >
                              > Ed

                              alt.sports.hockey.fantasy
                            Your message has been successfully submitted and would be delivered to recipients shortly.