Deleted Posts from the USCF Forum: Please Post Them here

Please Post here what the USCF does what its members to see

oldtimer wrote:
I've sent emails to moderators 1, 2 and 4 and no response yet. (2 days
for some)



Your PM to me was dated 3:27 EDT today. Your post to this thread was less
than 10 minutes later!

Considering the harsh (obscene) comments I have received in my first month
of volunteer service, I have become very hesitant to remove anything but the
most blatant violations, and only after careful consideration. I could write
more, but I've been yelled at once before in this very thread for making
personal statements.


Next, start a thread Another Example of Biased Moderation Under New Policies
which is just like this one so far and I get the following response.

The above mentioned thread as well as the one titled "Specific Examples with
Problems of Moderators" have been pulled for FOC review. Please remember
that personal attacks are not allowed on this forum.

Quote:
Don't Attack Each Other: Because this is so important, we'll say it
again: Don't attack others. Personal attacks on others will not be
tolerated. Challenge others' points of view, record, statements, and
opinions, but do so respectfully and thoughtfully ... without insult and
personal attack. Harassing, offensive, vulgar, abusive, threatening, hateful
or bashing communications -- especially those aimed at sexual orientation,
gender, race, color, religious views, national origin, or disability -- will
not be permitted. Before you post a message that is intended to embarrass,
humiliate or harass another person or group of people, stop and THINK first!
Harassing communications are continued persistent attacks that are intended
to cause distress, embarrassment, unwanted attention or other discomfort.



Meanwhile Hillery is calling me stupid in another post but is that deleted.
No. This is the moderator that claims he only deletes the most vile posts.
Does anyone find my stuff vile. I think maybe moderator 4 is vile. At this
point not only does moderator4 begin to attack my intelligence in the forum
but so does kbachler, hillery, jonnybear and so we know the moderators are
actually working with some of the posters. I posted for quite some time and
people attacked me but not my intelligence. This was clearly a concerted
effort organized by moderator 4.

30 hours later Moderator finally responds to me
I responded So here's a good one. I start a topic. The superposters come
with completely off topic material and I request that it be removed. A new
thread is started but the comments are not removed from the old thread. So I
contact various moderators about this over several days. No one responds but
then moderator 4 finally says that he is not going to do anything because no
one has posted on the thread for 30 hours so lets have it die a natural
death. You see the moderators support only the superposters or are
manipulated by them. They destroy the thread and then the moderators say
well the thread is dead so we don't have to do anything about it. Of course,
that wouldn't be their response if it was one of their special people so
this is yet another example how the forum moderators completely fail to even
do the job they claim they are doing.

oldtimer wrote:
Could somebody please clean up the ratings/inflation thread. There is
an offtopic discussion of Fischer/morphy and apparently a new thread was
created but the discussion was not removed from the ratings/inflation thread



Considering that there have been no posts in either of the ratings/inflation
threads in the past 30 hours, I think I will try to let them die a natural
death.

What a sleaze. He locks the thread after I wait 30 hours for him to
respond. I start a new thread that has received over 40 new posts on the
same topic proving that moderator4 is not doing his job.



Your posts in the "Member Generated Posting Guidelines" is off topic. You
were politely warned about this by Mr. Channing as well.

I will let the FOC determine if your lengthy post about kbachler constitutes
a personal attack or not.


oldtimer wrote:
I believe we have already proven that the current moderators do not
work. Below is an example of the sort of the thing they delete so it's clear
we cannot find moderators that follow the rules so I suggest that we go back
to the way things were. I would like to see something different but it's
clear that moderation doesn't work.

Below is an example that moderator 4 was aware of and did nothing
about except to try and prevent me from posting about the topic by locking
the prior thread. I don't see how anyone can support the current system when
it's obvious the moderators are not following the rules.

-------------------------------
Oldtimer This theory rests upon many assumptions including the
assumption that younger players do not become stronger at the same rate
across generations.


No such assumption is required. I don't know how you came up with
this.

Insulting: Kbachler is not a professor as I have been. He simply
claims I'm incorrect without any evidence. Furthermore, I believe simply
logic of my post would lead any reasonable person to believe that I am
correct on that but this was also explained in detail to kbachler in the
prior the thread. Obviously, if younger players are unusually strong in this
generation middle-aged players ratings would go down.


Oldtimer Several posters proposed that in the current era younger
players have become much stronger and so they are driving the older players
ratings down and so the methodology that the ratings committee has been
using is unsound for the current era.


Not at all. The point is simply that it's been shown on several
occasions and anecdotally that younger players tend to have more volatile
ratings, and that they also tend to improve more rapidly. Therefore a group
with more younger players tends to suffer more deflation because improvement
in general causes deflation in the Elo system.

Insulting. "Not at all" kbachler ignores what the other players said.
He gives his argument without acknowledging what was previously said


Oldtimer said

"An alternative approach to determine rating inflation would be to
compare the abilities of older players with similar ratings to new players.
After the opening is played, you could look at the average number of
blunders made across a sample of games from different eras. A computer could
be used to identify blunders of greater than .50 centapawns, for example. If
the # of blunders is higher today than for similarly rated players then you
have a case for inflation. On another thread, this idea was suggested and
Kenneth Sloan claimed you would need millions of games collected over a
30-year study to do an appropriate analysis. Obviously, this is not true but
you would in my opinion have to analyze at least 1,000 games. Of course,
that's not that big a deal with today's computers. If this study was done, I
believe it would be fairly obvious that over a 30-year period some rating
inflation has occurred. I recall 2400 rated Seirawan winning many major U.S.
tournaments and I strongly suspect that a young Seirawan would do the same
today but of course then his rating would quickly rise to over 2700. Of
course, I admit I could be wrong about that. I recall that Jim Tarjan's USCF
rating got over 2700 but he did that by rarely losing a game. He was quite
dominant in California for a time. When Dzhindi won the U.S. championship in
1983 he was not even 2500."


Kbachler Basically, your last paragraph is nonsense.
Insulting

Oltimer Evidence to support the strength of younger players is a
dramatic increase GM titles, IM titles, FIDE and USCF in ratings. There are
only really only two possibilities to account for such a dramatic increase.
1. Improved strength of young players 2. Rating inflation kbachler said
quoted my next sentence which was oldtimer If hypothesis 1 is correct than
the USCF demographic studies are unsound.
Hypothesis 2 is neither correct nor possible.

There is no need to address the rest of this post. It's completely
off-point.

Depending on what your first hypothesis means - since it is stated
poorly - it is either irrelevant or proves our point.
Now he's saying the last part of the post is too bad to address the
last part of the post. After claiming that

oldtimer wrote:
I would say most of the recent posts suggest a confusion about how to
define rating inflation.


kbachler wrote: Generally, that seems to be true primarily of your
posts.

Insulting

Oldtimer said In particular, the USCF argues you cannot look at
averages ratings across time and claim that inflation has occurred.


kbachler This isn't just a USCF argument. People knowledgeable about
ratings already know this as a given

Insulting kbachler is implying that because I stated the USCF argument
that I somehow don't understand that it's an argument that other people use.
He cites no evidence to support that claim.

Kbachler goes on to say "Inflation and deflation have causes. The
immediate cause of deflation is that a player's (or more) performance
improves. The immediate cause of inflation is that a player's performance
degrades.

If you are going to argue that a pool is inflated or deflated, you
should be able to point to a significant segment of the pool and describe
where the above is happening."

insulting. I was talking about young players improving. That is, I did
exactly what he accuses me of not doing.


Kbachler makes the same point again. "USCF doesn't define inflation or
deflation differently from other knowledgeable entities. Lay people often
get confused and mis-define it."

Insulting. Kbahcler knows I'm a retired college professor with
expertise in statistics and psychometrics. I'm not a lay person.

Oldtimer said "but that doesn't mean the entire pool is deflated.


kbachler When over half of the pool are youth, with volatile ratings
and a lot of improvement it does. Every time a youth improves and deflates
an adult (and other youth for that matter) they have a ripple effect to the
next person. To argue otherwise with these demographics is absurd on its
face.

Insulting: He's calls my argument absurd. He says nothing about
whether young players are deflated or inflated and even acknowledged that.
Since they are also the largest part of the pool according to kibachler it
stands to reason that the entire pool's standing on inflation/deflation
depends largely on the younger pool and not the older pool.

Oldtimer The problem is your methodology doesn't address the question.


It does. The problem is that you still don't understand it.
What makes you think the only portion of the pool examined is adults
in a specific age? That's just the portion simplest to explain, and you
aren't grasping that yet. Their were several adjustments made to the rating
system to correct for ratings integrity. The one considered most important,
because we were losing adult members, along with the simplest one, is the
one mentioned.
Insulting First he again claims that a college professor cannot
understand his thinking. He add what makes you think the only portion of the
pool examined is adults in a specific age?

What made me think that was is that he previously agreed with me on
that point. In fact, it's the only time he agreed with me.

oldtimer wrote:
In contrast, USCF argues for deflation in the past 10 years. As far as
I can tell this is the result of looking at middle-aged male players whose
ratings are generally stable. This is the methodology they have used in
several studies in recent years. The theory is that if these players ratings
go down we have deflation and if they go up we have inflation.


kbachler Fairly on track.


Oldtimer: The maximum liklihood analyses suggest that inflation has
occurred.


kbachler Actually, it suggests that deflation occurred at most parts
of the system. insulting. I gave two references to this and both support
argue for inflation (Chessmetrics.com) is the primary reference. If kbachler
has another reference he doesn't give it.


oldtimer wrote:
The Glicko system gives extra points to players who rise rapidly. The
purpose is to address concerns about deflation. That is, it quickly puts the
stronger player at a better rating but if you have a whole group of stronger
players entering the system you get a great influx of rating points that
might overall lead to inflation.


kbachler There was no influx of stronger players. I don't know where
you keep getting that. The influx is at the low end of the system.

Moreover, the USCF system now has similar protocols and it is
currently reinflating the system.

Insulting. He basically states that there is no stronger players
without any evidence. Note I have already stated by evidence that there is
(e.g., a ten-fold increase in GM norms in the past 10 years). He then claims
the USCF has similar protocols as if I don't acknowledge that when I
explicitly am referring to the Glicko system.

Oldtimer: Of course, other posters should realize that the unhearalded
kbachler is criticizing me (an academic with many statistical publications)
for having a lack of statistical knowledge.


kbachler Right now I don't care about your credentials, I care that
you still aren't grasping simple concepts. I suggest that until you do, you
might be better off not leveling criticism.

Insulting

Oldtimer Inflation occurs because more points enter the system.


Insulting kbachler This single statement alone demonstrates you
haven't a clue.

.. At the VERY low end of the system I believe it was an issue,

First he takes the statement out of context in that I also refer to
other conditions that are required for inflation. He later acknowledges that
my point has some validity after accussing me of being clueless, which
demonstrates that he is liar. He goes on to accusse me of several things
that I didn't say and therefore are also lies.

Kbachler Simply ADDING players does not per se cause inflation.



Insulting. Kbachler- May I suggest at this point that you quit the
online dialogue, because your lack of fundamentals is a waste of time for
you and others. Write to the rating committee, and ask for an explanation of
their analysis. REad Elo's book. Get some fundamentals. In a prior post he
acknowledges that he knows I have read Elo's book.


George. In terms of your computer idea that's very similar to what I
initially proposed and Tanafl proposed in another thread. I believe your
basically correct. However, opening knowledge changes so what you'd have to
exclude that part in some what. Perhaps you could have a standard set of
middle and endgame positions that people could play against (A human test
suite). Kenneth Sloan claimed it was a bad idea because you would have to
the study constantly on all games at ICC for 30 years. I don't understand
his logic on that one. Another way to do it is to simply have the computer
analyze games from players within rating categories doing a blunder check in
the middle and endgame using fixed depth searches to control for processing
speed and using only existing tablebases at the beginning of your five year
period and perhaps an algorithm to extend the depth in endgame positions
where greater depth may be needed--Chess Assistant 9 has one. The analysis
should find approximately the same percentage of errors across ratings
categories in the end and middlegame as it did five years before if no
inflation/deflation has occurred. Of course, one can go back in time and
analyze old games from the past and do the same analysis to determine if
inflation has occurred in the past. P.S. There are all these coaches out
their for kids. Maybe they need some coaches for old people like me too.


kbachler The idea is fundamentally unsound. It assumes that rating X
today (say 1600) has the same knowledge as rating X in a prior era. That
assumption is false.

Ratings compare performance, not knowledge. Someone who maintains
exactly the same knowledge would decrease in strength relative to the pool
over time as general chess knowledge increases. I don't believe the pool has
ever been corrected for this "implicit inflation".

Insulting The idea is fundamentally unsound. Also, I acknowledged his
point in what I said and he's lying again by claiming that I didn't.
However, opening knowledge changes so what you'd have to exclude that part


oldtimer wrote:
Of course, it's true that opening knowledge improves over time but as
i said earlier you simply don't use moves that are book (at that time) for
your analysis. Has endgame play improved over the past 30 years for the
average player? In my opinion it's more likely it's gotten worse. Have their
been advances in the middlegame. A few, but not as many as a younger player
might think. For example, would a Capablanca get blown out knowledge of the
middlegame. Probably, the reverse is more likely. In fact, some have
suggested that is how Fischer was able to defeat the Russian jugernaut. That
is, he studied non-Russian greats while the Russians tended to only study
Russian greats. Of course, if you discovered that was a problem there are
ways around it. What probably hasn't changed since the dawn of chess is the
error rates. Actually, several senior citizen centers have asked me to teach
chess but I declined as I seldom teach and at the moment I feel very rusty
but I believe there is a market for senior chess, especially with all the
books out about how chess might help people with alzheimer's and other forms
of cognitive decline. If you see it that way, you might have some success.
Personally, I believe the reason the ratings committee does not participate
in this discussion is because they now realize they've been doing it wrong
in so far as using demographic analyses to determine whether there has been
rating inflation/deflation. Hopefully, they will have the courage to change
things. I did read Tannafl's idea on the other thread which Kenneth Sloan
pooh poohed but his arguments were unsound. He's an expert an computer
science and as far as I can tell has not published anything on methodology
or advanced statistics. Glickman and I of course have and Glickman has
admitted in his recent interview in chesslife that he doesn't know whether
there is inflation or deflation occurring and he unsure if his system is
working. he basically responded to administrator's who felt they were losing
middle-age membership because their ratings were going down against strong
young players. The new system gives bonus to people whose ratings are
improving quickly. This seems like the best system but I mentioned earlier
how this could have some negative side effects. Speed of processing
definitely slows with age in everyone and cognitive decline begins as early
as ago 40 for people who do not keep themselves in shape or are stressed.
This would definitely have a negative effect on one's rating with faster
time controls. Another interesting study would be to look at if the
middle-aged groups decline was related to the percentage of tournaments that
were moving to faster time controls. One has the data to look at it and if
one knows something about sampling theory you could do a quick and dirty
study looking at only about 300 people at two points in time.


k-bachler- I really suggest that you would be better off in a private
discussion with the rating committee, rather than spreading incorrect
information and untested theories to the average player. Besides, if you are
so certain that you are correct, then posts here are meaningless anyway. A
direct discussion with the rating committee would be much more useful for
you. If you are correct, it would be much more likely to effect change. And
if you are wrong, it would be much less likely to confuse the average
player.
_________________
Kevin Bachler

Insulting
Oldtimer, no one is afraid of your questions. Just annoyed. Please
stop posting misinformation.
_________________
Kevin Bachler

Repetious, Insulting

Kibachler You don't seem to understand what inflation is. Please go
back to the fundamental post earlier in this thread that explained inflation
and deflation.

Inflation presupposes that players are rated higher than their
performance indicates. The only way that increased knowledge creeps into the
situation is via increased performance. If the performance is increased,
then there isn't inflation.

Knowledge creep in the pool is not inflation.

Repetitious, Threatening insulting and a lie


What absurd statements? And what have I said that would concern the
moderators?

BTW, I already put in this thread why the computer approach doesn't
work, and the answer is fairly obvious.

Again, you would be better off if you learned the fundamentals and had
a direct discussion with the ratings committee.
_________________
Kevin Bachler

Reptitious, insulting

Todd Millier needs to take a chill pill.

And yet, when Susan Polgar called me dishonest she was not sanctioned, nor
was her post deleted. Rather, when I told her that she was in danger of my
suing her for defamation my post was deleted. Now, the FOC has ruled that
asserting ones legal rights is considered and attack and will be sanctioned
in the future. This is what the FOC sent me:

NOTIFICATION OF SCOPE OF AUG RULE ON THREATENING
COMMUNICATIONS

Date: March 23, 2007
To: Brian Lafferty

Your Post:35160 to the USCF Forums on Sun Mar 18, 2007 3:09 pm
(Subject: "A Follow Up Financial Question for Ms Polgar ") in the USCF
Issues Forum was in violation of the Acceptable Usage Guidelines for the
USCF Forum.

"Harassing, offensive, vulgar, abusive, threatening, hateful or bashing
communications -- especially those aimed at sexual orientation, gender,
race, color, religious views, national origin, or disability -- will not be
permitted." - Mike Nolan (Mon Feb 13, 2006 11:18 pm)
http://www.uschess.org/forums/viewtopic.php?t=702

No sanction is being applied at this time because we consider it possible
that you are not aware that legal threats are included.

Accordingly, the Forums Oversight Committee has chosen to inform you
of the situation and to notify you that, in the future, a sanction will
result
from the posting of a legal threat in the USCF forums.

Sincerely,
The Forums Oversight Committee



"Todd" wrote in message
...
Please Post here what the USCF does what its members to see

oldtimer wrote:
I've sent emails to moderators 1, 2 and 4 and no response yet. (2
days for some)



Your PM to me was dated 3:27 EDT today. Your post to this thread was less
than 10 minutes later!

Considering the harsh (obscene) comments I have received in my first month
of volunteer service, I have become very hesitant to remove anything but
the most blatant violations, and only after careful consideration. I could
write more, but I've been yelled at once before in this very thread for
making personal statements.


Next, start a thread Another Example of Biased Moderation Under New
Policies which is just like this one so far and I get the following
response.

The above mentioned thread as well as the one titled "Specific Examples
with Problems of Moderators" have been pulled for FOC review. Please
remember that personal attacks are not allowed on this forum.

Quote:
Don't Attack Each Other: Because this is so important, we'll say it
again: Don't attack others. Personal attacks on others will not be
tolerated. Challenge others' points of view, record, statements, and
opinions, but do so respectfully and thoughtfully ... without insult and
personal attack. Harassing, offensive, vulgar, abusive, threatening,
hateful or bashing communications -- especially those aimed at sexual
orientation, gender, race, color, religious views, national origin, or
disability -- will not be permitted. Before you post a message that is
intended to embarrass, humiliate or harass another person or group of
people, stop and THINK first! Harassing communications are continued
persistent attacks that are intended to cause distress, embarrassment,
unwanted attention or other discomfort.



Meanwhile Hillery is calling me stupid in another post but is that
deleted. No. This is the moderator that claims he only deletes the most
vile posts. Does anyone find my stuff vile. I think maybe moderator 4 is
vile. At this point not only does moderator4 begin to attack my
intelligence in the forum but so does kbachler, hillery, jonnybear and so
we know the moderators are actually working with some of the posters. I
posted for quite some time and people attacked me but not my intelligence.
This was clearly a concerted effort organized by moderator 4.

30 hours later Moderator finally responds to me
I responded So here's a good one. I start a topic. The superposters come
with completely off topic material and I request that it be removed. A new
thread is started but the comments are not removed from the old thread. So
I contact various moderators about this over several days. No one responds
but then moderator 4 finally says that he is not going to do anything
because no one has posted on the thread for 30 hours so lets have it die a
natural death. You see the moderators support only the superposters or are
manipulated by them. They destroy the thread and then the moderators say
well the thread is dead so we don't have to do anything about it. Of
course, that wouldn't be their response if it was one of their special
people so this is yet another example how the forum moderators completely
fail to even do the job they claim they are doing.

oldtimer wrote:
Could somebody please clean up the ratings/inflation thread. There is
an offtopic discussion of Fischer/morphy and apparently a new thread was
created but the discussion was not removed from the ratings/inflation
thread



Considering that there have been no posts in either of the
ratings/inflation threads in the past 30 hours, I think I will try to let
them die a natural death.

What a sleaze. He locks the thread after I wait 30 hours for him to
respond. I start a new thread that has received over 40 new posts on the
same topic proving that moderator4 is not doing his job.



Your posts in the "Member Generated Posting Guidelines" is off topic. You
were politely warned about this by Mr. Channing as well.

I will let the FOC determine if your lengthy post about kbachler
constitutes a personal attack or not.


oldtimer wrote:
I believe we have already proven that the current moderators do not
work. Below is an example of the sort of the thing they delete so it's
clear we cannot find moderators that follow the rules so I suggest that we
go back to the way things were. I would like to see something different
but it's clear that moderation doesn't work.

Below is an example that moderator 4 was aware of and did nothing
about except to try and prevent me from posting about the topic by locking
the prior thread. I don't see how anyone can support the current system
when it's obvious the moderators are not following the rules.

-------------------------------
Oldtimer This theory rests upon many assumptions including the
assumption that younger players do not become stronger at the same rate
across generations.


No such assumption is required. I don't know how you came up with
this.

Insulting: Kbachler is not a professor as I have been. He simply
claims I'm incorrect without any evidence. Furthermore, I believe simply
logic of my post would lead any reasonable person to believe that I am
correct on that but this was also explained in detail to kbachler in the
prior the thread. Obviously, if younger players are unusually strong in
this generation middle-aged players ratings would go down.


Oldtimer Several posters proposed that in the current era younger
players have become much stronger and so they are driving the older
players ratings down and so the methodology that the ratings committee has
been using is unsound for the current era.


Not at all. The point is simply that it's been shown on several
occasions and anecdotally that younger players tend to have more volatile
ratings, and that they also tend to improve more rapidly. Therefore a
group with more younger players tends to suffer more deflation because
improvement in general causes deflation in the Elo system.

Insulting. "Not at all" kbachler ignores what the other players said.
He gives his argument without acknowledging what was previously said


Oldtimer said

"An alternative approach to determine rating inflation would be to
compare the abilities of older players with similar ratings to new
players. After the opening is played, you could look at the average number
of blunders made across a sample of games from different eras. A computer
could be used to identify blunders of greater than .50 centapawns, for
example. If the # of blunders is higher today than for similarly rated
players then you have a case for inflation. On another thread, this idea
was suggested and Kenneth Sloan claimed you would need millions of games
collected over a 30-year study to do an appropriate analysis. Obviously,
this is not true but you would in my opinion have to analyze at least
1,000 games. Of course, that's not that big a deal with today's computers.
If this study was done, I believe it would be fairly obvious that over a
30-year period some rating inflation has occurred. I recall 2400 rated
Seirawan winning many major U.S. tournaments and I strongly suspect that a
young Seirawan would do the same today but of course then his rating would
quickly rise to over 2700. Of course, I admit I could be wrong about that.
I recall that Jim Tarjan's USCF rating got over 2700 but he did that by
rarely losing a game. He was quite dominant in California for a time. When
Dzhindi won the U.S. championship in 1983 he was not even 2500."


Kbachler Basically, your last paragraph is nonsense.
Insulting

Oltimer Evidence to support the strength of younger players is a
dramatic increase GM titles, IM titles, FIDE and USCF in ratings. There
are only really only two possibilities to account for such a dramatic
increase. 1. Improved strength of young players 2. Rating inflation
kbachler said quoted my next sentence which was oldtimer If hypothesis 1
is correct than the USCF demographic studies are unsound.
Hypothesis 2 is neither correct nor possible.

There is no need to address the rest of this post. It's completely
off-point.

Depending on what your first hypothesis means - since it is stated
poorly - it is either irrelevant or proves our point.
Now he's saying the last part of the post is too bad to address the
last part of the post. After claiming that

oldtimer wrote:
I would say most of the recent posts suggest a confusion about how to
define rating inflation.


kbachler wrote: Generally, that seems to be true primarily of your
posts.

Insulting

Oldtimer said In particular, the USCF argues you cannot look at
averages ratings across time and claim that inflation has occurred.


kbachler This isn't just a USCF argument. People knowledgeable about
ratings already know this as a given

Insulting kbachler is implying that because I stated the USCF
argument that I somehow don't understand that it's an argument that other
people use. He cites no evidence to support that claim.

Kbachler goes on to say "Inflation and deflation have causes. The
immediate cause of deflation is that a player's (or more) performance
improves. The immediate cause of inflation is that a player's performance
degrades.

If you are going to argue that a pool is inflated or deflated, you
should be able to point to a significant segment of the pool and describe
where the above is happening."

insulting. I was talking about young players improving. That is, I
did exactly what he accuses me of not doing.


Kbachler makes the same point again. "USCF doesn't define inflation
or deflation differently from other knowledgeable entities. Lay people
often get confused and mis-define it."

Insulting. Kbahcler knows I'm a retired college professor with
expertise in statistics and psychometrics. I'm not a lay person.

Oldtimer said "but that doesn't mean the entire pool is deflated.


kbachler When over half of the pool are youth, with volatile ratings
and a lot of improvement it does. Every time a youth improves and deflates
an adult (and other youth for that matter) they have a ripple effect to
the next person. To argue otherwise with these demographics is absurd on
its face.

Insulting: He's calls my argument absurd. He says nothing about
whether young players are deflated or inflated and even acknowledged that.
Since they are also the largest part of the pool according to kibachler it
stands to reason that the entire pool's standing on inflation/deflation
depends largely on the younger pool and not the older pool.

Oldtimer The problem is your methodology doesn't address the
question.


It does. The problem is that you still don't understand it.
What makes you think the only portion of the pool examined is adults
in a specific age? That's just the portion simplest to explain, and you
aren't grasping that yet. Their were several adjustments made to the
rating system to correct for ratings integrity. The one considered most
important, because we were losing adult members, along with the simplest
one, is the one mentioned.
Insulting First he again claims that a college professor cannot
understand his thinking. He add what makes you think the only portion of
the pool examined is adults in a specific age?

What made me think that was is that he previously agreed with me on
that point. In fact, it's the only time he agreed with me.

oldtimer wrote:
In contrast, USCF argues for deflation in the past 10 years. As far
as I can tell this is the result of looking at middle-aged male players
whose ratings are generally stable. This is the methodology they have used
in several studies in recent years. The theory is that if these players
ratings go down we have deflation and if they go up we have inflation.


kbachler Fairly on track.


Oldtimer: The maximum liklihood analyses suggest that inflation has
occurred.


kbachler Actually, it suggests that deflation occurred at most parts
of the system. insulting. I gave two references to this and both support
argue for inflation (Chessmetrics.com) is the primary reference. If
kbachler has another reference he doesn't give it.


oldtimer wrote:
The Glicko system gives extra points to players who rise rapidly. The
purpose is to address concerns about deflation. That is, it quickly puts
the stronger player at a better rating but if you have a whole group of
stronger players entering the system you get a great influx of rating
points that might overall lead to inflation.


kbachler There was no influx of stronger players. I don't know where
you keep getting that. The influx is at the low end of the system.

Moreover, the USCF system now has similar protocols and it is
currently reinflating the system.

Insulting. He basically states that there is no stronger players
without any evidence. Note I have already stated by evidence that there is
(e.g., a ten-fold increase in GM norms in the past 10 years). He then
claims the USCF has similar protocols as if I don't acknowledge that when
I explicitly am referring to the Glicko system.

Oldtimer: Of course, other posters should realize that the
unhearalded kbachler is criticizing me (an academic with many statistical
publications) for having a lack of statistical knowledge.


kbachler Right now I don't care about your credentials, I care that
you still aren't grasping simple concepts. I suggest that until you do,
you might be better off not leveling criticism.

Insulting

Oldtimer Inflation occurs because more points enter the system.


Insulting kbachler This single statement alone demonstrates you
haven't a clue.

.. At the VERY low end of the system I believe it was an issue,

First he takes the statement out of context in that I also refer to
other conditions that are required for inflation. He later acknowledges
that my point has some validity after accussing me of being clueless,
which demonstrates that he is liar. He goes on to accusse me of several
things that I didn't say and therefore are also lies.

Kbachler Simply ADDING players does not per se cause inflation.



Insulting. Kbachler- May I suggest at this point that you quit the
online dialogue, because your lack of fundamentals is a waste of time for
you and others. Write to the rating committee, and ask for an explanation
of their analysis. REad Elo's book. Get some fundamentals. In a prior post
he acknowledges that he knows I have read Elo's book.


George. In terms of your computer idea that's very similar to what I
initially proposed and Tanafl proposed in another thread. I believe your
basically correct. However, opening knowledge changes so what you'd have
to exclude that part in some what. Perhaps you could have a standard set
of middle and endgame positions that people could play against (A human
test suite). Kenneth Sloan claimed it was a bad idea because you would
have to the study constantly on all games at ICC for 30 years. I don't
understand his logic on that one. Another way to do it is to simply have
the computer analyze games from players within rating categories doing a
blunder check in the middle and endgame using fixed depth searches to
control for processing speed and using only existing tablebases at the
beginning of your five year period and perhaps an algorithm to extend the
depth in endgame positions where greater depth may be needed--Chess
Assistant 9 has one. The analysis should find approximately the same
percentage of errors across ratings categories in the end and middlegame
as it did five years before if no inflation/deflation has occurred. Of
course, one can go back in time and analyze old games from the past and do
the same analysis to determine if inflation has occurred in the past. P.S.
There are all these coaches out their for kids. Maybe they need some
coaches for old people like me too.


kbachler The idea is fundamentally unsound. It assumes that rating X
today (say 1600) has the same knowledge as rating X in a prior era. That
assumption is false.

Ratings compare performance, not knowledge. Someone who maintains
exactly the same knowledge would decrease in strength relative to the pool
over time as general chess knowledge increases. I don't believe the pool
has ever been corrected for this "implicit inflation".

Insulting The idea is fundamentally unsound. Also, I acknowledged his
point in what I said and he's lying again by claiming that I didn't.
However, opening knowledge changes so what you'd have to exclude that part


oldtimer wrote:
Of course, it's true that opening knowledge improves over time but as
i said earlier you simply don't use moves that are book (at that time) for
your analysis. Has endgame play improved over the past 30 years for the
average player? In my opinion it's more likely it's gotten worse. Have
their been advances in the middlegame. A few, but not as many as a younger
player might think. For example, would a Capablanca get blown out
knowledge of the middlegame. Probably, the reverse is more likely. In
fact, some have suggested that is how Fischer was able to defeat the
Russian jugernaut. That is, he studied non-Russian greats while the
Russians tended to only study Russian greats. Of course, if you discovered
that was a problem there are ways around it. What probably hasn't changed
since the dawn of chess is the error rates. Actually, several senior
citizen centers have asked me to teach chess but I declined as I seldom
teach and at the moment I feel very rusty but I believe there is a market
for senior chess, especially with all the books out about how chess might
help people with alzheimer's and other forms of cognitive decline. If you
see it that way, you might have some success. Personally, I believe the
reason the ratings committee does not participate in this discussion is
because they now realize they've been doing it wrong in so far as using
demographic analyses to determine whether there has been rating
inflation/deflation. Hopefully, they will have the courage to change
things. I did read Tannafl's idea on the other thread which Kenneth Sloan
pooh poohed but his arguments were unsound. He's an expert an computer
science and as far as I can tell has not published anything on methodology
or advanced statistics. Glickman and I of course have and Glickman has
admitted in his recent interview in chesslife that he doesn't know whether
there is inflation or deflation occurring and he unsure if his system is
working. he basically responded to administrator's who felt they were
losing middle-age membership because their ratings were going down against
strong young players. The new system gives bonus to people whose ratings
are improving quickly. This seems like the best system but I mentioned
earlier how this could have some negative side effects. Speed of
processing definitely slows with age in everyone and cognitive decline
begins as early as ago 40 for people who do not keep themselves in shape
or are stressed. This would definitely have a negative effect on one's
rating with faster time controls. Another interesting study would be to
look at if the middle-aged groups decline was related to the percentage of
tournaments that were moving to faster time controls. One has the data to
look at it and if one knows something about sampling theory you could do a
quick and dirty study looking at only about 300 people at two points in
time.


k-bachler- I really suggest that you would be better off in a private
discussion with the rating committee, rather than spreading incorrect
information and untested theories to the average player. Besides, if you
are so certain that you are correct, then posts here are meaningless
anyway. A direct discussion with the rating committee would be much more
useful for you. If you are correct, it would be much more likely to effect
change. And if you are wrong, it would be much less likely to confuse the
average player.
_________________
Kevin Bachler

Insulting
Oldtimer, no one is afraid of your questions. Just annoyed. Please
stop posting misinformation.
_________________
Kevin Bachler

Repetious, Insulting

Kibachler You don't seem to understand what inflation is. Please go
back to the fundamental post earlier in this thread that explained
inflation and deflation.

Inflation presupposes that players are rated higher than their
performance indicates. The only way that increased knowledge creeps into
the situation is via increased performance. If the performance is
increased, then there isn't inflation.

Knowledge creep in the pool is not inflation.

Repetitious, Threatening insulting and a lie


What absurd statements? And what have I said that would concern the
moderators?

BTW, I already put in this thread why the computer approach doesn't
work, and the answer is fairly obvious.

Again, you would be better off if you learned the fundamentals and
had a direct discussion with the ratings committee.
_________________
Kevin Bachler

Reptitious, insulting

On Mar 26, 12:14 pm, "Todd" wrote:
Kevin Bachler

Reptitious, insulting

That is certainly the Kevin Bachler we knew and loved here for so many
years.