"Mike Murray" wrote in message
...
On Thu, 15 May 2008 08:38:15 -0400, "Chess One"
wrote:


On Sat, 10 May 2008 09:34:49 -0400, "Chess One"
wrote:

... I discovered that Ken Sloancould offer no explanation for the
complete
absense of quality control at USCF's rating department which
permitted
the situation, and secondly, that
those who proposed Tanner's award for a master title also couldn't
spot
thecheat.

Phil confuses quality control with fraud detection.

'Confuses?'

Doesn't lack of control data quality actually permit fraud? In this
instance, not only was there no quality control whatever, but an award was
made to a politico for what Mike Murray prefers to call 'fraud'.

For the first issue, clearly Tanner is responsible, but USCF's ratings
department is /as/ responsible, since after all, if you keep geting rated
by
chess officials, can you be doing anything wrong?

The folks who receive the rating reports, enter the data and publish
the ratings can hardly be expected to determine that some of the paid
USCF members in the crosstable don't really exist, or didn't really
play in the event.

They can't be expected? But I thought that was the entire point of official
ratings? Some expectation exists of 'officiating'.

And the spurious events didn't all come in a
batch -- they were nicely separated both in time and space.

But a visual scan of data of ratings would immediately make that clear. That
seemed to be a benefit of the XPR system I wrote about.

As I
remember, Tanner was strong enough that the fake events didn't stand
out as egregious anomalies. Some nice detective work by Sam Sloan.

Because the data is recorded as number. As a visual presentation in a
chart - it would be immediately apparent.

But if this is indeed USCF's system of /recording data/ where is any quality
control to occur?

Secondly, on awarding a master rating and rating floor, /still/ no-one
noticed or checked Tanner's record [if indeed, we are to believe that].

But the 'fraud', as Murray terms it, is the reponsibility of the 'awards'
system. Not only for Tanner, but for the gent writing to in the NY Times
about getting his own award, with no apparent invigilation whatever - no
questions, no data = master title and rating floor.

If this second instance happened as you describe, some guy just writes
in and gets a 2200 floor -- yeah, that's a big deal, much worse. Who
was the guy?

I forget his name. But it was in a NY Times Gambit blog. Ah - I researched
it, and here are 2 relevant paragraphs:

Chess history has been destroyed. The ratings for past individuals (prior to
1990) have been destroyed. Some say it was an accident. Others say it was
done to cover up false rating reports submitted by coaches to inflate their
chess ratings in order to gain respect and money by claiming superior chess
skills (e.g., I can coach you because I am a master).

Goichberg weakened the title of chess master by claiming that anyone with a
rating over 2200 (for only a few games) can call themselves a master for
life. The original life master title was for 300 games over master. Of
course, who is a life master has apparently been lost when other records
were destroyed. Yet, when I requested a rating floor on the basis of my life
master title it was given to me?? The USCF didn't ask me to send my
certificate or any documentation whatsoever.
- Posted by Todd MIller

http://gambit.blogs.nytimes.com/2008...#comment-13299

And there is the URL.



So there you go! Motive, Opportunity, and the Smoking Gun, plus Who Cares
behavior by USCF who do not apologise or even explain this state of affairs.

The paradoxical aspect of this issue is that I would not have looked further
that Sam Sloan's reporting of the cheat if USCF rating's boosters had not
[abstractly] rubbished other ratings services for their 'lack of QC'.

But at the time of Tanner's exposure, it was only Larry Parr and myself in
these newsgroups who restrained themselves from simply blaming Tanner -
since evidently USCF's system encouraged him passively in not vetting his
rating, and then actively by proposing an award. All this took place without
anyone noticing.

My first question to Ken Sloan on these issues was 'how many such examples
are there?' Which he blew off with his normal mono-syllabic dismissal.
Therefore, the unlooked for comment by Todd Miller is a strong indication
that Tanner's situation is not a singular one.

Journalistically, it seemed right to investigate other options for rated
play, both for the player in terms of speed, cost and detail of information,
and in terms of the rating agency, their own procedural activities. After my
interview with Russ Mollot of CXR I stopped short of recommending his
service, but certainly encouraged chess players, TDs and coaches to /look/
at it, then decide for themselves if it fit their own needs, budget, and if
in fact they would lose anything if they switched?

If indeed 'official' ratings have no officiating to them whatever - then I
propose this is another example of decayed USCF infrastructure - and
something of a challenge to their stewardship of American chess ratings. I
have even heard it argued by a delegate that USCF's rating service is
actually a 'brand'.

I think this is an unlikable idea, but by these accounts perhaps not an
untrue or atypical one. A brand now operated by a non-profit in competition
with for profits will be challenged on at least three criteria:

a) If the non-profit brand is at least 3x more than commercial services,
then in what way is this a service to chess players?

b) If management of the non-profit official brand has less quality controls,
especially of such non-Elo factors as awards and rating floors, then should
there be an independent assessment made of the type of service due paying
customers?

c) In terms of value for money, and also speed of service [the reader will
note that official correspondence ratings were 'delayed' at least 6 months
last year] should an independent ombudsman make a determination, not only on
what has gone before but on what suits the future? [which was the
recommendation of the correspondence master]

The minimum change necessary, it seems to me, is to /independently/
investigate these issues, and the system on which they are based, which is
now so far from Elo's original idea on inter-determining ratings by play
that it as to corrupt his very idea by 'awards and floors' all conducted in
the dark of USCF's recesses, and where we are asked to believe no one looks
even at master level ratings to determine if they contain the least
veracity.

Phil Innes