|
||||||||||||||||||||||||||||||||||
|
The Black Hole in Chess Ratings – Part III By Russ Mollot (founder, Chess Express Ratings) In case you missed them: Part I and Part II As
mentioned in earlier articles, children in a scholastic chess program are
typically assigned initial "club ratings" that are exceedingly low. They
then begin to learn the game at a very rapid pace, gaining knowledge,
experience, and skill. The participants generally are playing exclusively
amongst themselves, which results in some students’ ratings rising, some
falling, but the overall average rating within the group remains the same --
and way too low. These low scholastic ratings are slowly bringing about a
massive deflationary trend in chess ratings in general, as more and more of
the young chess scholars make their debuts in adult tournaments. What is needed are modifications to scholastic ratings which recognize both the young players’ initial lack of skill, and their ultra-rapid acquisition of the same. This objective has accomplished, with some elegance I might add, by Chess Express Ratings (CXR), and has been part of our rating process for approximately two years. The CXR system recognizes that our young chess scholars are learning and improving much faster than their ratings would ordinarily reflect (as demonstrated in Parts I & II of this series.) Consequently, some sort of extraordinary means was needed to upwardly adjust their ratings. Such adjustments needed to be fair, sensible, and gradual. Our original specification (called “Project Orange”) envisioned five components:
Note that at the present time, Standardized Aptitude Tests for chess have not become widespread -- in fact, this idea is still considered heretical by the "mainstream". Chess Express Ratings, Inc. will continue to be a proponent of standardized testing for chess scholars (just as such testing is standard in virtually every other academic field). Until such tests have been widely adopted, we are omitting the Learning Points component. Recommended Initial Ratings
We have, unfortunately, encountered heavy resistance to our attempts for standardization of initial ratings. Although our customers understand our reasoning, ultra-low ratings are already so widespread that CXR’s recommended initial ratings just seem too high. The urge to conform is strong. We do not seek to override the judgment of our affiliates, so we have had to adapt to the environment now prevalent. Practice Points Every scholastic player rated below 1000 is awarded two extra rating points after each of the first 100 games he or she plays – win, lose, or draw. This is in recognition of the fact that playing a complete game provides beginning students with much-needed practice, reinforcing their understanding of the basic concepts. A student who remains in the program and plays at least 100 practice games will have thus gradually gained 200 extra rating points, reflecting his or her significant over-the-board practice. Achievement Points (a.k.a. "Victory Points") Every scholastic player rated below 1000 is awarded three additional rating points for each of his or her first 100 wins. This recognizes that the student has attained enough skill to visualize and successfully execute a winning plan of action. Note that this includes games beyond the first 100 games played. The successful student can thus gain up to 300 Achievement Points, in addition to 200 Practice Points, for a total of 500 rating points. Safety Net CXR’s scholastic players benefit from a safety cushion preventing their ratings from dropping precipitously. Any student whose rating has fallen below 400 will never lose more than 10% of his or her rating points as the result of a single game. While not the same as a "rating floor", the Safety Net makes it very unlikely for a child's rating to fall below 100 or to remain there for very long. Examples Note: ratings and formulas here refer to those used in the CXR System.
The maximum impact of the program is upon lower-rated scholastic players (especially pre-teens), because players whose rating is 1000 or higher receive no extra points. However, such players still benefit indirectly, as many of their opponents will no longer be as badly under-rated as before. An added benefit of our special treatment of scholastic competition is that youngsters can see their progress in chess reflected in their ratings. A class of second graders starting with a 200 rating (knowing nothing about chess) can see their ratings climb as high as 700, on average, as they learn and play -- even if they compete only amongst themselves.
Our critics will accuse us of "tinkering" with rating tradition. We must plead guilty. Traditional rating methodologies were not designed to work with the chaos that is scholastic chess today, just as Sir Isaac Newton's laws of physics were unprepared to deal with Relativity and black holes. We believe our approach is warranted. It recognizes the realities of scholastic chess. It is certainly better than doing nothing, as the status quo would allow the massive rating distortions to continue and accelerate. |
The
Advertise to Single insert:
|
|||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||
