Characterizing Rating Levels

Sort:
exigentsky
Gonnosuke wrote:

I think that most players below expert level are primarily tacticians.  For the most part, their rating is a reflection of their tactical skill and their board vision.  Most of their victories are the result of executing winning tactics (or plundering blunders) and they often have very little positional or endgame knowledge for the simple fact that they've never needed that knowledge to be successful! 

As ih8sens alluded to, tactics will only get you so far.   Once these players start to face off against players who know more tricks than they do, they either get serious and dedicate themselves to further chess study or they get discouraged and figure they've reached their peak.


If you're referring to OTB ratings, this is a huge underestimation of players below Expert level. Class A players are typically competent positional and tactical players with a high degree of chess understanding.`Your generalization better fits Class C and below. Although, even Class C players understand the rudiments of positional play and don't just rely on tactics.

JohnClayborn

Okay, just my $.02 here...

First, I would have to agree about the chessware...it sucks. I can play OTB and online somewhere between 1300-1400, but if I play a 1200 Chessmater 10th edition player, I get beat down like a bad habit. Crazy.

But, getting back to the real issue at hand, and tossing in my own personal experience I would have to say that a lot of the below 1200 rated players have a psychological disadvantage. I have a pretty decent understanding of endgame theory. And I can do well in the middle game. I have limited understanding of openings at all and often times Im just tossing my pieces out there hoping I dont screw up too badly. when I first started playing I would fight to the last man and hope for a stalemate when I was obviously beaten. Now I just resign. I can see forks, and sometimes traps. I think 4-5 moves ahead (although it doesnt always work out the same way OTB as it does in my head). One of the biggest things that I notice when playing lower rated players is that they not only dont see the forks coming, but they are afraid to loose pieces. Chess isnt about not loosing pieces. It's about trading pieces for position, or trading pieces for better pieces, etc. There have been many a game where, if my opponent had just called my bluff or moved his power pieces down to threaten me back or just took my queen and lost thiers instead of running away the game would have turned out far differently.

Another thing that is slightly difficult to quantify and measure, but should be tried, is that each person thinks differently. We have different analytical processes, we have different preceptions and different preconcieved ideas. Some of them cultural, others not. As a result, we will also be more inclined to develop our "area of expertise" that suits us based on a combination of all of those factors. An example, I know that I suck at attacking, so I prefer to black black. I play defensively fairly well. I play offensively rather poorly. I can see the attacks coming from my opponent and stop them a lot easier than I can plan my attacks out on the board. I win by allowing my opponents to make enough mistakes that I can exploit to beat them. With other players the opposite is true, they will just attack and attack and attack and keep the opponent on the ropes and psych them out. The psychological aspect of the game, in my opinion, has at least as much to do with rating as understanding the rules and principles and techniques.

Im also very interested in this and I'm willing to help you undertake this project in any way that I can.

Zenchess

If I'm being serious, of course you can't classify players by rating.  A rating is a single number that cannot possibly quantify the complex amount of information that is necessary to convey what you understand and what you don't. 

And how well you'll do in competition is also far more complex.  Even people with a high degree of understanding can fail miserably in competetion.

queenofdeath

think the concept is interesting...

  you stop at 1500 wonder why? anyways...

  I Think one aspect to NOT over look is how does this person get to be called

1700 for example... I know some well balanced 1700 players that if given a certain opening or end game appears can play for a win vrs 2000 ranked players.

  Just my view but THE YEARS you have put into chess counts... I mean years

in regards to the amount of games, books youve read, games studied, end game

practice... so on and so fourth...

  1600 LEVEL players to me know most openings but still get caught by the better players... they learn to transition from mid game to end game... at least the concept... they see most pins, and can see many moves ahead

 1700 level is good at openings,  can shape mid games, end game is advaced

  1800 is expert at opening, mid game is hard to beat, solid end game...

  1900 - 2000 knows vast number of games, can see the 20 moves ahead less

or more... has a second to none end game but can't put it all together verse

the CHESS GODS...

 2100 -- openings depict win or loose with any player higher than 2000...

they hardly loose with perfected openings which are a few for white few for

black... mid game is solid... needs to  use end game to shape exchanges

to win a higher percent of games... the creative reasoning so to speak about

how to get through to a win from mid game stance is the roadblock... the reason why they are 2100

2200 expert in 20 openings for white or black, mid game transition almost flawless... end game is perfect... few beat them, small mistake game over...

2300 and above understand I would say 200 chess openings...

can remake 500 games from memory... end game is flawless and may write

articles about games and ways to improve games to win that were played

by the worlds best... the mid game is one of art and due to playing 100,000

or more games... they have super hi I.Q ( ON TESTS ) get OVER it as well,

i.q counts in this example...  they recall a hi percent of chess books they read and by this allone can beat 2200 and below just by opening therum and carrying a solid mid game transition...

  I also feel the immagination of a player with such a recall is second to none...

for example if you have no stress cause you know what will occur.. then you are left with a creation based focus.. I feel this is the reason why certain players

are the 1 out of 500,000 players that have this high a rating day in and day out... against the worlds best...  and last but not LEAST... THEY love the game..

they play the game, and they put the time in...

2500 and above... NO COMMENT... can't imagine what creates someone at this level... I am sure chess is their life at this point... small understandings and discoveryies are why they breathe I supose... I think they know all in the opening... ( even to the point of creation of somethinig new LINE WISE )

I think the mid game is trivial to them... they see all...

and the end game is so keen, they shape moves 1-20 around it... minus some

giant threat that stalls the intention of mid-game-end-game transition...

   As well, I think they are GIFTED IN MORE THAN... 1 area of super mind power... LIKE the photographic recall... ok, given, HI i.q.. given...

( I FEEL THEY HAVE ONE OR TWO MORE GIFTS AS WELL.... )

   May be genetic rapid eye movements... may be ADD, OR SUPER ADD..

may be an over logical mind that can't stop tearing a puzzle apart till it's solved.. ( the INSANITY FACTOR ) that made the worlds best artists, musicians,

inventors and any other art form... I feel these same chess players would have

been great at something else as well... even worlds best and it's in this 10 super

attributes that they are who they are... I think other players work themselves

to death but NEVER CRACK... 2100...  I THINK we understand it when someone cant bench 600 pounds...but we almost dont fully give when thinking of mind power... but it's true some will never crack 2100... just dont  have the mental

juice to pull it off... anyways.. this is my thoughts... could be out of my mind

but at least I am havin fun.. enjoy everyone...

Cobalt_T

very interesting idea but hard to work out I think

buttonc

It's a brilliant idea and could lead to a general improvement in the level of play by members of the site.  Just the publishing of the list is a good thing in itself.  A player can look up "hanging of pieces" for example if he/she does not know what that means, and can then become aware of an inherent fault in their play.  Grouping the list by rating makes it more personal and more readable. 

artfizz

There are several fascinating threads to this discussion running in parallel - all to do with getting the bare rating number to make more sense.

One theme (courtesy of hicetnunc) is Rating Class Labels or ranks. The United States Chess Federation [USCF]  uses these:

National Class A

USCF (1800-1999)

Top Amateur Class

National Class B

USCF (1600-1799)

Strong Tournament Player

National Class C

USCF (1400-1599)

Average Tournament Player

National Class D

USCF (1200-1399)

Strong Social Player

National Class E

USCF (1000-1199)

Social Player

It shouldn't be too hard to come up with a more relevant set of ranks for informal use within chess.com - and to extend the range above 2000 (which, in the real chess world, is covered by International or Professional Titles: GM, IM, FM, SM, NM and Expert/CM).

artfizz

Another strand to the discussion is that the rating is an average over the different aspects of the game (ozzie_c_cobblepot, gonnosuke). Analogous to a school report card, the interesting part is in the details: e.g.

SUBJECT

SCORE

History

65%

Science

45%

English

75%

Math(s)

60%

Art

80%

Technology

30%

 

 

Final Mark

59%

If CHESS is the whole curriculum, what are the individual fields or SKILL AREAs? The PHASES of the game: Opening, Middle game, End-game. Key concepts: Development? Co-ordination? These start to get ill-defined! What is this list of HIGH-LEVEL skills?

artfizz

Another aspect that has been mentioned (chawil, costelus) is that rating is the average against all players. We don't consistently beat players rated 100 below us - or lose to players rated 100 above us. In fact, the reality may be more like this:

OPPONENT

His/her rating

Outcome

Fred

1300

He always beats me

Sue

1400

She always loses to me

Skip

1500

I beat him 50% of the time

Joe

1600

I beat him 50% of the time

What’s my rating?

As zenchess and others have pointed out, different situations give rise to different levels of performance: e.g. OTB, CC, competition (tournament), ...

artfizz

The main thrust of the discussion is the checklist of specific skills that can can be separately practiced and 'mastered'.

There are many glossaries on the web that define the jargon of chess (e.g.  http://en.wikipedia.org/wiki/Chess_terminology). These cover terms like: pin, passed pawn, sacrifice, connected pawns, forks, discovered attack. There is clear value in understanding the terminology. I would characterise this as AWARENESS.

The checklist of skills is not the same as this. It is one thing to understand what forks are. It is quite another to be able to anticipate them and employ them. One assessment approach that is commonly adopted is this:

SKILL or Competency

Aware

Competent

Expert

Forks

 

 

 

Fianchetto

 

 

 

En passant

 

 

 

Smother mate

 

 

 

What is this list of techniques/competencies?

The individual techniques could be grouped as suitable for: novices, better players, strong players, etc. - although this will tend to be subjective. We then arrive at a table like this:

 

SKILL or Competency

Aware

Competent

Expert

Novice < 1200

defending pieces

 

 

 

 

avoids weak openings (e.g. a3)

 

 

 

Apprentice < 1400

castling

 

 

 

 

forks

 

 

 

 

en passant

 

 

 

Competent < 1600

Openings (e.g. Sicilian)

 

 

 

 

 

 

 

 

Developed < 1800

 

 

 

 

Does this ring any bells with industry workplace assessment?

JohnClayborn
artfizz wrote:

Chess Analysis Pro 7 (http://www.chess-analysis.com/chess-software/chess-analysis.html) looks interesting. They have identified a list of factors e.g. (Played Too Fast, Fell To Counter, Missed Opportunity, Poor Central Control, Poor Development) which acts as a checklist. You have to analyse your own games manually, ticking the boxes in their checklist. After a while, a pattern should begin to emerge. As far as I can make out - from this tutorial (http://www.youtube.com/watch?v=KEY_lY8YTf8), you have grasp what these factors mean - and recognise when they have occurred - there is no automatic analysis. It sounds like the type of approach that should increase chess awareness.


 Hmmm...that site seems like it has a good idea in principle, however I have to question the logic of having people analyze thier own games...theyre going to be limited by thier own knowledge. They're going to be missing things that they dont know that they dont know. For instance, I was analyzing a game of mine and a 1800+ rated player was looking over the moves and I made, what I thought, was a perfectly legitimate move and he said "oooo...you shouldnt have done that..." He obviously saw something that I missed. Since we are all a giant community that laregly gets along, and we're all (for the most part) interested in progressing the game and the knowledge and love of it, I definitely think that higher rated players should rate games of someone at least 1 level beneath them whenever possible. I think you'll get a more accurate breakdown that way.

thegab03

Ratings for me is only in the head & only in the head so forget your head & you'll be free!

JohnClayborn
Zenchess wrote:

...A rating is a single number that cannot possibly quantify the complex amount of information that is necessary to convey what you understand and what you don't....


 While I do have to agree with Zenchess on this point here, I do however, still believe that a lot of work can be done in this area. I do not believe that it will be as cut and dry and black and white as was initially proposed, but I still think that it would be doable. I think that all of the dynamics of the game need to be taken into account:

Opening Moves
Middle Game
Endgame
Types of Checks
Types of Mates
Types of Forks
Pins
Traps

etc.

I have a feeling that the average and lower rating players will score highly in one of these main areas of classification, but not in others. I also think that the area in which people are going to score well will differ vastly from person to person (partly due to the differences in the way that people process information, partly due to thier past chess experiences and knowledge, and partly due to culture). Some people, like me, are going to SUCK at opening moves, be okay in the middle game, and have a solid grasp of endgame techniques. Im much better at defensive techniques than I am at offensive techniques. I understand that I dont understand positional play very well, but I dont know what it is that I need to learn. Then, I know other people, that are the exact opposite...they wallop me in the opening of the game because they understand openings and attacks, but Im able to recover lost ground and beat them in the end because they dont know how to hold their position and close the deal. And, as artifzz was pointing out from another article, sometimes it varies from person to person, largely based on which of these areas they understand and what they lack. Case in point, at my work there are roughly 20-30 players who play chess outside on our break patio every day (we live in the desert so it's tolerable all year round). Some of these people I can beat easily, others spank me like a small child. Others it goes either way. Some of the people that beat me lose to people that I can beat and vice versa. It seems to boil down to: how does your knowledge and stragety counter that of your opponents. Kind of like paper, rock, scissors, but with chess principles instead. It's definitely something that we want to keep in mind moving into this.

Im definitely serious about helping out. Im good with excel and I have quite a bit of psychology and teaching experience (I teach people for a living) so I might be able to help break it down from the psychological and learning process standpoints.

thegab03

JohnPaladin

Im definitely serious about helping out. Im good with excel and I have quite a bit of psychology and teaching experience (I teach people for a living) so I might be able to help break it down from the psychological and learning process standpoints.

I teach dogs for a living, it does not really help my chess but rather helps myself!

JohnClayborn
thegab03 wrote:

JohnPaladin

Im definitely serious about helping out. Im good with excel and I have quite a bit of psychology and teaching experience (I teach people for a living) so I might be able to help break it down from the psychological and learning process standpoints.

I teach dogs for a living, it does not really help my chess but rather helps myself!


 ....okay....I'm sure that the psychology behind why dogs learn is different than the psychology behind why humans learn. All Im trying to say is that I understand the learning process in humans from this psychological standpoint and that there is definitely a correlation between how much we have learned and our chess rating.

thegab03

I'm a confirmed ornitholigist aswell, I train parrots to play chess!

artfizz
thegab03 wrote:

I'm a confirmed ornitholigist aswell, I train parrots to play chess!


"I find that hard to believe."

Rael

           
            > SQUAAAK <
     ~  Polly wants to capture ~

YeOldeWildman

The thing that makes this sort of thing difficult (not to mention potentially misleading) is that ratings aren't a one-size-fits-all sort of thing. In a recent thread on the validity of ratings, I observed that people use correspondence chess in many different ways and this will have a considerable effect on their ratings. 

For example, there are people who only play a few games, spend an hour or more on at least some moves, use books and databases, and generally do everything they can to make the best move they can every move of every game. Other people play socially, chat a lot with chatty opponents and don't really take the chess part all that seriously.  Still other people play to improve their OTB play and don't use books or the analysis board. Yet other folks have more than 100 games going on at once, since this quantity-over-quality approach can be used to improve one's playing strength even if one isn't putting forth one's best effort in any particular game.  And so on... 

So when you look at someone's rating on a CC site like chess.com, you have to realize that the person's rating is valid given their chosen playing method -- but it may not reflect their level of chess knowledge very well at all.  I remember playing against a National Master (US) who had 125 or so games going at the time saying in a post he played here to keep himself tactically sharp and to plug holes in his opening repertoire.  I played two tournament games against him;  I earned a draw in one and he offered me a draw from a better position (it was a gift, I was in big trouble) in the second since (I assume) he needed half a point to lock 1st place in the tournament. 

In OTB play he's going to be rated 2200+ USCF and my most recent USCF rating is 1702.  I have no illusions how I'd fare in an OTB tournament game against him or which of us has the greater chess knowledge and the superior chess skill.  But me busting a hump on each move was (barely) good enough to hang on for two draws against him when he was maybe spending 1-2 minutes per move.  And I lost rating points since I was higher rated on chess.com at the time!  Smile

My conclusion #1:  Using OTB tournament ratings would be much a better idea than using CC ratings for the sort of study folks here are proposing.  In OTB, everyone has the same amout of time and the full attention (presumably) of one opponent so you'd be cutting out a lot of irrelevant factors.

However, even using OTB ratings may not produce the sort of clean results one would hope for.  That's because (as several people have observed), there are many ways (i.e., skill combinations) to get to any particular rating.  Ratings are a measure of how you generally stack up against the ratings pool you're swiming in;  they don't define you as a chess player (let alone as a human being -- though you wouldn't know it listening to some people...).

I had an acquaintance in junior high school who was always rated substantially higer than I was (maybe +200 points UCSF).  He never beat me in a game that mattered:  tournaments, inter-school team matches, etc. I always won.  He probably was a better player overall, and I think he went on to become a Senior Master (USCF) after taking a break to go to college.  But he couldn't beat me and it drove him crazy.  IMO, it was simply a matter of style.  He simply couldn't cope with my particular skill set.

My conclusion #2:  Whatever the ultimate chess "truth" about playing strength ultimately is, it can't help but being a lot more complicated than "Class C players know X, Y, and Z."  Jeremy Silman in The Amateur's Mind had a lot of interesting observations to make about how his students of various rating levels think about chess.  If someone is really serious about researching in this area, that would be a great place to start.  Good luck!

artfizz
YeOldeWildman wrote: ... My conclusion #2:  ... it can't help but being a lot more complicated than "Class C players know X, Y, and Z." ...

I agree. However, if you turn the whole thing inside out and view it as a syllabus: a schedule of what everyone at that level SHOULD know rather than what they DO know - it becomes potentially useful as a study framework.

There are several interesting chess syllabuses out there. e.g. The National Chess Syllabus™ featuring The Bandana Martial Art Exam System™ is a fun & funky way of learning how to play chesshttp://www.trafford.com/04-2350 (see page 2).

http://www.ipswichjuniorchess.btinternet.co.uk/syllabus.html

http://www.chesskids.com/syllabus04.shtml