Characterizing Rating Levels

Yeah, good post, thanks for the insight.

This is a good thread.  Is there a way to categorize Tactics Trainer problems
into types, not just by rating but by similarities?

It doesn't appear so. The problems are tagged using a 7-digit numerical ID. When viewing the list of problems, you can sort the list by:

• Rating (lowest)
• # Attempts
• Rating (highest)
• Avg. Time (shortest)
• Avg. Time (longest)
• Pass% (highest)
• Pass% (lowest)

Even if the tags had some significance (e.g. 00001xx -> en passant), you cannot sort the problems by tag.

I think it better, to go by what a game is rated. This thread shows what people's perspective is on ratings, so here's mine

900-1100- You hang a piece twice; you miss mate twice; Your strongest tactics are obvious to the next group.

1200-1300- You hang a piece once in the game; you miss mate once; you only know tactics, no strategy. Your tactics are of a 1-3 move, forced kind at best, or an error.

1400-1600- You rarely hang pieces to those in your rating range. The three areas of chess: opening, middle, and end game become apparent; You favor at least one of those areas without knowing.

1750-1850- You have become more diciplined; you are creating tension in the position; Habit has evinced intuition; Strategy comports itself.

1900-2050- You begin to think instead of see, and you are always asking yourself, which to trust; Positional understanding appears for the first time; You become more aware of how certain positional, or tactical ideas 'create' the opportunity for your opponent to err.

Since I've never gotten beyond that rating, I have to stop. It just wouldn't be genuine. This is how I view ratings based on what I thought of chess as I was given these various numbers at various chess sites.

The USCF is going to have to put some work into those title labels. 2nd category, 3rd category, etc. are not as catchy as the ones they used to have (post #50) - nor as those suggested in post #71.

@artfizz (Responding to post #50) The CM title, while in spirit closer to the Expert title, in strength is closer to the NM title. From what I know: USCF 2200+ is NM. FIDE 2200+ is CM. My perception is that it is more difficult to get 2200+ FIDE than 2200+ USCF, which would make the CM title more difficult than the NM. I suppose the problem is that "master" in FIDE indicates FM (FIDE 2300+), but "master" in USCF indicates NM (USCF 2200+).

"We use words like title, master, expert. We use these words as the backbone of a life spent defending something. You use them as a punchline. I have neither the time nor the inclination to explain myself to a man who rises and sleeps under the blanket of the very freedom that I provide, and then questions the manner in which I provide it. I would rather you just said thank you, and went on your way, Otherwise, I suggest you pick up a weapon, and stand a post. Either way, I don't give a damn what you think you are entitled to."

Another way of characterizing rating levels is to consider what one would be prepared to sacrifice in order to achieve it ...

I'd like to resurrect this old thread, because I've had a go at doing this comprehensively, and I'd like to know what others think. This is my effort, mainly based on observing players I know in OTB chess.

600 - More or less random moving of the pieces, by no means always legal chess moves.

700 - Mostly legal moves, but sometimes kings are left in check, castles through check, etc.

800 - Knows how to move all the pieces without making illegal moves, occasionally saves hanging pieces.

900 - Saves hanging pieces most of the time, can deliver simple one-move checkmates and captures hanging pieces.

1000 - Can do simple tactics, usually makes sensible enough opening moves for the first two or three moves.

1100 - Can sometimes do 2 or 3 move tactics, can mate with a K+Q or K+R, vague understanding of basic principles.

1200 - Knows some openings without understanding much about them, mostly gets tactics right, knows some positional rules.

1300 - Sometimes plays according to basic principles, but panics and stops doing so when put under slight pressure.

1400 - Typically plays according to basic principles without much understand how or what they are doing.

1500 - Typically plays according to basic principles with a certain understanding of how they apply to winning the game.

1600 - Understands basic principles well, but not dynamics or what features of a position matter more than others.

1700 - Understands basic principles well, and that some positional features matter more than others, but no sense of which.

1800 - Starting to gain some understanding of what features matter the most, but not how to play accordingly.

1900 - Starting to gain some vague understanding of what proper chess is.

2000 - Has a fair understand of how to play chess properly in theory, but gets it woefully wrong in practice.

2100 - Has a good understand of how to play chess properly in theory, but still gets it woefully wrong in practice. 

2200 - Understands in theory how to play proper chess, and occasionally gets in right over the board, but not often.

2300 - Potentially a strong player, has two or three major weaknesses keeping them from professional standard.

2400 - Potentially a strong player, has one major weakness keeping them from professional standard.

2500 - Strong player, plays chess properly, but without the genius of the top players.

2600 - Very strong player, plays proper chess with occasional sparks of brilliance.

2700 - Unbelievably strong player, plays proper chess with regular sparks of brilliance.

2800 - As far as is possible for a human, understands everything that is happening on a chess board.

2900 - Magnus. His brain must just be wired differently from the rest of us.

3000 - Computer.

Amazing that all the input, except for last , entry is over 5 years old but still interesting. There was a comment that fast chess with a low rating and slower chess a much higher rating smelled of engine use. This may be true but it could also point to the effects of the ageing process. I am 70 and useless at fast chess. Are there any other senile old gits out there who will back me up?

madhacker wrote:

I'd like to resurrect this old thread, because I've had a go at doing this comprehensively, and I'd like to know what others think. This is my effort, mainly based on observing players I know in OTB chess.

600 - More or less random moving of the pieces, by no means always legal chess moves.

700 - Mostly legal moves, but sometimes kings are left in check, castles through check, etc.

800 - Knows how to move all the pieces without making illegal moves, occasionally saves hanging pieces.

900 - Saves hanging pieces most of the time, can deliver simple one-move checkmates and captures hanging pieces.

1000 - Can do simple tactics, usually makes sensible enough opening moves for the first two or three moves.

1100 - Can sometimes do 2 or 3 move tactics, can mate with a K+Q or K+R, vague understanding of basic principles.

1200 - Knows some openings without understanding much about them, mostly gets tactics right, knows some positional rules.

1300 - Sometimes plays according to basic principles, but panics and stops doing so when put under slight pressure.

1400 - Typically plays according to basic principles without much understand how or what they are doing.

1500 - Typically plays according to basic principles with a certain understanding of how they apply to winning the game.

1600 - Understands basic principles well, but not dynamics or what features of a position matter more than others.

1700 - Understands basic principles well, and that some positional features matter more than others, but no sense of which.

1800 - Starting to gain some understanding of what features matter the most, but not how to play accordingly.

1900 - Starting to gain some vague understanding of what proper chess is.

2000 - Has a fair understand of how to play chess properly in theory, but gets it woefully wrong in practice.

2100 - Has a good understand of how to play chess properly in theory, but still gets it woefully wrong in practice. 

2200 - Understands in theory how to play proper chess, and occasionally gets in right over the board, but not often.

2300 - Potentially a strong player, has two or three major weaknesses keeping them from professional standard.

2400 - Potentially a strong player, has one major weakness keeping them from professional standard.

2500 - Strong player, plays chess properly, but without the genius of the top players.

2600 - Very strong player, plays proper chess with occasional sparks of brilliance.

2700 - Unbelievably strong player, plays proper chess with regular sparks of brilliance.

2800 - As far as is possible for a human, understands everything that is happening on a chess board.

2900 - Magnus. His brain must just be wired differently from the rest of us.

3000 - Computer.

3100 - bigbird419

How would somebody rated 400 play?

Not at all, really

bigbird's post was funny BTW, it deserves a 

If we want to be really mean and go below the cut-off of 600 that I used, then...

500 - Totally random moves, knights moving diagonally and the like. No concept of check or mate.

400 - Doesn't finish all their games, loses attention halfway though.

300 - Doesn't finish any of their games, loses attention halfway through.

200 - Doesn't know how to set the board up.

100 - Doesn't know what chess is.

0 - Doesn't exist.

There we go...

-100 - Never will exist.

-200 - Can't possibly exist in this universe.

-300 - Can't possibly exist in any universe.

However bad things are, they can always get worse! 

madhacker you're a legend!

Yeah that's an issue, but I thought the best approach was one based on observation of other people - not just how they play chess but how they talk about it. I don't personally know anybody above 2500 but there's still the internet as a resource.

I'm rated 1800-1900 in slow but 1200-1600 in bullet. I never use an engine. Some fast gamses I just can't move my mouse fast enough.

I have to disagree. Basically the slower the game the better I play, and when the time available drops below 5 minutes my results are hardly better than those of a beginner. My OTB strength just now is KNSB: 1935, FIDE: 1977, but 30 (gulp) years ago I had, for several years, a BCF rating equivalent to somewhere between FIDE 2100 and 2200 so although I am no chess genius I am no idiot at the game either (although I play my fair share of idiotic moves).

At the time of writing my 3-day rating on chess.com is meaningless as I have played too few games. However, my current rating at 10-minute chess on chess.com is 1830, but my bullet rating is stuck in the mid-900's. I don't cheat  at slower play by using a chess engine. The trouble is that even if I could think fast enough (I can't) I just can't move fast enough to get anywhere in bullet games, especially on the web where you cannot move andf bash the clock anywhere near as quickly as with a real board and set and a physical chess clock.