Master versus class A player

[Alex Chambers]

What is the difference between a master chess player (2200) and a
class A player (1800) in terrms of strength. What does a master 'see"
during a game which a class A player. I don't understand the
difference.
they both seem strong.

[Andy Walker]

On 12/12/12 22:34, Alex Chambers wrote:
> What is the difference between a master chess player (2200) and a
> class A player (1800) in terrms of strength. What does a master 'see"
> during a game which a class A player.

The 2200 player is likely to have a wider and more secure
knowledge of openings, a much better understanding of positional
play and strategy generally, a quicker grasp of tactics, and a
much better endgame technique. No part of this is universal; but
as a four-part package, most of it must be present, else the 2200
player wouldn't typically beat the 1800 player so comprehensively.

It's less what the stronger player sees, more the speed
and efficiency with which it is seen. In other words, the stronger
player is much quicker to cotton on to the most important features
of a position, and to concentrate on things that matter.

> I don't understand the
> difference.
> they both seem strong.

If you are rated, say, 1400, then they both are strong;
they both have better technique than you, they see things more
quickly, their judgement is better. If you are rated 2600, they
are both weak, for the same reasons but swapped around. If you
are rated 2000, then one seems strong, and the other weak. The
2000 player is quite likely to be at least as good as the 2200
player in some aspects of the game, which is why he can expect
to draw reasonably often and even win a few; conversely, he is
likely to be no better than the 1800 player in some aspects,
which is why he sometimes draws or loses.

Somewhere around the 1400-1500 rating, the picture seems
to change. In other words, players below that sort of level do
simply miss things that the other player sees, and would continue
to miss them even with ample time to analyse or after a stronger
opponent has shown the refutation. But players above that level
know all the things they really need to know -- they understand
pins and forks, pawn structure, development, the centre, king
safety, active rooks, ..., they just don't apply that knowledge
as effectively as stronger opponents.

A large part of the effectiveness of analysis by strong
players is due to "chunking", in other words patterns made by
several pieces or several moves in combination. By seeing the
pattern, you don't have to see the separate parts, so you have
to do much less analysis to see the same outcome, or to know
better what that outcome means. There is a fair amount about
this in the psychology literature and on the web.

--
Andy Walker,
Nottingham.

[William Hyde]

I'm not a master but I jumped from B player to strong expert fairly
rapidly. Suddenly A players simply seemed weak - they did not plan
well, and they didn't understand endings. Just what an IM would say
about me - and he'd be right.

Of course they made all these errors when I was a B player, but then I
was terrible at tactics, had no idea what a plan was, and rarely got a
chance to show my endgame abilities.

The one area I improved in the most was probably planning and
positional play. I never had a real advantage over 1900 players in
tactics - they're about as good as I am. But when your position is
better a favourable tactical shot is more likely.

So I don't think there is a qualitative difference between A players
and weak masters (IMs are a different story). The master plays better
in some or all aspects of the game but not everyone is a master in the
same way. One might be a superb tactician but positionally rather
weak, while the other is an adequate tactician but excellent
positionally.

Someone once told me that tactically I was about 1900, positionally
about 2100, and 2200 in the endgame. I might quarrel with the
numbers, but the general idea is reasonable.

William Hyde

[Martin Brown]

On 14/12/2012 01:32, Andy Walker wrote:
> On 12/12/12 22:34, Alex Chambers wrote:
>> What is the difference between a master chess player (2200) and a
>> class A player (1800) in terrms of strength. What does a master 'see"
>> during a game which a class A player.
>
> The 2200 player is likely to have a wider and more secure
> knowledge of openings, a much better understanding of positional
> play and strategy generally, a quicker grasp of tactics, and a
> much better endgame technique. No part of this is universal; but
> as a four-part package, most of it must be present, else the 2200
> player wouldn't typically beat the 1800 player so comprehensively.

A 400 ELO point difference amounts to the stronger player winning about
3 points out of 4 games and is about the right ELO difference to set an
engine strength above your own ability to avoid being beaten too
comprehensively and always learn something from it. Analysis should be
done at full strength with training mode on.

A computer playing at nominal ELO is different to a human in some subtle
and some not so subtle ways. Certain engines get the right statistics on
average by playing mostly GM level but with the odd crass blunder that
no normal player would ever make.

>
> It's less what the stronger player sees, more the speed
> and efficiency with which it is seen. In other words, the stronger
> player is much quicker to cotton on to the most important features
> of a position, and to concentrate on things that matter.

The more advanced player tends to know more types of set piece attack
and defensive tricks to allow planning deeper combinations. Notably
setting up positions where the weaker player will fall into traps.

>
>> I don't understand the
>> difference.
>> they both seem strong.
>
> If you are rated, say, 1400, then they both are strong;
> they both have better technique than you, they see things more
> quickly, their judgement is better. If you are rated 2600, they
> are both weak, for the same reasons but swapped around. If you
> are rated 2000, then one seems strong, and the other weak. The
> 2000 player is quite likely to be at least as good as the 2200
> player in some aspects of the game, which is why he can expect
> to draw reasonably often and even win a few; conversely, he is
> likely to be no better than the 1800 player in some aspects,
> which is why he sometimes draws or loses.

You can also be good at getting draws against (much) stronger players -
at one time I specialised in this in school matches.

>
> Somewhere around the 1400-1500 rating, the picture seems
> to change. In other words, players below that sort of level do
> simply miss things that the other player sees, and would continue
> to miss them even with ample time to analyse or after a stronger
> opponent has shown the refutation. But players above that level
> know all the things they really need to know -- they understand
> pins and forks, pawn structure, development, the centre, king
> safety, active rooks, ..., they just don't apply that knowledge
> as effectively as stronger opponents.

I think the point of changeover is a bit higher than that, but we won't
split hairs. The argument is sound and beyond a certain point the
important thing is having an intuition about what will work and looking
mostly at the right moves to either find a line or quickly refute ones
that do not work.

>
> A large part of the effectiveness of analysis by strong
> players is due to "chunking", in other words patterns made by
> several pieces or several moves in combination. By seeing the
> pattern, you don't have to see the separate parts, so you have
> to do much less analysis to see the same outcome, or to know
> better what that outcome means. There is a fair amount about
> this in the psychology literature and on the web.

I wonder if this is why chess appeals so much to natural pattern
matchers like mathematicians. TASC chess trainer is excellent for school
level chess even it is a bit long in the tooth now.

Regards,
Martin Brown