Thursday, September 28, 2006

Chess Positions to Analyze

Robin Grochowski writes:

In a recent OTB game, I (playing white) reached the following position:

I spent too much time debating whether to take black's bishop with my pawn (dxe5) or my castle (Rxe5). I took with my castle. I took with the castle (Rxe5) because I was concerned that in the upcoming endgame, I did not want my bishop blocked with my own pawns. I did not want a bad bishop later in an endgame.
Any ideas from anyone whether Rxe5 was the best move? Perhaps taking with the pawn would have been better? But if taking with the pawn is better, why is it better?


Southwest Chess Club said...


I like taking with the Rook on e5. Not sure if it's enough to win, but here's why I like Rxe5:

* you don't let Black exchange 1 or 2 pairs of Rooks down the d-file(easing his cramped position)
* you maintain the option for a possible d5 pawn break at some later point
* you maintain pressure against his funny pawn structure on f7/e6/f5.

You have a nice Bishop, which could be a factor later on.

Anonymous said...

Shredder Analysis:

Best Line (0.91): 1. dxe5 Rxd1 2. Rxd1 Rd8 3. Rxd8+ Kxd8 4. Kd3 Kd7 5. b3 c5 6. a4 Kc6 7. Bd2 Na6 8. f3 Nc7

Second Best Line (0.61): 1. Rxe5 Ne8 2. d5 cxd5 3. cxd5 Nc7 4. dxe6 fxe6 5. Rc1 Kb7 6. a4 Rgd8 7. b3 Kc6 8. Ree1 Rd5

All other possible moves for white, leaves white in a negative eval.

Ashish Vaja said...

This is a very interesting position, where you must both use some calculation and understanding to figure out what the strongest continuation for white is. I think that you must not be too quick/bias to judge this position. We are told do not put pawns on same color as bishop, keep rook active, use outposts, but we cannot blindly follow these GUIDELINES. You must go into this position objectively without any bias for one move based on your intuition. I know that you are told to use positional judgment, but this is by a 2400+ strength player who has trained his intuition to the point he can look at a position and moments later know all the features of the position. The average player cannot do this, and until he can, should objectively weight the positional features himself then draw his conclusion. Here is my thought process when I see this position.

It is clear that we are in an endgame and our thought process must go along the lines of planning and positional judgment rather than tactical/ deep calculation. This is important step in assessing a position; we must know whether deep calculation is important or planning/positional judgment.

The next step in this position would be positional assessment. The question what is going on in this position? There are many factors that can be considered when doing this, here are a few examples:
1) Material: is one side ahead in material? Is the material on the board the same?
2) King position: Is the king poorly placed for whatever phase of the game you are in? Meaning in middle game is the king weak? Or in endgame is the king far from the center?
3) Pawn Structure: Are there any passed pawns? Are there any potential passed pawns? Are there any weaknesses in pawn structure?
4) Weak squares: Are there any weak squares which our pieces can use as jumping off squares? Is there a complex of weak squares?
5) Control of center: self explanatory
6) Control of open lines: Does the bishop, rook, or queen have complete control over open lines?
7) Poor positioning of opponents pieces?
8) Poor coordination of pieces/lack of plan?
9) Space advantage?

(A note: This seems like a long process, but you can train yourself to do this very quickly. All strong players can just look at a position and within a minute familiarize themselves with all these factors and begin the next phase.)

So when I see this position, here are the positional factors that I believe are of importance:

1) Material: There is clearly a huge difference in material we have a bishop vs. knight situation if the rooks are traded. In this position it is clear the bishop is the much stronger piece. The knight is poorly placed and seems to have trouble finding an active square. If the knight can find an active role in this position it will make whites job much harder. White should try to take away all advanced outposts for the knight; black must try to get his knight out of its passive position.
2) King position: both kings are equally close to the center; do not need to take this into consideration.
3) Pawn structure: There are no passed pawns or potential passed pawns, but black has weak pawn on h7 on the kingside, and the doubled pawns are no longer mobile so they are also weak.
4) White does not have any weak squares, but black on the other hand does. It is important to know in these positions the value of the d6 square for both sides. Also the dark squares on the kingside are also easily accessible to the white bishop.
5) Control of the center: does not really apply in a practical sense.
6) Open lines: the d-file is open it should be noted, but no one can really claim it for themselves without trading rooks, when no one has it. Black has the g-file, but I do not see a practical use for it. Ideas of f4 can be met with g4 when black has only created a new weak pawn on f4. If black tries to play h5 first then white must have options like Rd4 to stop black’s activity. The bishop is clearly more powerful than the knight since it controls diagonals everywhere it goes.
7) Blacks knight is poorly placed, I am not only considering the knight itself, but the pawns in relation to the knight, and where it may go in the future. In this position this is a crucial point to consider, as if the knight has no place to go, white will be up a piece.
8) Both sides have a plan: white must try to imprison the black knight, gain space, take advantage of the weaknesses on the kingside, and squeeze. Black must try to get the knight into the game and create counter play.
9) White has a space advantage.

Ok so now we have assessed the position. Next step is plans for both sides, based on our assessment of the position:

White: Keep the black knight hemmed in, prevent attempts of counter play for black, gain more space, take advantage of the weakness on the kingside, and find a role for the white king.

Black: Get knight into the game, create counter play somewhere, if possible transfer king to the kingside to defend the weaknesses, but this seems too time consuming.

Next what plans/candidate moves do we have? Also ask yourself which move fits in with the positional assessment we have made.

A) 1. Rxe5: This move on the surface seems like the “natural move” for many people, but it does not take into account blacks plans, and white’s advantages. We should ask ourselves what is black’s main objective right now? Get the knight into the game! Can black do this now? Yes, 1…Ne8 is a very strong move, now the knight can use his f5 pawn, and jump into e4. White cannot kick the knight from e4 with bishop because it’s on a light square. If white uses his f-pawn then we have to worry about the g-pawn. White now has two options let the knight come to e4 or take immediate action neither is appealing to me, I give the following continuations: 1. Rxe5? Ne8!

A1) White can act quickly with 2.d5 cxd5 3. cxd5 Nc7 4. dxe6 fxe6 and white has the advantage, but it is a long road to converting this sort of position. Black has clearly found a useful role for his knight. If there is nothing better I would choose this line.
A2) White can wait with 2. Be1 Nd6 3.b3 Ne4 and the knight has become quite the nuisance for example 4.g4 is met with Ng5 with the idea of Nf3 equalizing.

White is not worse in either of these lines, but I am not impressed.

B) 1. dxe5: Immediately we notice that the f6 and d6 square have been taken away from the black knight, also we can now defend any attempts on the king side with a move like Rd4. Also we should realize by now exactly what did the e5 square bring us in line A? Nothing! It accomplished no goal, unless your idea was d5. So we should now again ask how will black get the knight into the game? Then we must ask what can black do? This is how I came up with the following variations:

B1) 1…Na6 2.Rxd7 Kxd7 3.Rd1+ Ke7 4.a3 Nb8 and white has a strong advantage you can follow up with something like: bxc5, rb1, rb7.

B2) 1…Rdg8 2.Rxd7 Rxd7 3.g4 b5 4.b3 (Of course not 4.cxb5 conceding the d5 square.) Kb7 5.Rg1 white has a strong advantage, it is not so clear how black can continue, his knight is stuck, and the rook has no methods of penetration. While white is invading on the kingside.

B3)1…f4? 2. Rxd7 Kxd7 3.g4 and white has very big advantage, the f4 pawn is a new target for white.

B4) 1…Rxd1 2.Rxd1 Rd8 (2…c5 3.a3 followed by b4; 2…h5 3.Rd4 +/-) 3.Rxd8+ Kxd8 4.b4 when we enter an ending of bishop vs. knight and white has very good prospects of winning this position. The bishop is clearly superior, and there are many weaknesses in black’s position. The most important thing is that for black to get his knight out now, he must create a weakness through which the white king can penetrate.

I know there are not enough variations to back up this analysis, but it should give you an idea of the difference between the two moves. We can summarize: do not go into a position like this thinking one move is better than the other without weighing all the positional factors. Garry Kasparov can look at the position and rapidly say what the correct move is, but then again we are not Garry Kasparov. If you would really like to understand this position I would recommend analyzing this position more thoroughly and trying to find holes in my analysis or reasoning. Do not just say: “I do not agree” put some time into it and back it up with calculation. We cannot find the truth in a position without putting our idea’s up to scrutiny, so scrutinize away!

--Ashish Vaja

Tom Fogec said...

Outstanding analysis by Ashish Vaja. That was like a chess lesson !!

Jim Coons said...

I like taking with the pawn, and then there is a rather nice but slow plan for white of exchanging a pair of rooks on the d-file and playing the remaining rook to e3-f3-f4-h4

Jim Coons said...

Another good plan after taking with the pawn, is exchanging a pair of rooks on the d-file and playing the remaining rook to e3 then instead of playing it to f3-f4-h4 playing g4 which is bound to create weakness for black. It seems like black needs to get his king to the king side in a hurry to defend the weak pawns which result.

In either case I really prefer taking with the pawn.

Patrick Van Dyke said...

Nice analysis in general by Ashish here of this position.

I do have a question about the analysis in B1, however.

In looking at 1. dxe5 Na6; 2. Rxd7 Kxd7; 3. Rd1+ Ke7; 4. a3, is that really the intended move?

It seems like Black can respond to this move with 4. ...Nc5!, intending entry into e4 as in the lines in Variation A. 5. f3 wouldn't work to prevent such entry, because g3 hangs.

Perhaps the intended move was 4. b4 here?

But even in the event of 4. b4, the repositioning of the knight to d7 via b8 makes the knight significantly more useful than it was before (at the expense of giving White additional space). It could be used to discourage queenside pawn action from white, it pressures the e5-pawn, forcing White to pay some attention to this isolated pawn [even though it's not isolated in the strictest sense of the definition, the f-pawn really can't conveniently come into play to guard it in the near future], and could possibly work to consolidate some of the kingside/central pawn mass with an eventual ...f6 (supported by the Ke7 and the Nd7).

I point this out, not because Ashish is incorrect to assign advantage to White, or even to recommend 1. dxe5, but to point out that even with the best move, White still has a complicated battle ahead to convert the point.

Frankly, I worry that placing a pawn permanently on e5 removes the possibility of using that square (and accordingly, much of the long diagonal) for the bishop, and think that placing a pawn here invalidates a lot of White's advantage.

* * * * *

As to the point by "southwest chess club", the idea of easing the cramped position doesn't apply so much in endgame positions such as these. The cramped nature of Black's position is due to the lack of the knight's mobility, and exchanging the rooks doesn't change this.

Rather the endgame principle for the side that is behind (and in this case, the limited mobility of the knight and the isolated nature of the h-pawn is enough to put him behind) states that one should trade pawns, not pieces. Black can't completely give up the d-file after 1. dxe5, and will likely have to exchange one set of rooks, but should likely keep the other set of rooks to make use of the semi-open g-file to hinder White's plans.

One should also note that the "funny pawn structure" on f7/e6/f5 should really only prove to be a factor if White can put pressure on any point. That pressure can only be applied by playing d5, and is rendered somewhat moot if the e-file doesn't remain semi-open for White's use. Since immediately opening the file (1. Rxe5 Ne8, 2. d5!?) doesn't pay immediate dividends, and delay is probably not possible, this really isn't a significant consideration for the short-term.

Patrick Van Dyke
SWCC Club Member '99 - '02

Anonymous said...

Ashish Vaja said...


Yes, I am sorry that was incorrect, I meant b4 instead of a3, but a3 is not as easy to dismiss as you may think. I analyzed a3 for some time before going to b4 in that position. We must not forget the idea of exchanging one advantage for another, but it turns out according to my analysis black can draw this ending. The only thing is black must play very actively and precisely at multiple stages in this rook ending, so white is clearly better, just not winning.
I do not understand how the knight is "more useful" it plays no significant role. Can you please provide some sort of analysis showing where it is more useful, thanks.
Note: if you are over 1600 and are working on endings, I would reccomend spending 3+ hours analyzing this ending. First go through the possibilities and main plans that can arise for both sides, this will take 20 minutes or so. Then go a little deeper and find the critical positions which must be analyzed in depth, for example in the line I was talking to Patrick about with Bb4 with the resulting rook ending, this would be a critical position to look into. This will take between 1-2 hours. You just laid the foundation for what must be done in the last part, you do not need to find the truth behind the positions just figure out the "turning point" positions or position that decides wehther one line works or not. Then tear apart those critical positions and try to find the true evaluation of the position this will take 2+ hours.
This seems like a lot of work, but will pay off many times in your games if you are to ever reach a similar position. I have created 10 training positions for my students, and even more calculation/assesment excersizes from this ending, so its instructional value is huge.

--Ashish Vaja

Anonymous said...

Quick glance from a patzer I'd play dxe5 with thoughts of holding black's e and f pawns later. Your bishop is not a bad state at this point.

Jeff Cooper said...

Computers might like strange moves but flesh and blood play demands Rxe5 because the decision tree it hands Black makes it increasingly difficult for Black to deal with all of White's potential breakthroughs. In a nutshell, you optimize your Rook positions, keep your pawns on the light squares and your position (almost)plays itself. (Jeff Cooper)

Raven said...

I'd take with the Rook without any hesitation.

First, in the endgame to follow, you have the clear advantage.

The pawns on opposite sides of the board and you have a Bishop vs. his Knight.

Your pawn structure is better than Black's.

The rooks shouldn't be exchanged off just yet, as they can force further concessions out of Black and continue to pressure his pawns.

You also maintain your pawn duo for an eventual pawn push.

Your King is more centralized and opening up the D file grants Black control of the D file with Rgd8. Then the King is blocked out for the time being.

Under cover of your central pawns, your King can jet over to the Queenside if needed or infiltrate into the various Black holes in Black's central pawn mass.

Raven said...

Hi Jeff. Glad to see you are still around. We may meet again, as I have decided to return to chess after a decade off.

Bob W said...

I have been looking at this for a while . . . after reading ashish's lesson it is hard to disagree with him (and I won't). I guess this position is an example of how a weakness is meaningless if it can't be taken advantage of, and dxe5 seems to claim those weak f6 and d6 squares.

In my first look I liked dxe5 because of a possibility of lifting a rook to the fourth rank would be limited by Rxe5.