Improving Evaluation

Sort:
ThreeSteps
Martin_Stahl wrote:

I actually have the book but haven't read through it very recently. I'll have to see if I can find that part.

 

Page 20 is particularly relevant.

In terms of improving evaluation, I'm gradually going off using engine analysis since they have no concept of how hard it is for a human to play a particular position.  There are many positions that engines regard as equal but yet it is much harder to play for one side than the other.

Instead I'm finding the study of GM games to be helping me.  I recently looked at an Alekhine game where I underestimated a passed pawn and overestimated completing development in that specific position. Similarly I look at what pawn structures Karpov allows in his games, etc.

One of my weaknesses with evaluation is letting one factor dominate too much, such as overvaluing the bishop pair while ignoring a loss of time to maintain them.  Or choosing a line that removes a weak pawn of mine while forgetting that it also removes one of my opponent's weaknesses too.  Of course, every position is specific so I'm not suggesting general rules.

Martin_Stahl
dodgernation wrote:

Those steps actually will help you with evaluation a position.  Afterall, what is evaluation a position? 

Material: The first advantage/disadvantage most people learn. Simple piece counting. A consideration of all position, but only one of many. You can have all the material you want, but it is useless, if it is undeveloped and can't come to the king’s defense in a time of need! Sounds strangely like making sure your pieces are safe, and actively placed.

Pawn Structure: Look for ways to damage your opponents pawn structure and always be aware of threats to your own. Too many pawn islands, isolated, and double pawns are bad. #1 through #4 could apply here.

 

Development: Having better development can often lead to having an initiative that can carry over into the middle game and sometimes even the endgame. #4 piece activity

King Safety: This one can NOT be over stated. If your king isn’t safe, what does it matter if you have the better pawn structure or more material! Be very careful about any pawn moves in front of your king. Once a weakness is created pawns can’t move backward so you have to live with it.  #1-3-4 apply here

Space: Both an advantage and disadvantage. Having more space often leads to greater attacking chances because of the greater freedom of movement for the pieces. #4 piece activity

Control of Center: Very closely related to space. A piece’s value is not an absolute. If a piece control good squares it becomes valuable. #4 again...piece activity

1. Make sure all your pieces are safe - This will help you to examine the entire board.  It makes you see where all the pieces are, and any threats.  You cant evaluate a position, unless you can see any, and all threats first.

2. Look for forcing move: Checks, captures, threats.  You want to look at ALL forcing moves (even the bad ones) this will force you look at, and see the entire board - Again...you are forcing yourself to see the entire board.  You cant evaluate the postion, if you arent seeing the position, and what it holds.  Forcing moves should be played first, as they limit what your opponent can do.

3. If there are no forcing moves, you then want to remove any of your opponents pieces from your side of the board - If your opponent has pieces on your side of the board, that increases the chances of tactical possibilities.  Any and all evaluation cant be accurate if you fail to see a mate in 1 or 2 or 3 or...

4. If your opponent doesnt have any of his pieces on your side of the board, then you want to improve the position of your least active piece - A quiet position can still hold many surpises.  But if it doesnt, and you dont immediately have any tactics, attacks, etc. Then improve the activity of your least active piece. 

All good points and certainly will help with analysis and calculation. I'll have to see if I can't combine ideas in a way that will help me determine the overall evaluation at the end of that. Because, once the position reaches a static (or less dynamic place) and I'm not concerned yet with finding another move in the line, I need to be able to try and figure out who is better and approximately by how much.

I'm not even thinking I can get near 100% accuracy, but I'm hoping I can get a lot closer than I am right now. I don't mind being one step back so much, it's when I have the complete opposite evaluation of what the position is like that concerns me.

Martin_Stahl
ThreeSteps wrote:
Martin_Stahl wrote:

I actually have the book but haven't read through it very recently. I'll have to see if I can find that part.

 

Page 20 is particularly relevant.

In terms of improving evaluation, I'm gradually going off using engine analysis since they have no concept of how hard it is for a human to play a particular position.  There are many positions that engines regard as equal but yet it is much harder to play for one side than the other.

Instead I'm finding the study of GM games to be helping me.  I recently looked at an Alekhine game where I underestimated a passed pawn and overestimated completing development in that specific position. Similarly I look at what pawn structures Karpov allows in his games, etc.

One of my weaknesses with evaluation is letting one factor dominate too much, such as overvaluing the bishop pair while ignoring a loss of time to maintain them.  Or choosing a line that removes a weak pawn of mine while forgetting that it also removes one of my opponent's weaknesses too.  Of course, every position is specific so I'm not suggesting general rules.

I flipped through the book and found that I think. Where he talks about analysing positions from games that already have some analysis and then comparing it to what lines were given. Pretty much the same idea.

vanessagdias

Nice thread! :3

Martin_Stahl
Martin_Stahl wrote:
axersa wrote:

I too would be interested to read your evaluation of the positions. Could you update this thread with your blog post once you're done?

I know you're playing online games but have you considered playing slow live games as well? I find the time pressure helps to focus, and after all, most strong masters improved their calculation and evaluation primarily through playing.

Also, I believe they are called Stoyko exercises after FM Steve Stoyko.

Yes, I'll post an update to this when I do the blog entry. Though, I'm sure the quality isn't that great (for my first one) and I already know my evaluations weren't the best.

 

I posted the blog in 3 parts. Most of part 1 and all of part 2 is the exercise and part 3 is just my follow-up thoughts to doing it.

Not even close to the best but it was interesting and I plan on doing more (though I probably won't post them Smile). Here is the first part.

http://www.chess.com/blog/Martin_Stahl/taking-a-stab-at-stokyo-part-1

RoobieRoo
dodgernation wrote:

 

 
Now your mission if you chose to accept it.  Find the move that Aronan played.

What is the solution?

maxwalker2003

I think Rfd1? Am I right?

RoobieRoo
maxwalker2003 wrote:

I think Rfd1? Am I right?

 

well lets think about it Max and put the authors theory to the test,

Checks? Nf6, Qxf7 both do nothing

Captures? Qxh6, Nxe6, Bxe6, Nxc6 all do nothing

Attacking moves? a3 attacks the bishop which is on our side of the board

Improve the position of the least active piece? Rfd1

So it appears if we have followed the theory then either we should consider a3 or Rfd1.  Is it not the case?

33eric2008

              Please be relevant, helpful & nice!

TheAdultProdigy

Hi, Martin. I have one suggestion that may help. One issue I think some have when going through annotated Master games is not being mentally active enough or being as philosophically inquiring as necessary about why some moves are if aren't good. What I do to develop better evaluation skills, and this is also something the ICS does, is: get thoroughly annotated games (e.g., Chernev's Logical Chess), then get the PGN of the first game and blank notebook paper, and begin writing down every aspect of the position that you think is relevant, candidate moves suggested by your intuition and by features of the position, write out variations stoyko style, and then evaluate the position before moving to the next move in the PGN. You should have guessed the move for the side on move and the opponent's response before rolling to the next move. Be sure to explicitly write the evaluations on each move. If you are thorough enough, it could take you 10 to 15 hours on each game. Doing just s couple of games will bring your time down a great deal, because, in my experience, you learn quickly what to calculate, what not to, and you grow your sense of "what matters most in a position," to use Soltis' words. I am at the point where I am simulating long game controls when going through Master games in this way. When you are all done, look at the Master's annotations, see what you wasted time on (or didn't, as you might be more thorough than the annotator), and compare what the Master thought mattered most vs your assessments.

Martin_Stahl

Thanks. Just need to find more time in the day

TheAdultProdigy

Oh, just to be clear, you don't have to make it through the PGN in one sitting. But yeah, time is a resource that is limited.