1. Standard memberSwissGambit
    Caninus Interruptus
    2014.05.01
    Joined
    11 Apr '07
    Moves
    92274
    23 Mar '09 22:091 edit
    If anyone has doubts about the merit of the point-count system, just look at computers. It seems to work rather effectively for them.
  2. Joined
    21 Sep '05
    Moves
    27507
    23 Mar '09 22:16
    Originally posted by SwissGambit
    If anyone has doubts about the merit of the point-count system, just look at computers. It seems to work rather effectively for them.
    Computers use numbers for every part of their evaluation... king safety, pawn structure, space, etc. They have no alternative. Humans don't think in such numbers for chess, and I see no reason why it should apply to material except as a very crude guideline for beginners.
  3. Standard memberWulebgr
    Angler
    River City
    Joined
    08 Dec '04
    Moves
    16907
    23 Mar '09 22:201 edit
    Two Rooks are slightly stronger than a Queen. They are slightly weaker than two Knights and a Bishop, and a little more so than two Bishops and a Knight. The power of the Knight decreases as the pieces are changed off. The power of the Rook, on the contrary, increases.

    Jose R. Capablanca, Chess Fundamentals, p. 25.
  4. Standard memberSwissGambit
    Caninus Interruptus
    2014.05.01
    Joined
    11 Apr '07
    Moves
    92274
    23 Mar '09 23:28
    Originally posted by Varenka
    Computers use numbers for every part of their evaluation... king safety, pawn structure, space, etc. They have no alternative. Humans don't think in such numbers for chess, and I see no reason why it should apply to material except as a very crude guideline for beginners.
    Wow.

    The last statement is just completely ignorant. There are so many counterexamples: David Tebb in this thread [a strong master still using point-count], Kasparov saying they taught him two passers on the 6th rank is worth a rook [as if that is still a poignant comment! or maybe he is reverting back to his old 'beginner' ways 🙄], masters analyzing their games to try and prove compensation for the exchange [showing that they obviously know they are 'down' points], GMs laboriously making sure they don't lose even so much as a pawn in many of their games, etc. etc.
  5. Joined
    24 Aug '07
    Moves
    48477
    23 Mar '09 23:28
    Queen = 9 Generally

    It's easy to remember because queen + pawn = 10 and rook + rook = 10. There is much debate whether the queen and pawn or rook are better. It definitely depends on the position.

    Although queen (9) = 3 minor pieces (B + B + N), the queen usually has its hands full. Again, it definitely depends on the position.

    Sometimes two bishops even compensate for a queen.

    The main thing that assigning numerical value to the pieces does is give you some idea of when you are making a better trade (exchange). You really can't depend for other factors.
  6. Joined
    21 Sep '05
    Moves
    27507
    24 Mar '09 00:29
    Originally posted by SwissGambit
    Wow.

    The last statement is just completely ignorant. There are so many counterexamples: David Tebb in this thread [a strong master still using point-count], Kasparov saying they taught him two passers on the 6th rank is worth a rook [as if that is still a poignant comment! or maybe he is reverting back to his old 'beginner' ways 🙄], masters analyzing ...[text shortened]... iously making sure they don't lose even so much as a pawn in many of their games, etc. etc.
    Just show me one example of Kasparov using actual *numbers* during discussion of a game. We're not talking about "two passers on the 6th rank is worth a rook", wer'e talking about if Kasparov said "two passers is equal to 5 points, which happens to be the same as a rook". But of course, the "5 points" part is never part of his thinking. He assesses; he compares; but he doesn't do arithmetic.

    Sure, GMs will sacrifice a pawn and say that they have compensation for the material BUT nowhere in their thinking are they assigning numbers to anything. Computers assign numbers to everything.

    Seriously, do you look at a pawn and think "1"? You don't just look at it and get a feel for it's worth?

    Only beginners sit and actually think e.g. "3 + 3 + 5 + 1 +...." during a game of chess.
  7. Joined
    21 Sep '05
    Moves
    27507
    24 Mar '09 00:38
    Originally posted by paulbuchmanfromfics
    The main thing that assigning numerical value to the pieces does is give you some idea of when you are making a better trade (exchange). You really can't depend for other factors.
    But beyond the stage of being a beginner, experienced players skip doing any arithmetic explicitly.

    Supposing we were to have a fight 🙂 and you could have a gun or a knife, which would you prefer in general? I bet you make your decision without asking the value of each weapon. You just know what each is capable of.

    Similarly, experienced chess players have a feel for the capability of: a queen; two rooks; two doubled pawns; two connected passed pawns on the 6th rank; the fianchetted bishop in the King's Indian Defence; etc. without having to give them all an arithmetic value.

    And this is what most chess players do when assessing a trade. It doesn't involve 1s, 3s, 5s, etc. Sure, they might look to see if they're a minor piece down at the end of a variation, but they're not thinking "-3" as such but instead just "a piece down".
  8. Hollow earth
    Joined
    29 Apr '08
    Moves
    2472
    24 Mar '09 00:531 edit
    Imo the point count is a necessary evil when first learning the game.Later one automatically forgets about it.A pawn is a pawn,not 1 point.Consider a pawn able to deliver checkmate,it's value is infinite.Would you trade such a pawn for a queen?After all,you'd win 8 points! 😉
  9. Joined
    23 Sep '07
    Moves
    23415
    24 Mar '09 01:152 edits
    pawn = 1 or itself
    knight/bishop = 3 pawns
    rook = 5 pawns
    queen = 9 pawns

    naturally you'll see bishops being 3.68 or knights 3.33 or rooks 5.5, etc, etc.. everyone has their own views.. Then you get the people that take 2 million games of grandmasters 2500+ and try to somehow figure out what each piece is worth. The bottom line is it depends on the position.

    As Joel Benjamin has said...3 pieces (ieL 2 bishops+knight are better than a queen) even though it seems like they're worth 9 pawns each. Just that beginners would prefer a single piece the queen as it's easy to control. The 3 pieces are better but you would have to know how to coordinate them properly.

    2 rooks should be about a queen and a pawn...although this depends on the position again, sometimes either one can be a lot better.

    in general if you're giving up the exchange and loosing a rook, make sure you get atleast a pawn. Joel Benjamin says about bishop+1.5 pawns would be a rook, or knight +1.5 pawns to a rook. 1 Pawn would be too little, 2 would be an advantage. Again depends on position.

    We've all seen GM games where a knight or 2 knights dominate a rook or 2 rooks, showing it's all about the position
  10. Joined
    01 Mar '09
    Moves
    1415
    24 Mar '09 01:19
    i don't think the points system for counting material was ever intended to apply to every situation, nor has anyone really advocated for it this way. Even a lot of beginners are familiar with the basic ideas of gambits and sacrifices (I can say this with certainty because I am one).

    it was always meant to be a rule of thumb and it does this job well.
  11. Standard memberAThousandYoung
    or different places
    tinyurl.com/2tp8tyx8
    Joined
    23 Aug '04
    Moves
    26660
    24 Mar '09 03:22
    9.75

    http://home.comcast.net/~danheisman/Articles/evaluation_of_material_imbalance.htm
  12. Joined
    12 Sep '07
    Moves
    2668
    24 Mar '09 07:33
    I haven't looked at ATY's site there, but i think i vaguely remember looking at it sometime ago...
    Anyways, i use
    P=1
    N=B=3.25 (I like Bishops better, but the only reason for that is to sac them at f7/f2 or h7/h2...)
    R=5
    Q=9.75

    note: beginners are usually taught 1,3,3,5,9, but then according to that R+P=B+N. I tend to disagree, two pieces are worth more than a rook and pawn, especially earlier in the game
  13. Joined
    09 Mar '09
    Moves
    27
    24 Mar '09 08:531 edit
    might have asked this before but is an exchange of a queen for a rook and bishop in a fairly equal posistion good?

    i have played this a few times and lost with the queen . it can get outnumbered.
  14. Joined
    19 Jun '08
    Moves
    8115
    24 Mar '09 18:122 edits
    a queen and knight is equal to a rook plus two bishops and a pawn

    two rooks and a bishop is equal to two knights a bishop and 2 connected wing pawns

    a bird in the hand is worth two in the bush
  15. washington
    Joined
    18 Dec '05
    Moves
    47023
    24 Mar '09 18:15
    screw the point system. a queen is worth 3 pieces bishops and knights and thats it. I have rarely seen a good game where one player trades a queen for two pieces and win unless it was a sacraficial line leading to mate.
Back to Top

Cookies help us deliver our Services. By using our Services or clicking I agree, you agree to our use of cookies. Learn More.I Agree