Quite an insightful post. As my chess understanding is increasing, I am seeing the importance of this principle even more in my games. Material actually is a comfort zone for us humans as it gives us something tangible to hold on to in a game full of complexities. That's why I heed Kasparov's advice of playing gambits as beginners(Not always but from time to time) as they allow one to take risks and let go of material in exchange of rapid development and more dynamic chess play. I had a game today where I employed the smith morra against a Sicilian. Even though I was down a pawn, by Move 10 I had a big advantage (+2) due to strong activity of my pieces. I won comfortably after that.
In the newest AlphaZero paper they find that there is a huge change when it learns the value of the pieces, and then it mostly learns all the other concepts after that. So it does seem there is something fundamental about material, not just for us puny humans, and you may have to learn it before you can unlearn it.
When one can accurately see 30+ ply ahead as computers can, or can calculate at world championship level as the top players do, this emphasis on activity is a fine thing. When one is a mere mortal club player, the most gains are to be found in emphasizing accuracy in analysis and maximizing material. Strategy (of which activity is a major part) is the tie-break in choosing candidate moves in equally safe positions -- safety considerations dominate in almost all human games up to about master level.
They are different concepts. You can have wildly active pieces, all of which are en prise and thus not safe, or your king could be subject to an unstoppable mate threat, and have very active pieces. Activity is clearly very important, but only in the context of not losing material or getting checkmated. Unless, of course, you can see 30+ ply ahead with perfect accuracy, then by all means, emphasize activity within that context.
I think activity is a useful idea well below 30 ply. If my pieces are more active than yours, it's more likely that I will be able to do something nasty to you, and less likely that you will be able to do something nasty to me, even if I can't see exactly how yet.
Don't get me wrong. Activity is of course useful. After having written this to you yesterday, the universe (or the chess gods, or whomever), having a sense of humor or justice or something, conspired to give me a game where I was at a material deficit, but had basically full development and activity of all my army, while my opponent's pieces were almost all cramped, undeveloped, pinned, and defensive. This occurred via an initial miscalculation on my part, wherein I ended up trading a my knight for a pawn, but his "extra" knight was pinned for most of the game and his pieces were tied down to defending it (valuing material!).
I completely understand that material that can't do anything useful should generally not be counted (or counted at considerably less than Reinfeld values) in an evaluation function.
Despite considering myself losing, I went on to win the game. (Not because I was able to keep up the pressure and mount a winning attack by force, but because my opponent's attention lapsed for one move and he allowed a tactic where I won his Queen for a Bishop). In the end, the game was decided on material superiority brought about by the very human foible of not considering the opponent's replies to his candidate move.
My main point was always that: from a practical standpoint, it is very difficult for humans, especially below-master-level players, to win in such positions (down material, up activity). Of course, in selecting candidate moves, finding moves that maximize one's own activity at the expense of the opponent's activity is definitely very important.
But in emphasizing how club players can most improve, there is no doubt that the maximal bang fo the improvement buck is found in getting tactically stronger and developing skill in calculation/analysis, together with a thought process that avoids losing material for no compensation.
Quite an insightful post. As my chess understanding is increasing, I am seeing the importance of this principle even more in my games. Material actually is a comfort zone for us humans as it gives us something tangible to hold on to in a game full of complexities. That's why I heed Kasparov's advice of playing gambits as beginners(Not always but from time to time) as they allow one to take risks and let go of material in exchange of rapid development and more dynamic chess play. I had a game today where I employed the smith morra against a Sicilian. Even though I was down a pawn, by Move 10 I had a big advantage (+2) due to strong activity of my pieces. I won comfortably after that.
In the newest AlphaZero paper they find that there is a huge change when it learns the value of the pieces, and then it mostly learns all the other concepts after that. So it does seem there is something fundamental about material, not just for us puny humans, and you may have to learn it before you can unlearn it.
When one can accurately see 30+ ply ahead as computers can, or can calculate at world championship level as the top players do, this emphasis on activity is a fine thing. When one is a mere mortal club player, the most gains are to be found in emphasizing accuracy in analysis and maximizing material. Strategy (of which activity is a major part) is the tie-break in choosing candidate moves in equally safe positions -- safety considerations dominate in almost all human games up to about master level.
Maybe having more active pieces helps keep you safe.
They are different concepts. You can have wildly active pieces, all of which are en prise and thus not safe, or your king could be subject to an unstoppable mate threat, and have very active pieces. Activity is clearly very important, but only in the context of not losing material or getting checkmated. Unless, of course, you can see 30+ ply ahead with perfect accuracy, then by all means, emphasize activity within that context.
I think activity is a useful idea well below 30 ply. If my pieces are more active than yours, it's more likely that I will be able to do something nasty to you, and less likely that you will be able to do something nasty to me, even if I can't see exactly how yet.
Don't get me wrong. Activity is of course useful. After having written this to you yesterday, the universe (or the chess gods, or whomever), having a sense of humor or justice or something, conspired to give me a game where I was at a material deficit, but had basically full development and activity of all my army, while my opponent's pieces were almost all cramped, undeveloped, pinned, and defensive. This occurred via an initial miscalculation on my part, wherein I ended up trading a my knight for a pawn, but his "extra" knight was pinned for most of the game and his pieces were tied down to defending it (valuing material!).
I completely understand that material that can't do anything useful should generally not be counted (or counted at considerably less than Reinfeld values) in an evaluation function.
Despite considering myself losing, I went on to win the game. (Not because I was able to keep up the pressure and mount a winning attack by force, but because my opponent's attention lapsed for one move and he allowed a tactic where I won his Queen for a Bishop). In the end, the game was decided on material superiority brought about by the very human foible of not considering the opponent's replies to his candidate move.
My main point was always that: from a practical standpoint, it is very difficult for humans, especially below-master-level players, to win in such positions (down material, up activity). Of course, in selecting candidate moves, finding moves that maximize one's own activity at the expense of the opponent's activity is definitely very important.
But in emphasizing how club players can most improve, there is no doubt that the maximal bang fo the improvement buck is found in getting tactically stronger and developing skill in calculation/analysis, together with a thought process that avoids losing material for no compensation.