Engines vs. Classical Chess Principles |
| Written by Administrator | ||||||||||||||||
| Wednesday, 18 September 2013 | ||||||||||||||||
 By GM Lars Bo Hansen, PhD, MBA  In the most recent issue of New In Chess, Jan Timman notes an interesting observation he has made: I have noticed that young top players make a habit these days of playing weakening pawn moves for no clear reason. As examples he shows excerpts from the games Grischuk-Mamedyarov, Beijing 2013 (26g5); Kamsky-Karjakin, Beijing 2013 (10h5); and Kunin-Edouard, Helsingør 2013 (22 b5). I tend to agree with Timman; I too find these moves surprising or even doubtful. Take a look at these positions and make up your own mind (without computer analysis).  Grischuk (2780) - Mamedyarov (2761), 2013, 1-0 26...g5  Kamsky (2763) - Karjakin (2776), 2013, 1/2-1/2 10...h5  Kunin (2502) - Edouard (2662), 2013, 1/2-1/2 22.b5 But curiously, when running the games through his analysis engine, in each case it turned out that the analysis engine was much less critical of the pawn moves than Timman had expected. It should be noted that two of the three games (Kamsky-Karjakin and Kunin-Edouard) were later drawn; only Mamedyarov eventually lost. View the games It may very well be that Timman and I, both trained in classical chess principles (such as not weakening your pawn structure unnecessarily) in pre-computer times, over-emphasize the consequences of the weakening pawn moves in these specific examples. But more generally, Timmans observation leads to an interesting question which I have also addressed in some of my books, e.g. The Lost Art of Common Sense in Chess: what is the impact of powerful analysis engines on the way top Grandmasters make decisions and is the technological progress only positive for chess? As with most new technology, progress comes with some caveats. Clearly, the emergence of databases and analysis engines has contributed greatly to chess progress. For example, it is much faster to access games these days which has made it much easier and faster to study chess. And because analysis engines by now have pointed out so many exceptions to the old classical principles, we have come to understand chess at a much higher level than before and realize that weird-looking candidate moves should be considered. But the fact that classical principles apply far less frequently than we may have thought in the past does not mean that they never apply. We should take care not to become slaves of analysis engines and make sure that we still apply common sense to chess. Houdini and other engines are incredibly strong but (still) not always right, especially when it comes to transferring the engines suggestions to competitive chess between two humans! In my view, the best (human) chess players of the future (and present) will be those who manage to strike a fine balance between exploiting databases and analysis engines in the most optimal way while combining these insights with the great powers of human intuition, common sense, and reflection. If you like the article, you can learn more about GM Lars Bo Hansen & his books at Amazon.com Related materials: Accumulating small advantages Patterns & biases h2-h4 revolution How to beat higher-rated players Rook and pawn vs. rook Thinking in schemes Does the "Draw with Black, Win with White" approach work anymore? Boris Gelfand & maintaining a strong center How to react to a chess novelty A lesson from the Ukrainian Chess Champion Carlsen-Anand @ Tal Memorial Strategy of Restriction
Write Comment |
||||||||||||||||
| Last Updated ( Wednesday, 18 September 2013 ) | ||||||||||||||||
| < Prev | Next > |
|---|