I’d think that the best chess players would be able to create their own best strategies/moves that the world hasn’t seen, instead of memorizing/replaying strategies that have been documented by older Grandmasters.
Also, how complicated is chess… and why is it so important to study/analyze past plays, and if everyone is just following one play out of the set of documented plays, doesn’t that automatically make the game more predictable/less exciting?
Finally, how good is Carlsen and what makes him special… I’m trying to understand his genius compared to the others?
In: Mathematics
Actually one of the ways tournaments try to detect cheating is by analyzing the originality of the move. Basically the more original or creative the play was the more likely it was done by a computer algorithm then by a human player. So basically in chess originality is considered a trait of a computer player.
As a laymen and certainly no where near an expert at chess, I think what it basically boils down to is: millions of humans have analyzed chess over hundreds of years. Over the past few decades, we’ve been using computers to analyze the game as well.
Some games – like tic-tac-toe – are referred to as “solved” games. That is, we’ve analyzed every possible choice you can make in the game, and come up with the “optimal” strategy. For example, if played correctly, [x will always win (EDIT: or tie) at tic-tac-toe](https://xkcd.com/832/).
Chess is not a solved game – there are too many possible combinations. But parts of the game are partially solved. For example, at the beginning of the game there is a very limited set of moves (you can either move your pawns, or move your knights, and that’s it). This means that we’ve had plenty of time to study many many different openings and how to respond to them, and we understand the weaknesses and strengths of various choices.
Since every game starts with pieces in the same position, you can analyze past games to understand what are “optimal” moves. Since we’ve played so many games, it’s likely that your game will take time to deviate from a game that has been played in the past. So, if you’ve studied those past games and understand what moves are good and what moves are bad (based on the analysis that we and computers have done), you can rote memorize the optimal play from a specific position that has been played in the past.
But, it’s still important for us to analyze past games (or to think about the games we’re playing now) – especially new games – because it allows us to check for even better moves that no one has thought of before. Some of the excitement of chess comes from the fact that you don’t know exactly when one of those moments of genius will occur. And as games progress into the mid and late game, those moments become more and more common because the positions are less likely to match a previous game.
> everyone is just following one play out of the set of documented plays
That is definitely not happening. Or rather, it might be happening, but VERY rarely. Chess has a huge number of (viable) variations, and the moment you opponent does one move different from your memorized play, it will play out very differently.
Almost every game by good players will be something the world “hasnt seen” (aka none of the players have the moves memorized, in the huge pool of games that are played there might ofc be some that are exactly identical).
The initial phase of the game – “the opening” – works kind of like that (because it starts from a fixed situation, so there isnt so much variation yet. This phase lasts maybe 3 to 8 turns.
> and why is it so important to study/analyze past plays
You try to understand general strategies for long term plays. You also try learn to recognize situations that allow you to apply certain tactics. Looking at games from good players and trying to understand their moves is a good way to do that.
Cant comment on the Carlsen-question.
A lot of chess boils down to making your opponent choose the lesser of two evils. They either block with a pawn and lose their bishop, or make a different move and lose their queen.
The previous working strategies and tactics are established, efficient ways of backing your opponent into having to make these choices. If you don’t use them, you risk falling into them yourself. It’s a part of chess theory.
That doesn’t mean a strategy becomes the same set of moves in the same order the whole way through– it’s just the same general idea. Unfortunately I don’t follow the game quite closely enough to have an opinion on Carlsen, so I’ll leave that part to someone else.
Happy playing. 🙂 You might also want to look up chess subreddits for a more focused set of answers.
> everyone is just following one play out of the set of documented plays
This is not what occurs in high level chess play.
I highly recommend the youtube channel [agadmator.](https://www.youtube.com/user/AGADMATOR) In most of the games he reviews, he usually comments on if/when a new position is reached in the game (a position that hasn’t been seen before in any of the recorded games in the chess engine he uses).
That being said, there is a lot of theory involved in the game. A lot of times, the first few moves are played by following certain openings. The various openings (there are hundreds, if not thousands) have been studied for centuries, leading to modern day theory on which openings are best, how to counter them, and how to get a good piece or positional lead going into the midgame.
There are certainly times when two high level players will purposefully copy the same moves of a previous game, but they will have heavily analyzed that game to try and figure out why a certain player lost or won. At some point, one of them (usually the one playing on the side that lost) will divert from that game and then it comes down to a little bit of theory and a lot of improvisation.
Most of the time, the midgame is where you find yourself in a completely new game. That’s where it takes the most understanding of chess fundamentals to get/keep a lead.
There are a lot of “solved” endgames, such as how to checkmate with 2 bishops vs a king, where you can play off of memorized moves.
The reason why it is important to study old/famous games is to try and build a deeper understanding of theory. Additionally, you may find yourself in a game that is identical/highly similar to one and if you know how that game went, it can give you a distinct advantage.
Those kinds of things are what contribute to how Grandmasters are able to seemingly outmaneuver lower rated players without them even realizing its happening. Also, modern day chess engines have gotten much better at analyzing games and providing insight into better moves in certain positions. For example, in a [100 game set between AlphaZero and Stockfish 8](https://www.chess.com/news/view/google-s-alphazero-destroys-stockfish-in-100-game-match), A0 almost never used the King’s Indian Defense, leading to a lot of analysis as to whether or not it was a subpar opening.
Latest Answers