Hi everyone,
As a player working on tactical pattern recognition, I've been experimenting with solving puzzles extracted directly from historical Grandmaster games rather than standard generated puzzle sets.
Real GM games feature natural positional setups and piece coordination that frequently happen in actual matches, which feels quite different from purely engine-generated positions.
I'd love to get your thoughts and feedback on a few questions:
1. Do you find real value in solving puzzles from authentic historical GM games compared to synthetic tactic sets?
2. Do you feel calculating tactics from these real setups translates better to your actual over-the-board or online games?
3. What features do you look for when analyzing tactical blunders from master games?
Looking forward to hearing your experience with training on real master positions!
Take a look at the Woodpecker Method.
The positions are all taken from games of all he world champions from Steinitz to Carlsen. Volume 1 is devoted to tactics and Volume 2 to positional play. Aagaard has a book on tactics where all the positions are taken from games with the Sicilian defence. The Chess Informant series always has a section on winning combinations from recent games and these are collected together in their Encyclopaedia of Chess Middlegames.
The goal of puzzle solving is to find the best move and train pattern recognition. They can be composed or from real games, that's your choice. Endgame studies, for example, have a cleaner feel as there are minimal pieces.
Non-existent unless one is at least IM-level, in my humble opinion.
This topic has been archived and can no longer be replied to.