Hello and welcome. I am Polugaevsky, and I am glad to sit down for a game with you. Set up the pieces, choose your color, and let us see what ideas appear on the board. Good luck, and may the game be interesting for both of us.
Polugaevsky is a Chessiverse bot that simulates the style of Lev Polugaevsky, a Russian chess legend. It is an independent historical simulation, not the person himself, and it is not endorsed by the player or his estate. The bot's estimated Chessiverse rating is 3113, which places it at super-grandmaster and engine tier, strong enough to beat nearly all human masters. That figure belongs to the bot alone and should not be confused with the historical player's FIDE rating or his actual decisions at the board. The bot's repertoire on file centers on Indian-style structures. With White it has 741 games in the Indian Game mainlines, and with Black it has 407 games in the Najdorf, plus 276 in the Queen's Indian and 186 in the Nimzo-Indian. Its profile shows an aggressive temperament, a taste for keeping tension, and a slight lean toward sharp positions. Anyone sitting down against Polugaevsky should expect precise, near-flawless play with active pieces and very few easy exits from complexity.
How Polugaevsky plays
Polugaevsky leans aggressive (0.55) and prefers to keep tension rather than trade down (0.65 toward complicating). Its positions are mildly sharp rather than extreme, and its opening choices sit near the middle between mainline theory and offbeat ideas. Typical battlegrounds include the Najdorf as Black and the Indian Game mainlines as White, where the bot builds pressure and punishes inaccuracies.
Who should play Polugaevsky
Players rated roughly 2400 and above, including titled masters and serious preparation students, will find Polugaevsky a fitting test at 3113. Because the bot is aggressive and keeps tension, it suits players who want to practice defending under pressure, calculating in complex middlegames, and converting scarce chances. Lower-rated players can still use it as a benchmark, but should expect to be outclassed and treat each game as a learning exercise.