Crazy perfect! Fabi, don't fear the Reaper!

Great play by fabi, not perfect according to the silicon monsters, but these are humans who miss one or another perfect move sometime. Fine analysis Josip, as usual you give us an insight into the strategical ideas behind the scanes. thanks and keep up the good work.

Nice

Thank you once again for a great analysis. Basically, Fabiano Caruana did not have an advantage in the endgame anymore, he lost his advantage before. Please allow me a piece of advice: If there are only 7 pieces on the board, I would suggest that you use 7 men tablebases for your analysis, each position has a predetermined outcome. “Normal” computer analysis will sometimes not give you the correct picture of some specific endgame positions with 7 men. This position (time stamp: 17:50) was drawn. Then MVL blundered in this position (time stamp: 18:52) with Nh6. Fabiano gave back the favour in this position (time stamp: 19:06) with Kg4 and the position was again drawn.

Fabi, the baby faced killer. Prob not so good with a shovel or hammer but give him a couple of Kings, one to defend and one to destroy and suddenly you get Super Boy.

Nepo vs Giri was a bit anti-climatic but made sense in the context of the tournament. I think it was lucky for us spectators that Fabi was out for blood today.

WARNING! A long Explanation is coming. =) I am sure that Fabi prepares the line Bc4 sacrifice @ 9:33. The Bd6 that engines preferable does not give an advantage for Fabi as it only gives equal for both sides. He needs to bust the Sicilian-Najdorf of MVL and proves that is a bad opening. Most commentators and other YT analysts including Magnus like the idea of Bc4 preparation. As the commentators also point out that the current Chess Engines does not like the Bc4 line in the meantime. However, sure it will be the top choice in the upcoming future development of AI. These AI's today are probably processing their algorithm deeply on how the Bc4 will effectively work in the upcoming years. MVL probably hoping Fabi to play the engine's suggested line Bd6 which gives the game an equal match if not an advantage. As I watch the game, Magnus, David, and Tania spotted this line too. However, commentators point out that it gives no advantage to Fabi for winning a guarantee in the first place. It may also result in a much favorable advantage in the blacks' camp and squeeze entirely the position of the white in favorable cases. The line suggested by the engines is pure taboo. Fabi did the right thing to play Bc4 as it im-balance the game without guarantee of the draw and advantage for the black, and it did capitulate the game entirely. The Bc4 line is a sign that Fabi has opted for the win as he is behind points of MVL in round 8. The Bxd4 suggestion @ 11:10 has a significant drawback in terms of the long-term plans. Leaving the dark-squared bishop at d7 leads to pure disaster for the Black and the result will be an instant win for Fabi. Your Bxd4 suggestion is a bad choice. Let say after Fabi plays Rxd4 gaining tempo and deflecting the white's queen from c4 to b3 (Qb3), castling will be a guaranteed win making room for Rook to create a checkmate pattern at f7 weak pawn and Black could not castle because of the guarding bishop at d7. If Black plays Nf6 covering the weak f7 pawn, White will capture the defending knight (Nxf6-xf6) resulting in an oblivion checkmate pattern sequence of the king-hunt method (Qf4-f5, Qe5-Rg8, Qf6-Kd7, Qe7-Kc6, Qc7-Kb5, Qc5# checkmate). Black may survive in some instances very short but not enough such as (Nxf6-xf6, Qf4-Nd7, Rc4-Kd8, Bc7-Ke8, Qd4-Rg8, Rxd6-Nc7, Qf4-Rf8, Bd7....) were white is clearly winning giving black a near zugzwang position. We called this the "Deceased King". If this happened, the game ended shortly because of the black queen's greed and pride at b3. Therefore Bxd4 is not an option to play. It is likewise the reason that MVL plays immediately Nf6 instead of capturing the knight at d4. It is possible that MVL also takes consideration of Bxd4 but he did not choose because of the danger infiltrating at its kingside of the board. Lastly, with the condition of pawn structure of the board @ the endgame, this is in a fortress situation and clearly result in a drawn endgame. The reason is that the pawns at kingside are facing near to each other (g2 & f6) + rook and knight battle. MVL did a terrible blunder @ 18:54 with Nh6. He should play Ng7 instead to claim a fortress position. Fabi cut off the king connection of the pawn at g-file and as well the knight by infiltrating its rook at g3 allowing its king to sneak at the g-file and h-file disconnecting the king-pawn-knight connection slowly but effectively. It is predetermined in this positional case that if the black pawn is along at a-b-c file, it is automatically drawn by a fortress even with the presence of pawn at g2. However, if the black pawn is along d-file or e-file, black is winning even with the presence of g2 pawn of the white. Black will surely make resourceful use of his knight to march the pawns toward for promotion. I learned this before from a strong national master. It is also lucky that one of the KZfaqr GM mention also this Table Base principle. =)

Bc4 is not at all best of the moves i put it in engines observations are interesting it started at -2.3 after sometime settled at -1.5 then after long time it settles at -0.5 really dangerous preps by Fabiano(the prep god)

Ng7 may be create a fortress.king can not enter