# What makes game 5 of the 2018 World Championship a draw?

I was watching recently the excellent ChessNetwork commentary of the 5th game of the WCC 2018 between Magnus Carlsen and Fabiano Caruana. The end position looked like this and players shook hands after white played g4.

``````[FEN "8/5R2/5bp1/3rpk1p/8/4B2P/5PP1/5K2 w - - 0 1"]
``````

There is not much to grip onto on either side for sure, but it doesn't strike me as a draw either. The line I get with Stockfish is not straight forward.

``````[Title "{ [Stockfish 280218 64 POPCNT] 68:+0.00}"]
[FEN "8/5R2/5bp1/3rpk1p/8/4B2P/5PP1/5K2 b - - 0 1"]
[startflipped ""]

1... Ke6 2. Ra7 Rd6 3. Ke2 hxg4 4. hxg4 Rc6 5. f3 Bd8 6. Rg7 Kf6 7. Rg8 Bb6 8.Bh6 Rd6 9.Rf8+ Ke6 10. Re8+ Kd5 11. Be3 Bd8 12. f4 exf4 13. Bxf4 Re6+ 14. Rxe6 Kxe6 15. Kd3 Kd5 16. Ke3 Be7 17. Kd3 Bb4 18. Ke3 Bc5+ 19. Kf3 Be7 20. Be3 Bd8 21. Bf4 Be7
``````

Is there any simple explanation why this position appears as an obvious draw other than a lack of goals to pursue on both sides?

Some points to consider:

• No passed pawn
• All rook endgames with equal number of pawns in its own half are drawn (I do not know if this is a theorem, but it is almost a theorem).
• One of the players can probably force exchange of bishops and get it to rook endgame with equal number of pawns.
• Both players are 2800+ and it is a classical game, so enough time to think and make the right moves.

EDIT: You can look at the game discussed here, where one side has Rook along with 'a' and 'c' pawns, while the other side just has a rook and the game is still a draw.

Computer engines are of little use in positions like this. Still it is a pretty obvious draw, because of:

• equal (and very limited) material
• pawns on the same side of the board
• no weaknesses for either player
• active and safe kings

In order to win this, one would have to win material or promote a pawn, neither of which can be forced.

• I disagree computer engines are of little use in positions like this. Leela maybe, but Stockfish is spot on with the 0.00 eval. Commented Oct 25, 2019 at 5:29