After writing my most recent Endgames Essentials post, I decided to watch the European Individual Chess Championships and stumbled across an endgame that I thought was an effective model of the principles we’ve established thus far.
Before Ernesto Inarkiev managed to pull away from the pack and win the Championships, it looked like the event could go a number of ways – Saric, Navara, Jobava, Kovalenko, and Wojtaszek were all over 5/6, with a bunch of strong players at 4.5/6 with five rounds to go. While it wouldn’t determine the winner, David Navara’s game in round 7 against Baadur Jobava definetly impacted the course of the tournament.
In that game, we reached this drawish position:
Navara – Jobava, 2016
With a symmetrical pawn structure, it seems like not much can happen. There are no logical pawn breaks in the position, and though White’s king is more active than Black’s, Jobava does have control over the d-file. What you might notice though, is that it is White who is pressing. With the e- and f-pawns already advanced, it will be difficult for Black to expand effectively on the kingside and create weaknesses. Sure, White certainly cannot be considered winning here, but it is Black who must prove equality. As we’ve seen in many of Carlsen’s games, this is already enough to play for! To stop Black from entering the second rank, White brings his king to c3, and activates his rook on the b-file. 18. Kc3 Kf8 19. Rb1 Ke7
White has improved his position, but so has Black – in fact his king is coming into the game very quickly! If White isn’t careful, Black can bring his king to d6 and rook to d7 with the hopes of creating a fortress. When you’re trying to improve your position and push your opponent, it’s always important to consider their plans and see if you can stop them. Navara spent 20 minutes on this next, and made the most contesting move on the board. 20. e5!
At an artificial level, this move looks really weakening. White gives up the d5 square for Black’s rook and his structure could get undermined with …f7-f6 ideas. But if you think about it, Black’s rook is no better on d5 than it is on d8 because tactically it must always retreat to d7 following Rb1-b7+. With the pawn on e5 cutting out improving squares for Black’s king, Black will need this …Rd7 resource until further notice. It’s also important to note that a break on f6 arguably hurts Black more than White – it doesn’t improve his structure, and an open f-file wouldn’t change the nature of the position. One thing to remember when playing in equal positions is that in order to play for a win, you must give up something in return. In this case, Navara gives Black the d5 square to keep his winning chances. 20…h5 21. a4
Right now both sides are playing accurately. Black is trying to solidify his kingside, while White is trying to gain access to critical squares on the queenside. For example, were Black to stand idle, White could march his pawn to a6, giving him control of the b7 square for his rook to then win the game. Black’s logic here is that while he may be less active, if he can remain solid, White will not have enough to exploit his advantage. Let’s see if this holds true.
21…Rd5 22. Rb7+ Rd7 23. Rb8 Rd8 24. Rb4 Rd5
So what just happened? It just seems like the two players shuffled their rooks back and forth, but what was the point? As Grandmaster Sam Shankland says, no self-respecting Grandmaster makes a move without a purpose. Let’s go back to the position after Jobava played 21… Rd5:
In this position it is White to move, but in the current position after 24… Rd5 it is White to move. What’s changed? White got in Rb1-b4 and now has an extra tempo to improve his position. By infiltrating deep into the b-file, Black had to block out White’s rook from raiding the kingside pawns, so this line is actually rather forced. While Black should still be able to hold here, it’s small moments like these that count towards building a winning position. 25. g3 Rc5+ 26. Kb3 Rd5 27. c4 Rd2?!
Here’s where Black starts to go wrong. As we mentioned earlier, the rook was no better on d5 than it was on d8, and so the same applies to d2. Black cannot afford to allow White’s rook to enter the 7th rank without resistance, as the a-pawn will fall, and it’s White’s passed pawn that will matter more than Black’s. Still, White can’t play 28. Rb7+ yet, so he makes the one move that wasn’t possible just one move ago. 28. a5!
I’m going to guess that this move’s power was under estimated by Jobava, seeing as he spent 24 minutes on his next move. However after 27. c4 its impossible to effectively stop this pawn push and be able to retreat to d7. Black’s best hope was to create a fortress by retreating to d7 and bringing his king to c8. I messed around with Stockfish here to see how Black would hold, and the line goes 27…Rd3+ Very important – the king is pushed to the second rank before Black makes a bunker. 28. Kc2 Rd7 29. a5 Kd8 30. a6 Kc8 31. Rb1 g6
White will have to try and create a weakness on the kingside, but his rook can’t run too far astray since Black can play …Kc7-b6 and attack the a-pawn. It’s an ugly position to have to defend, but since White’s king can’t get to the kingside thanks to the Black rook, Black should have good drawing chances. So how is this so different than what happened in the game? It turns out that not inserting this one check before retreating to d7 still gives White something to play for with an active king. If Jobava had played 28…Rd7, White’s king can enter the fray through a4, then later b4 and c5 – but admittedly this is very difficult to win. Instead, Jobava offers Navara an oppotunity. 28…Kd8?
The last move underestimated a resource to draw, but this move gives White an opportunity to play for more! While White can’t win material, his rook would be much better placed on f8 or g8 than it is currently on b4. Navara wastes no time in reaching his desired position.
29. Rb8+ Kc7 30. Rf8 Rd7 31. Kb4
A tremendous improvement in White’s position. Black has no easy way to defend the a5-a6 idea, and White’s king is headed to the c5 square, where it cannot be touched by Black! The position still looks difficult to convert, but for the rest of the game (with the exception of one move), Navara spends less than a minute per move to convert the point! 31…a6 +-
Not exactly a better recommendation for Black in the position, but now the b6 square is weak. White’s goal now is to stretch out Black’s defensive resources with his rook and try to make Black run out of good moves. 32. Kc5 g6 33. Ra8 Kb7 34. Rf8
White’s repeating moves – but that doesn’t mean he doesn’t have ideas. With each move, not only does Navara gain 30 seconds, but repeating moves in a superior position can actually create a psychological advantage! I’ve had a few cases in tournament games where I’ve used this idea, and sometimes instead of repeating my opponent’s have completely collapsed! Of course you can’t failry compare the caliber of opponent I’m playing to the likes of Baadur Jobava…
34…Kc7 35. h3 Kb7 36. g4
Since Black cannot make any productive moves, White decides it makes sense to open the h-file so his rook has more options. In this position, Black has three weaknesses: the 7th rank, the d6 square, and the b6 square. At the precise moment, Navara will relocate his rook to attack Black’s weak queenside pawn structure.
36…hxg4 37. hxg4 Kc7 38. Ra8 Kb7 39. Rh8 Kc7 40. Rh1 Rd2
For a second it seems like Black’s rook has become active, but there’s a cute trick here to force the rook back to e7 (not d7)!
41. Rh7 Rd7 42. g5 1-0
Jobava was so dissatisfied with this endgame he actually resigned to the Czech Grandmaster! In this position, Black is more or less obliged to play 42… Re7 because after 42… Kb7 43. f5!! actually leads to forced mate. The pawn is poisoned since a move like 43…exf5 loses immediately to 44. e6 and White will have managed to trade rooks and gained a queen on the way.
So 42… Re7 is the only move that doesn’t lose immediately. However this move also fails to defend adequately because now when White plays 43. Rh2, Black can’t also activate his rook since it needs one tempo to reach the d-file again, so after 43… Rd7 44. Rb2 Rd3 45. Rb6, White will win Black’s queenside, and the win of the f-pawn doesn’t help Black.
Black’s endgame was actually difficult to hold, and after only one real mistake, Jobava completely collapsed. I thought this game was instructive for a couple of reasons. First it showed us how to press a minuscule advantage, while also using the idea of marching the a-pawn to use the b7 square. This game also showed us that sometimes its possible to hold difficult positions as long as we only have one weakness. I think Jobava may have seen this bunker idea, but thought it would fall apart in the long-run. For a human it may be difficult to hold, but it was really Black’s only real chance of saving the game. Lastly, Navara showed us the importance of gaining tempi at various points of the game. While an extra small improvement may not seem significant in a particular moment of the game, such extra moves add up and become overwhelming.
As I mentioned earlier this week, I’ll be in Washington DC for a tournament this weekend, so I won’t be able to post a video this weekend. Make sure to look out for a post next week on my performance! The Cherry Blossom Classic promises to be a tough tournament, and I’ll be hoping to continue my luck from New York!