Home > Publishing Schedule > A few ideas about fortresses

A few ideas about fortresses

In Endgame Play I think I had a few interesting insights, besides a lot of training material, in an already well-explored part of the game. I think the most interesting chapter in the book was probably the chapter on Fortresses. I have after I wrote the book been made aware of some interesting articles written on fortresses, but in general it is a part of chess that has not been explored fully. For example, Dvoretsky’s Endgame Manual is not very strong in this area, despite being sensational (and essential) in many others.

In Endgame Play I came with a few thoughts about fortresses that I personally find interesting to explore further. I am not sure I will do it personally, nor that they are accurate. But I found them useful in understanding the positions I had collected for the book.

The first thought is to see fortresses not as something that holds or not, but as a defensive technique. For the practical player this makes most sense anyway.

The second is to see how fortresses usually fail. Of course there can be a break that makes the fortress collapse, but in general what I found was that zugzwang was a big part of the picture. Surprisingly a lot of fortresses fail to exactly this position.

White has just played Kd5 and Black loses due to zugzwang. Previously White has probably exchanged a piece on f7. Obviously you could add h-pawns without ruining anything. But once the pawn is back on g3, things are a bit more difficult as we shall see.

The third observation is one I will just leave hanging in the air, but which you will see the value of if you look at the book. It is the only of these that can be said to have a real novelty appeal to it. The idea is that most bishop endings are defended through the fortress technique. As said, I will just hang it out there for you to ponder. My observations say that it is so.

I found the following recent game quite interesting. White was pressing all the way, but Black defended well. At one point he decided to go for a fortress.

Anish Giri – Mart Kraemer

Bundesliga 05.04.2014

92…Nc3+ 93.Kd3 Rxe5!

Black gives up a piece for two pawns. I am not sure if the position is ok without doing this; it might be, but it seems to me that this is a good practical chance.

94.Kxc3 Re3+ 95.Kd4 Rxf3 96.Ne4 Rf1 97.Ke3 Rf5 98.Rd7 Kf8 99.Rd8+ Kg7 100.Nd6 Re5+

100…Ra5 101.Rd7 g5 102.hxg5 Rxg5 103.Rxf7+ Kg6 104.Rf3 and White wins.

101.Kf3 Re7 102.Kf4

102…Re1!

A necessary move it seems.

102…Kh7 103.Kg5 leads to a winning ending for White. The exchange of pawns breaks for fortress: 103…Re3 104.Nxf7 Rxg3+ 105.Kf4 Ra3 (105…Rg4+ 106.Ke5 Rxh4 107.Ng5+ Kh6 108.Ne6 Kh7 109.Kf6+-) 106.Ng5+ Kh6 107.Ne6 Ra4+ 108.Ke5 Ra5+ 109.Kd6 Kh7 110.Rd7+ Kg8 111.Ke7 Rf5 112.Ng5 Rf1 113.Ke6 Rf2 114.Nh7

103.Rd7 Rf1+ 104.Ke4 Kf8 105.Ke3 Rf6 106.Ke2 Re6+ 107.Kf2 Rf6+

107…Kg8 looked simpler.

108.Kg1 Rf3 109.Kg2 Rf6 110.Ne4 Rf5 111.Rc7

Black seems to be holding, but at this very moment he lost his sense of danger.

111…Kg8??

111…Rd5 112.Ng5 Rf5

112.Nd6

White also won after 112.Rc5, trapping the black rook.

112…Rf6 113.Rc8+ Kh7 114.Rd8!

1–0

Black is in zugzwang. Quite a nice one, I should say, and difficult to predict with Black if you do not know what to look for. However, if you do, then it might have been possible. We do not know the time situation, but we can presume that Black was able to think for 2-3 minutes every few moves, after making a few quick moves. Not an impossible task, to predict this trap. Still, very impressive play by Giri.

Categories: Publishing Schedule Tags:
  1. J.
    September 8th, 2014 at 14:49 | #1

    @Jacob,

    For the attacking player, I would argue that it’s important to see fortresses as holding or breaking.
    So that he can anticipate/assess going into it or not. Ones the fortress has been established it’s indeed practical for both sides to see it as a defensive technique.

    I also wanted to add that with bishops, the possibility of zwugzwang becomes much smaller.

  2. Jacob Aagaard
    September 8th, 2014 at 22:25 | #2

    @J.
    I did not say that it is not important to say if they hold or not. But if they do not hold, do you still call them a fortress? If you view it as a technique, you do. So, it is a theoretical/practical stand point.

    Regarding bishops and zugzwang. Sorry, you got this one wrong :-).

  3. Ray
    September 10th, 2014 at 06:58 | #3

    Interesting game fragment – I like the theme of fortresses!

  4. Roy-in
    September 10th, 2014 at 10:48 | #4

    Nice post, thanks jacob

  1. No trackbacks yet.

 Limit your comments to