Opposite Colored Bishops

Opposite colored bishop endings are supposed to be boring and drawish in their raw form, but… Well, let’s have a look at my last tournament

In my last article, I mentioned I beat 3 GMs in a row with opposite colored bishops. Two with a pair of rooks on the board, and one with pure opposite colored bishops.

I’ve had my fair share of opposite colored bishop endgames, some with rooks and some without. Some were boring, but some were actually pretty interesting. Before the Philadelphia Open, I hadn’t had any opposite colored bishop endgames in a while, and it was time for compensation…

The opposite colored bishop content in the GM Paragua game wasn’t too interesting. Three connected passed pawns on the queenside are just too much.

The real opposite colored bishop deal began in my next round game against GM Shimanov.

First of all, even the decision of going into an opposite colored bishop endgame is worth analyzing.


I had gotten a powerful centralized position earlier in the game, and now it was coming to fruit. White’s f4-pawn is going to drop soon after black plays 35… Rd4 or 35… Qd4. Once the f4-pawn drops, the e5-pawn will drop too. A variety of endgames (pure opposite colored bishops, opposite colored bishops with rooks, maybe even with queens on) could come out. Which one is best?

OK, I wasn’t going perfectionistic here, because I was 99% sure the pure opposite colored bishop endgame was winning. The idea is that after I win the white e and f pawns, white’s king will have to babysit the h4 and g5 pawns which could easily get picked up by the black bishop if white’s king goes on vacation. Meanwhile, I have a majority on the queenside, and I’ll make a passed pawn which will overwhelm the white defenses.

My silicon friend doesn’t quite support my views, and I’m not surprised. Computers are not to be 100% reliable in opposite colored bishop endgames. But true, there might be a plan for black which is objectively better, but a win is a win. Besides, I thought a lot of other possibilities would likely boil down to the same opposite colored bishop endgame in maybe a slightly better version. Basically, I believed that just going into the opposite colored bishop endgame was good enough and there wasn’t much point in looking for something which might objectively be a tiny bit better.

The game went 35… Rd4 36.e6+!? (I had been expecting 36.Rc4 Bxf4+ 37.Kf2 Qxe2+ 38.Bxe2 Bxe5 39.Rxd4 Bxd4+ which is similar to the game) 36… Ke7 (36… Kxe6 can be met with something like 37.Kg2, where black can’t take the f4-pawn due to the awkwardness of the pin) 37.Rc4 Bxf4+ 38.Kf2 Qxe2+ 39.Bxe2 Rxc4 40.Bxc4 b5 41.Be2 Kxe6


Here we are in the opposite colored bishop endgame. However, it turns out not to be as easy as I thought it should have been. The problem is that it isn’t so easy to make a passed pawn on the queenside. If white gets his bishop on the a2-g8 diagonal, he can go to f7 and pick up my g6-pawn. That could through a wrench in the works, and I knew I had to be careful about that.

The endgame still turns out to be winning. I spent the next few moves dancing around, trying to get an idea what white’s defenses were like, admittedly without making any concrete progress. I missed easier wins on a couple of occasions, but here’s where I struck:


The black bishop on e3 is nicely placed; it restricts the white king from getting to the queenside. I went 52… Kd6 53.Be2 bxa4 54.bxa4 Ke5


White’s only resource here is 55.Bc4 going after the black pawns. That’s the problem I mentioned above. However, black gets through after 55… Ke4 (further restricting the white king) 56.Bf7 c5


If 57.Bxg6, white will have no choice but giving his bishop up for the pawn after 57… c4 58.Bf7 c3 59.Bb3 Kd3. GM Shimanov tried 57.Kg2, which puts up more resistance but ultimately does not save the game. Here’s how it ended.

In the middlegame, opposite colored bishop are good for attacking. The logic behind it is that the attacker attacks on the color of his bishop, and it is difficult for the defender to protect those squares. This holds true even if there are fewer pieces on the board, and in the very next round on the very same day, I got first-hand experience with that with white against GM David Berczes.


White is a pawn up and has pressure against the b5-pawn. The problem, however, is white’s king safety. The white bishop is pinned on d1, meaning that until white unpins, his rook is occupied. If Black goes 30… Rc1 31.Kg2 Rxc3 32.Rxb5, white still retains his extra pawn and his pieces are getting more coordinated. However, I was worried what would happen if black waited with 30… Kf7 (or Kf8). The point is that after 31.Kg2 Ra2+, white doesn’t have anything better than going back with 32.Kh1 (32.Kf3??? Rf2# is not a good idea; 32.Kh3? Bg1 is really asking for trouble; If 32.Kf1 Rf2+ 33.Ke1 Rxh2 34.Rxb5 Bf2+ black will grab a lot of pawns and white is in danger of getting worse). Instead of 31.Kg2, white has random waiting moves like 31.h4, but nothing really looks convincing).

Instead, the game went 30… Be3? 31.Kg2 Ra2+ 32.Kf3!


The difference here is that black doesn’t have mate with Rf2 because his bishop is hanging. White has a solid advantage here. Things further went my way, and we eventually reached this position.


Two pawns up and a nice passer, it should be winning for white, right? Well, it isn’t easy, again due to white’s king being weak. Black is planning to pester the white king with checks (Rf1, Rf2). White can escape by putting his king on h3, but that’s not reliable. Black will try to go Bg1, attacking h2 with nasty effects. I played 42.Bd3 but after 42… Rd2 I decided to repeat with 43.Bc2 Rf2 and then played 44.Rc6 attacking the black bishop, trying to throw a wrench in the works. The game went 44… Be3 45.Bd3 Rd2 46.Bf1


White has stopped the checks, but here liquidation started to occur after 46… g5

A few moves later, we reached this position:


Now, the king safety situation has been completely reversed! Black’s king is now more vulnerable than white’s.

Black can play 55… Rxb6 56.Re4+ Kd3, because the bishop defends the rook in case of discovered check. However, white can just play 57.Kg2, and it isn’t easy for black to play. Black’s king is not going to be safe anytime soon; there are a lot of opportunities for him to blunder something, and white has two extra passed pawns!

Instead, GM Berczes bailed out with 55… Bxf4? 56.Re4+ Kd5 57.Rxf4 Ke5 58. Rf1 Rxb6


This is the position I showed you in my previous article. White is winning, but it isn’t so easy. I managed to get through, but getting into this infamous endgame was a first.

The moral of the story is not all opposite colored bishop endgames are drawn! Don’t be afraid to go for a really promising opposite colored bishop endgame just because they are supposed to be drawish. As the defender, don’t automatically assume that you can easily draw all of them. Also, king safety matters in opposite colored bishops, even if there are only a pair of rooks on the board.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s