Rook vs. 2 Minor Pieces

I would like to continue exploration of material imbalances. Last time we dove into queen vs. pieces. This time we will analyze rook vs. two minor pieces. Rook vs. two minor pieces is a more common imbalance than queen vs. pieces, and it seems like a natural follow-up to my last article. Even though this imbalance is more common, the ability to correctly evaluate these kind positions doesn’t come easily, or at least it hasn’t for me.

There’s a reason why we’re taught as beginners not to go Ng5 followed by Nxf7 in the Italian. The two pieces are generally better than a rook and a pawn, even though according to the material scale we are all taught, two pieces and a rook and a pawn are both worth 6 points. However, there are cases when the rook is better than the two pieces, say if the rook side has better placed pieces, has better coordination, has better pawn structure, etc.

However, there’s a common theme that in the endgame, if there is a lone rook on the board, then it can be better than the two pieces. However, if you add an extra pair of rooks on the board, then the side with the two pieces is usually in great shape.

I’ve played plenty of games with this material imbalance, but I’ve decided to take a trip down memory lane to November 2014. The fall of 2014 was a good period for me; I made the push from the upper 2200’s to 2300. Yet, I had two games, played eight days apart, where I totally botched up the rook vs. two minor pieces imbalance. In both games, I had the rook and overvalued my chances.

Episode #1: Ignorance is bliss!

Brodsky, David (2277 USCF) – Barsky, Sam (2130 USCF) Marshall U2300 November 2014


White to play

I just snagged a pawn via a little tactical trick. However, if 26.Bd2, black has very active pieces and has pretty good compensation, say after 26… Re2. Therefore, I decided to play 26.Bxg7 Rxg7 27.Nxg7 Kxg7 28.Rf5 Rd8 29.Raf1


Black to play

OK, I’d like to quote my notes I made back in 2014 just to show how ignorant I was.

“In the endgame, a rook is usually stronger than the two pieces. I thought that I should be on the verge of winning here.”

Er… ehm… [insert guttural noise] *tries not to choke*

First of all, the first sentence usually isn’t true if there is an extra pair of rooks on the board…

When I run my engine, its evaluation is… 0.00!? OK, had I gone 29.Rg5+, I would have been +0.3, but that’s not the point. This is totally not “on the verge of winning”. But why did I think I was near-winning? That is the question. Well, I think I overvalued my active rooks and thought that the black minor pieces weren’t well coordinated. Still, was that sufficient justification? No, it wasn’t.

If 29… Be7, black is completely fine. A key idea is that if 30.Rf7+ Kg8 31.R1f5, black can calmly go 31… Re8!, and white is in trouble (I think missing that was part of my misevaluation). Instead, my opponent played 29… Bf8? 30.Rf7+ Kg8 31.Rxb7


Black to play

Now white has snagged another pawn and actually is on the verge of winning. However, it’s not so clear even here how to win. White has 3 pawns, but it will take a while to push them through. Meanwhile, black can get active, as he did in the game.

I soon snagged a fourth pawn but let my opponent get very active. Eventually, we reached this position.


White to play

Not bad activity! For black that’s it. A prime example of what you want to achieve with 2 pieces and avoid if you have the rook. This was a no-nonsense situation, especially with my king in trouble. I should have gone 38.h4!. The point is that if 38… Ne2, white has a perpetual check with 39.Rb7+, as the black king can’t escape via g5 anymore. Instead, I played the totally irrelevant move 38.a4? and 38… Ne2! was a pretty rude wake up call. I either get mated or lose a rook. I chose the latter and eventually lost.

What’s the moral of the story? First of all, don’t underestimate the pieces! Don’t assume the rook is better in the endgame, especially if there is an extra pair of rooks on the board. Second of all, don’t be too materialistic! I made the terrible mistake of grabbing a pawn but letting my opponent get very active, and it cost me big time.

8 days later…

Episode #2: My $1,000 blunder

Yes, this was for real and yes, I did type the right number of 0’s.

Brodsky, David (2276 USCF) – Shen, Arthur (2434 USCF) National Chess Congress 2014


White to play

Things had gone great for me, and I was just winning here. White is a pawn up and has a beautiful knight on d5; black’s king isn’t safe, and his pieces aren’t doing a great job… basically, white has the pawn AND the compensation.

However, there isn’t an immediate knockout punch. Still, had I done something reasonable like 26.Rf3, thinking of doubling up on the f-file and contemplating going Rc3, I would have been winning (+3 according to the engine). Black is near-paralyzed, and white can improve his position. Instead, I got carried away…

I went 26.Nf4 Re5 27.Ng6+?? hxg6 28.Bxe5 Nxe5 29.Qxb7


Black to play

White has a rook and two pawns for two minor pieces. The queenside pawns are somewhat advanced, and black’s bishop isn’t the greatest. However, the knight on e5 is very powerful, and as the game shows, the pawns aren’t that powerful. In simple English, white is still better, though it is a far cry from the winning position he had in the previous diagram.

The game went 29… Qc8 30.Qxc8 Rxc8 31.b5 axb5 32.axb5 Rb8


White to play

The b-pawn is far advanced. However, it won’t be queening any time soon. What to do? The best way to play is 33.Ra1! Kg8 34.Rfb1. With white’s a-rook well-placed, the position is unpleasant for black.

Instead, I blew what remained of my advantage with my next move 33.b6?. After 33… Kg8 34.Rb1 Nd7, I’m losing the b6-pawn. The position is about equal, though maybe white has slight pressure. I played 35.Rf3, and we agreed to a draw.

I’m not lying in the tile. It was the last round of the tournament, and had I won this game, I would have won $1,200. Instead, I won $200. My stupidity cost me $1,000 in prize money.

During the game a part of me knew that what I was doing was absurd. Yet I didn’t listen to it!

Is the rook really that bad?

So far I’ve portrayed the rook side of the argument quite negatively. Is that appropriate? Yes and no. The two pieces in general are more powerful than the rook and pawn. However, there are cases when the rook is better than the two pieces. For instance, I had a recent game at the Philadelphia Open, where I got this position.

Wang, Alex (2121 USCF) – Brodsky, David (2457 USCF) Philadelphia Open 2017


White to play

It’s fairly clear that black is on top. He has a rook and two pawns for the two pieces, and there isn’t an extra pair of rooks on the board. Also, white has some loose pawns that need protection. Black may not be technically winning, but he has a big edge, and I went on to win the game.

The rook can be good. However, in my experience, the two pieces have usually been better than the rook. Of course it all depends on the situation, but if you don’t have a good reason why the rook is better, then the pieces should be on top.

Leave a Reply

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

You are commenting using your 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