When is Brilliance Helpful?

  • NM danheisman
  • | Aug 26, 2012

Every year we hold the charity Holly Heisman Memorial Tournament to raise money for women in need. Thanks to our organizers, the Ramachandras, our TD Joshua Anderson, and Howard Stern, who donates a trip to his studio which we auction off on EBay (coming up this October!), we have raised about $75,000 over the past 10 years. The Holly Heisman Memorial Fund is presided over by The Philadelphia Foundation.

A few years ago the following position occurred in the event. As usual, try to figure out what you would do in a slow game before proceeding:

In this position, Black played 1...Qe2. Take a moment and see how you would rate this move:

   a) Brilliant!
   b) Unnecessary but OK
   c) Unnecessary but not quite OK
   d) Terrible!

In the game White thought for a little while and responded 2.Bd2. Black then proceeded to play 2...Qxd2 and won easily with his extra piece.

Hopefully you found both players' moves somewhat lacking. After 1...Qe2?? 2.Rxe2 Rd1+ 3.Ne1 Bf1 4.Qc2 White is just winning, and even after 1...Qe2 2.Rxe2 Rd1+ 3.Re1 Bxg2+ 4.Kxg2 (but not 4.Qxg2? Rxe1+ and Black wins the queen and regains superiority) 4...Rxe1 White is winning.
After the game I asked White why he did not play 2.Rxe2 and he replied that he would be checkmated. I challenged him to play White and see if he could checkmate himself after 2.Rxe2 Rd1+ and he could not, although a sui-mate is possible in lines like 2.Rxe2 Rd1+ 3.Ne1 R8d8 4.Rc2?? Rxe1#, but that is not germane to the story. But it is germane to say that if your opponent plays a move like 1...Qe2, even if he took a long time to do so, you usually don't just believe him, but ask "Maybe he made a mistake...is there ANY way I can get away with taking the queen?" and analyze very carefully because a sacrificial offer like this is almost always critical.
Let's evaluate the position in the diagram and see what is happening. Black has almost every conceivable advantage:

    a) Black is ahead a pawn
    b) Black has the advantage of the bishop pair (commonly just shortened to "the bishop pair", meaning that he has two bishops and his opponent does not - the average value of the bishop pair is about a half a pawn).
     c) Black's king is safer
     d) Black's army is much more active (more important than his also-stronger pawn structure).
     Together these amount to a winning advantage for Black. When you are winning, in general you want to keep things simple. Complications favor the player who is losing.
     Look at it this way: suppose you are ahead a knight and can create complications where it is certain someone will lose a knight, but it is 50-50 whether it will be you or your opponent. Then winning a knight to put you ahead two knights just goes from winning easily to winning more easily, while if your opponent wins the knight he goes from losing badly to even, so he has a lot more to gain!. Even if it's not 50-50, if you're winning, you still have a lot more to lose then he does. More on this, and other recommendations on how to best go after the full point when you're way ahead, in "When You're Winning, It's a Whole Different Game" at http://www.chesscafe.com/text/heisman13.pdf.

     In the diagram, Black does not have much incentive to look for, much less play, tricky moves like 1...Qe2. Tricky moves and complications are fantastic when you are losing and they give you chances to get back in the game. And grandmasters make a living finding clever moves when it looks like they are about even, but in fact they have some genius idea that shows that they really are winning. So if you are rated 2300+ and want to win with a flashy combintation to show everyone how good you are, we trust you can do it without much risk. But if you are already way ahead, then you have to be very careful that if you play something tricky, the one who will be tricked is not you! If you are not that good, you are taking a big risk by making "brilliant" but unncessary moves that raise the chances that you can miscalculate and throw it all way. When you have more to gain, that's when you want to be more careful, so the more ahead you are, the more careful you should be. Note that I am not saying the opposite, that when you are way ahead you should play very defensively and passively - that's not usually good either. But the "normal" middle ground should be sufficient.

     In this case, Black should play the "normal" 1...Rd1, offering not just a rook trade, but creating big back-rank problems for White. This move is not only much simpler, but it is obviously much stronger. Even if 1...Qe2 worked, it probably was unnecessary. It's great to show everyone how smart you are, but it's also good sometimes to show everyone how wise you are by avoiding unnecessarily "smart" moves. Smile

PS: If you would like to make a tax-deductible contribution to one of my charitable funds, follow the directions at http://danheisman.home.comcast.net/~danheisman/Main_Chess/donations.htm. Thanks!


  • 4 years ago


    if i like to play espectacular i would choose 1...Qd1 with the same idea of 1..Rd1 (Actually 1...Qd1 works quite well but my Houdini 2.0 gives more than 30!! moves that still working) 

    1..Qe2 its the typical move in which one sees a couple of variations that work and then he/she plays it without any doubt, a clear case of hope chess.

  • 4 years ago


    Qd1 fails to Kg1. - as the article says, unnecessary complications.

  • 4 years ago


    I was looking for some thematic brilliant idea, since White has so many light-square weaknesses on the kingside.  After a few minutes I settled on 1... Bf5 allowing me to follow up with e4 since I'm already up material and my position is better so even if a few winning chances slip my position is still ripe with ideas.

    As for 1... Qe2?? I found that 2. Rxe2 Rd1+ 3. Re1 was good enough (it's the same position as 1... Rd1 2. [nullmove] Rxe1+ which I considered and rejected when earlier looking for mate threats -- except in this variation Black is down a queen).

    In an actual game I'd probably see that I have rooks on my back row and want to develop them.  Maybe if I saw the diagram from Black's point of view that idea would have occurred to me.

  • 4 years ago


    Of course-firstly we need "to see" simple logic of our position....In such position -White must to RESIGNE (if his level is about 2000 ELO).

  • 4 years ago


    Terrible!Lost beautiful Queen!!!

  • 4 years ago


    If black wants to be flashy he could have played 1...Qd1, and the threath of 2...B:g2+ and 3...Q:e1 forces white to play 2. Bd2.

  • 4 years ago


    Found Rd1 almost instantly.  Black's position is just lovely!

  • 4 years ago


    Interesting article. Thanks!

  • 4 years ago


    Great stuff Dan, thank you!

  • 4 years ago


    Well I'm glad my intiial reaction (Rd1 is what to play) was right. I was worried for a moment after Qe2 was revealed (before I ran the computations), wondering if I had missed something.

  • 4 years ago


    Wow, what a great article. My thought was Rd1, with the wonderful queen support. When it was revealed that Qe2 was the move, I figured, I missed a brilliant move, again.

    I really enjoyed the discussion on the idea that if you are losing that to complicate matters could be a benefit. I never heard this before, and it makes a lot of sense to me. Like how I sometimes win complicated games when I am clearly losing. And vice versa.

    Thanks for the article, I am really enjoying the instruction.

  • 4 years ago


    I hit upon another too tricky for my own good move when I tried to solve the position. My first thought was 1...Bf5, hoping to trap the queen, but then she can still move away along the fourth rank. So I tried 1...Rd4, when 2.cxd4 Bf5 would trap the queen, and 2.Qc2 Qf3 3.Rg1 Rg4 is completely winning. I almost posted the analysis, but double-checking, I realised that after 2.exd4 Bf5 the queen can just escape to e2 or e3. Driving the point home that sometimes it's best to keep things simple.

  • 4 years ago

    NM danheisman

    Draconis, thanks! The players here may be a little weaker than the average reader. I am not sure yet how chess.com "slow" ratings compare with USCF (generally online ratings are higher, so you have to add to USCF ratings to get equivalent) but, at the time this game was played, Black was roughly 1300 USCF and White was 1200. Black was not too proud of his move: after he made it, he looked at me (the TD onlooker) as if to say "Don't say anything - I realize now that this was a big mistake!" After the game Black was a little embarassed, too, and we chatted about what I discussed in the article.

  • 4 years ago


    I chuckled in disbelief that Black played 1...Qe2 there and nearly LOLed when you said White responded with 2 Bd2. I figured 1...Rd1 was a decent move, and probably the one I would have played as Black.

    Your advice here, as always, is excellent. One dimension you add that few chess trainers do is that you consider the statistical element of chess. Perhaps this comes from your financial background. Because errors are likely when two humans play, there is definitely a statistical element to chess, and you should make decisions taking this into account. This "minimizing variance when unnecessary" lesson applies to more than just chess, of course, but it is definitely helpful in chess games too.

    Thanks for another excellent piece of coaching.

Back to Top

Post your reply: