Thursday, September 18, 2014

How to Lose a Drawn (or Won!) Position

It doesn't matter who you are, whether you are 600 rated junior, a 1500 rated club player, or World Champion Magnus Carlsen. At some point in your chess career, you have had an opponent totally crushed only to blunder in spectacular style and be left sitting with your head in your hands. I suspect it's happened to most of us more than we would care to admit and I know I have had my fair share of disasters where I have blown beautiful games by either overlooking simple tactics or, even worse, by relaxing and assuming a position will win itself.

I would like to share a few games of mine where this has happened and I had the black pieces in all four games. In the first two games, I was the one who ended up losing what should have been won or drawn games. In the final two games, I turned the tables on after my opponents blundered in vastly superior positions.

Game 1 - My First Swiss Game

The first position is one I have reconstructed from memory as I no longer have the game score. This was my first ever game in a weekend Swiss tournament. It was played in Portsmouth, England in about 2001. I had completely outplayed my opponent and had been sat in my chair for some time wondering when he was going to resign. I was up a rook and three pawns and just couldn't understand why he was playing on. I had just pushed my c pawn and my opponent moved his queen to e7, threatening checkmate by Qh7#  and Qf7#. I had foreseen this move and thought it was no big deal because I would just defend on the 7th rank. Surely then my opponent would call it quits!

NN - Hayes, Portsmouth 2001

Thinking I had everything under control, I flicked out 1. ... Rb7, which defended against both mating threats from white's queen. There is only one problem with this move, which I am sure you'll see. White was actually threatening a third checkmate and he slowly reached his hand towards the bishop on g6 and played 2. Bh7#.

Ouch! To make matters worse, my opponent even said, "I'm sorry" as he planted the bishop on h7. I sat there dumbfounded for a couple of minutes, wondering where it had all gone wrong. The truth was, it went wrong when I started to relax and didn't take white's threats seriously. I assumed I had a winning position, which was true, but I also assumed the game would win itself and that white would throw in the towel without making any effort. Big mistake!

Game 2 - Singing the Wrong Song

The position below was taken from a game I played earlier this year against NM Melandro Singson. I had comfortably equalized as black against his London System and we reached this position after white's 30th move:

Singson - Hayes, Arcadia 2014

Black should be completely fine here. In fact, Fritz thinks that black is just fractionally better (by less than 0.1 pawns). After the normal looking 30. ... Qd6, white would have to be a little concerned about his king's safety. It's nothing special but should be easier to play for black. Instead, I had a meltdown in time trouble and played the horrible 30. ... g5??  which loses on the spot to 31. Qxe7+. I had hallucinated that this move wasn't possible because of 31. ... Rxe7, winning white's queen, but of course white just plays 32. Rxe7+, which not only regains the queen but also picks up the a pawn to boot. Seeing that I would be down two pawns for zero compensation, I resigned right away and wished the game hadn't been on top board where so many people had crowded round!

Game 3 - Acon Artist

The third game is one where I should have been totally busted but swindled a win. In the Pacific Southwest Open in July 2007, I had had a miserable tournament. Going into the last round, I had only won one game and that was against a 1600 player. I was paired against Jennifer Acon, rated 1856 but she had been around 2000 before. She had obviously had a similarly miserable tournament and was eager to finish on a high note. For a long while, it looked as though that would be the case as she outplayed me in the middlegame and I was forced to give up the exchange to stave off her mating threats.

After 29 moves we reached the position below. In the long term this will be winning for white, not only because of the exchange but also because her queen is beautifully centralized, her rook on f1 controls the only open file on the board, and her king is much safer than mine. I had kept my knight on e5 to play for the only trick I could potentially see in the position and, fortunately for me, Jennifer fell right into it. Most moves should be good for white here. The computer assesses that both 30. Qd5  and 30. Kh1  give white a two pawn advantage. My opponent got a little greedy and decided to pile up on my d6 pawn, presumably thinking of playing for b4 and then c5 afterwards. She played the natural looking 30. Rd1. Try to work out why this is a terrible blunder.

Acon - Hayes, Burbank 2007

The answer is the surprising looking 30. ... Qxh3!!, leading to a highly favorable (probably winning) rook endgame for black after 31. gxf3 Nf3+, regaining the queen and leaving white with a shattered pawn structure. I eventually converted the ending without too many problems.

Game 4 - A Tasty Skewer

David Argall has been a regular opponent of mine for a number of years. He won most of our early encounters, back when he was a strong expert and long before I was rated over 2000. Back then, I was content to scrap for the odd draw among the numerous losses David inflicted on me. In the last few years, however, the tables have turned somewhat and, out of the last five games we have played, I have won four and the other was a draw.

In this game, I had suffered a catastrophe in the middlegame, had shed a pawn, and then bizarrely went into an endgame that was completely lost for me. I had just played 46. ... Ke6, which is a total bluff. At a glance, it looks like I am threatening to play Ra5+, picking up the rook, but this is an illusion. White should just play 47. c7  and it's all over. Sure, I can win the rook but what am I going to do when he queens his c pawn? The position is resignable for black but, then again, it had been for some time and I was playing for a swindle.

Argall - Hayes, Arcadia 2010

David obviously believed my bluff because he didn't play 47. c7. He disregarded that move, probably saw that his a pawn was inconveniently hanging, and figured that trading his a pawn for my g pawn was a fair exchange. I could hardly believe my eyes when he flicked out 47. Rxg5??, which lost on the spot to 47. ... Ra5+, skewing the king and rook. There is no hope that white can queen his c pawn afterwards either; it's simply not advanced enough and black will have no problem stopping it with his king and rook.

I hope you enjoyed these blunder-fests. It goes to show that we all make mistakes but, more importantly, that it's always worth playing on a while even in a bad position. I know some people will say it's disrespectful not to resign but let's take the first game for example. Yes, my opponent in Portsmouth was completely busted for a long time but who am I to tell him he should have resigned a game that he ended up winning? I don't believe in playing on until checkmate but I think one should at least make the opponent show that he knows how to win. Plus, you never know, he might overlook something and suddenly you have turned a lost game into a draw or win.

No comments:

Post a Comment