Three annotated games and a study

Update: Zenchess, a talented new addition to the chess blogosphere, has generously provided a detailed analysis of the first game here. I will have to wait until I'm done with work to look at his analysis closely, but on a quick reading it looks very helpful. He is an 1800+ player USCF (2000+ standard at ICC) and likes to analyze games, so he said to tell the Knights if anyone wants his help, throw a game his way. Thanks, Zenchess!

Two losses and a draw. Click on the description to be taken to the corresponding annotated game.

Game 1 (white/loss): Loss with Grand Prix Attack

Game 2 (black/draw): How to not win a won game

Game 3 (black/loss): A noble but unsuccessful parry

The following is the key position from Game 3, in which I had a strong opening and went into a comfortable middle game. My turn, as black, to move:


See Game 3 above to see what happened next. This type of position is quite common, so I should learn it well. The game was mine to lose, which I of course did quite flagrantly.

Howard Stern on the Kramnik Blunder

Howard Stern, Artie Lange, and Robin Quivers
Howard Stern discussed The Kramnik Blunder for almost 10 minutes this morning. It is available for your listening pleasure here (note he talks about something else for about 20 seconds at the beginning).

Stern has been talking about chess a lot lately. It is usually pretty funny and very entertaining: he has been playing for a few months and has obviously caught the chess bug. He plays at ICC daily, though doesn't reveal his handle (he probably doesn't want to receive 800 messages a day from his insane fans).

In the clip above, Howard talks about Dan Heisman, Bobby Fischer, in addition to the Kramnik Blunder. The woman talking is Robin, his long time sidekick. The guy who says he'd rather watch paint dry than continue discussing chess is Artie Lange (he used to be on MadTV, but is on Stern's show every day now).

If you think Stern is all about naked women, the clip above is more representative of what his show is usually like (though he still has naked women). Even my wife likes his show, and she hates the naked women stuff. Even Susan Polgar frequently mentions his show favorably. She often mentions it on her blog when he discusses chess on his show.

For those interested in listening to Stern, you'll have to subscribe to Sirius Satellite radio. I love it.

The Kramnik Blunder

It happened about ten minutes ago, in game two of the man versus machine match. Larry Christiansen called it the 500,000 Euro blunder. I call it the Bill Buckner move. It is move 34, and Kramnik has thirty minutes left on his clock. He is black, and it is his move:

In this drawn position, after a game in which Kramnik has played the QGA wonderfully as black, he moves 34...Qe3. Obviously, Fritz mates with Qh7#! Kramnik blundered the way we patzers think that only we can blunder. The commentators (audio and text) at ICC went crazy, everyone at first thought it was a joke. No joke. The world champion of chess, with thirty minutes left on his clock, missed a mate in one.

This is going to be a wonderful example for all future chess novices, probably becoming the best known single move of chess in the history of the game. It teaches the important lesson to all of us: even the World Champion can play ridiculously stupid moves.

Strategy Synergy via Pandolfini

I recently finished re-reading Pandolfini's Weapons of Chess, his book on strategy. While I have complained about its lack of systematicity (he uses a dictionary format rather than an organized system of presentation), it really is pretty good, and the little sections can be read in 5-10 minutes which makes it good bedtime reading.

Plus, the book meshes quite well with his excellent Russian Chess, which consists of six fully annotated games in which he focuses on strategic ideas behind each move. Thanks to Takchess for suggesting Russian Chess: it is quite a gem that I had never heard of. The two books make a nice synergy, and are aimed at anyone who knows some basic tactics.

Circles Update: Finished minicircles 2.4 and 2.5 (i.e., problem set two I've gone through five times now). I am starting to really do these problems fast now, so in addition to cranking through Problem Set 2, I have started doing five problems a day in Problem Set 3 (i.e., minicircle 3.1).

	# Circles	Percent Correct
Problem Set 1	14	98-99-100-100-100-100-100 100-100-100-99-100-99-98
Problem Set 2	5	90-93-96-99-99
Problem Set 3	0
Problem Set 4	0
Problem Set 5	0

NOTE: Circles done with CTB.

Kramnik vs Deep Fritz: my prediction

The first game between Kramnik and Deep Fritz begins tomorrow 9AM NY time. The official site is here. Whoever gets above 3 points first wins.

My prediction: the next two weeks will show, once and for all, that all humans, not just chess mortals like us, have been left in the dust. I predict 3.5 points for Fritz, 1.5 points for for Kramnik in the match. Kramnik will lose two and tie three, giving Fritz the win after five games. It won't need to go to the six alloted games, and humans will never give computers a real fight again.

(Note the above was modified from my original prediction as I didn't realize the deal was that whoever gets above 3.0 first wins).

Chess Visualization: Web and Software Tools

Patrick inspired me to think more about board visualization and variation calculation, and to look around for useful tools to build up these ephemeral skills.

First, check out this great site devoted to board visualization. It has lots of cool facts on it that have helped me simplify board visualization and also includes four visualization drills (at the bottom of the page). It starts out very basic but ends up with a bunch of stuff I had never considered before.

This sparked me to look for other visualization software. So far I've found a few programs. While they are generally built for people trying to learn to play blindfolded, when you are calculating variations in your head during a real game you are essentially playing blindfold chess where you are blind to the future board position. If you've used any of them, please let us know what you thought. I'll expand this list as we discover new programs.

Chess Eye: Chess Visualization Training is cool: there is a free trial version. There are many different types of exercices. In one they tell you where three pieces are (K vs KR) and you have to say whether the king is mated. You can try it for free online here. An earlier shareware (i.e., free) version is also available at Sourceforge.

The Push Up System. Quizzes you on your ability to reconstruct board positions, either of a previously seen board, or a previously seen board after certain indicated moves are made. Seems geared to playing blindfold.

Blindfold, a program for visualizing interactions between pieces. I couldn't get the demo version to work. For all I know I just installed a virus on my computer. If not, the program seems interesting and I'll post more once I get the damned thing working.

Please let me know what I've missed (I think one of the knights has some software they wrote but I can't remember whom). I especially appreciate links to software sites or good discussions for improving at visualization/calculation.

Needless to say, this is one of my weaker areas and I'd like to work on it. I know many people say training doesn't help, but given that the web site above has already helped me with visualization, they must be wrong (whether it makes a difference for ratings and actual play is a different question).

This has lasted longer than I expected

Since I went above 1200, I haven't dropped below. I went up to 1266 but have suffered a couple of losses so am at 1221 now (see previous post for one of the losses).

Psychologically, it has been a big lift to start thinking of myself as a 1200+ player. There are a few things that make it gratifying.

For one, in the ratings systems, Class E players are rated 1200 and below, so you feel like a loser being down in the basement of the class divisions. It is nice to feel like I am finally starting to move up the ranks. I also realize that it is not crazy for me to expect to reach 1500 or so within a few years. Second, good players are more willing to play me now: I've had quite a few games against players rated 1300+. The better my opponents, the faster I should improve (and the more I concentrate on playing well).

Perhaps most important, my goal when I started this was to reach 1200. I began this blog three months after learning all the rules of chess: I knew how the pieces moved since I was a kid, but not the en passant rule, I wasn't sure if pawn promotion was an 'official' move, and I didn't know how to castle (I didn't know if castling was an official move), or how the board was supposed to be oriented. I had read nothing about the game, and had probably played about 50 casual games in my life. My first few games (January 2005) I lost to scholar's mate a few times, and constantly got nailed with back rank mates. I then realized that I would need to build up some pattern-recognition skills, which is what attracted me to de la Maza's circles. In sum, it is nice to finally reach my goal (frankly, I thought I would get to 1200 within two or three months: boy was I naive!), especially given that I am not quite halfway done with the circles (note I didn't start the circles until a couple of months ago: see my training schedule).

Based on a suggestion over at Chess Training, I have used a useful formula at ICC to automate my game seeks:
set formula rating>myrating-100 & rating<=myrating+300 & established & noescape.
This sets it so that I will play established players rated between 100 points fewer and 300 points more than me (and 'noescape' constrains it so they have to have noescape on so they can't be jerks and disconnect after dropping a piece). For my seek options in the seek game window, I leave the minimum and maximum opponent rating blank: I don't have to reset it, ever, as the formula takes care of things for me.

Why do I bother?

I played a beautiful opening and was doing great, and the following position emerged on (my) move 12 as white:


What did I do? I captured his pawn on d5. It wasn't that I didn't think about my move. I was so preoccupied with the "key" d5 square that I didn't look at anything else. I suck at chess. Boy oh boy do I suck at chess. I ended up losing the game.

This is the type of problem I could probably solve in seconds if I were in the middle of a set of tactical exercises. This suggests there is a problem with my thought process. Note to self: before getting caught up in details like key squares, make sure I am not missing threats that a 400-rated 3 year-old would see in three seconds.

Man, how depressing.

King's gambit accepted: razor sharp

As black, I've decided to play the king's gambit accepted. I was playing KGD, but in a desultory fashion, never really studying beyond my typical response of 2...Bc5. There are two reasons for the switch: the KGA leads to much more tactical games with good fighting chances for black. Second, Emms' excellent book Play the Open Games as Black advocates for the KGA and devotes three nice chapters to it. My other black book (Davies' Play 1 e4 e5!) has about two-thirds of a single chapter with very little useful analysis (and it advocates the KGD). That book is mostly about the Sicilian and Ruy Lopez for black. The rest is treated in a superficial manner.

Transformation said he loves when black plays the QGA against him. I probably know why: I play it as black and sometimes end up in a horrible bind on the queenside, even though I know enough not to fight to hold onto the pawn. While I think QGA is good for black, you really have to know what you're doing. I expect it to take a year or so before I am comfortable with QGA. But for the reasons I discussed previously, I am now committed to it and will learn the ins and outs. In favor of QGA, white is usually less booked up there than for the QGD.

Circle 2.3 Finished

I finished mini-circle three of the second stage in Chess Tactics for Beginners. I am getting faster while my percentage of correct answers slowly creeps up. A couple more times and I think I'll be nailing the problems with little effort. I'm now doing 40 problems per session and recognize them almost immediately. Despite recognizing the problem, I don't always remember what to do; however, once I reach the board-recognition phase that means in the next circle I will likely recall the right move fairly fast.

Each day, when done with my problems, I've started to redo the ones I got wrong. This seems to help a lot.

I am now very happy I chose the Queen's Gambit Accepted. While I am not good at playing it yet, I am starting to like the feel, starting to learn some basic plans and features of the position. White players tend to not be all that booked up on on it either, so the game quickly becomes chess rather than opening book regurgitation.

	# Circles	Percent Correct
Problem Set 1	13	98-99-100-100-100-100-100-100-100-100-99-100-99
Problem Set 2	3	90-93-96
Problem Set 3	0
Problem Set 4	0
Problem Set 5	0

NOTE: Circles done with CTB.

Compare and contrast two books

I got the Aagaard book today 'Meeting 1d4', which I thought was going to be a good general repertoire book for black. Such books on how to respond to 1.d4 are very hard to come by for the patzer.

Unfortunately, Aagaard's book is simply awful for beginners. Over a hundred pages on what to do after move five in the Tarrasch. At my level, I don't even want to go much past move five in my repertoire (in focusing on tactics, I build a broad and superficial repertoire that I refine over time as I see variations in actual games).

This book should be called 'Metting 1. d4 d5 2. c4 e6 3. Nc3 c5 4. cxd exd 5. Nf3 Nc6', not 'Meeting 1 d4'. There is woefully inadequate coverage of what to do if white doesn't play 2. c4. Plus, it is a worst-case kind of opening book: annotated game dump that doesn't even include an index of variations. This leaves you the fun of searching through games for variations you want to find. Screw that.

Contrast this with James Rizzitano's simply amazing book How to beat 1d4 (reviewed very favorably at Chesscafe here). Over half the book is about what to do when white doesn't play 2. c4. The coverage for when white does play c4 is so packed with information that it feels like Rizzitano could have been greedy and gotten two books out of this.

While I have superficially played the QGA and gotten my rump roasted, I think that because Rizzitano's book is so far superior to any beginner 1. d4 response book I've seen, I'm gonna learn that one. I just don't have the time to pull together different responses from different sources. This seems to be a reliable and sound reference.

OK, hopefully I've made up my mind.

More Tarrasch talking

Due to some advice and excellent discussion on what beginners should play in response to d4, I've decided to stick with the classical stuff. If you are wondering about your response to d4, and are a beginner or a chess teacher, I highly recommend that discussion (it is at Chesspublishing.com).

I think I'm going to the Tarrasch. This, of course, leaves me having to figure out what to play if white doesn't play 2. c4. More will be forthcoming, I'm sure.

As a side-note, the discussions of opening theory at Chesspublishing are excellent. Many GMs are very generous with their knowledge of the opening, and often the authors of the opening books will come on and answer questions. A great resource.

Hat tip to Takchess for the Tarrasch pun.

Is this argument sound? (Flashback post)

From page 23 of Rapid Chess Improvement by Michael de la Maza:

One of the most frequently heard objections to focusing on tactics exclusively is: 'In order for a tactical opportunity to arise, positional/middlegame/strategic (PMS) play is required.' This question is asked in many ways, but the core concern of the question is always the same: tactical opportunities do not appear out of thin air, players must work hard to create them.

This may be true at the Master level, but is certainly not true at the class player level. A simple two-step argument suffices to prove this point:

1) Class players know next to nothing about PMS play.
2) Tactical opportunities (and the failure to recognise them)
decide virtually every game between class players.

Clearly if PMS abilities were required to create tactical opportunities, both of these statements would not be true.

Great new chess training blog

Creatively called Chess-training, it is a new chess blog with lots of good stuff for improving club players. It has very well-written and helpful posts. One of the few blogs in recent memory that I liked so much I felt compelled to go back and read through all the old posts. I'm sure someone else has probably mentioned it when it first started (when it included only comments on the chess World Championship Match), but it has much more than match commentary.

BTW, does anyone know who the author of the blog is? His name is Mark. Based on the writing, I'd guess he is 2000+. (Note added: it appears he is markgravitygood at ICC, with a present rating of around 1800...thanks to Fierabras for finding that. It just goes to show that you don't have to be a GM to write good chess.)

Dancing with the patzers

After consulting Sam Collins' excellent book Understanding the Chess Openings, I have tentatively decided to give the Two Knights Tango a go as my black response to 1. d4 (it normally goes 1.d4 Nf6 2.c4 Nc6!?).

It has the ingredients that have made me like my other openings. It is fairly uncommon, it is sound (i.e., GMs play it), and the move sequence is fairly natural after the first two. No crazy sharp lines that I need to memorize 30 moves deep. It is not so unusual an opening that white can't transpose into more comfortable lines, but as long as I am comfortable with those lines, that's OK.

I've ordered Palliser's popular book Tango! A Complete Defense to 1 d4 and once I've booked up on it I'll post a game or two with an evaluation.

Circle 2.2 Finished

Only went up three percentage points in my success, but in my third mini-circle I think the answers are starting to pop out a little more. I predict that by mini-circle five I'll be at 100% correct.

Also, I went through the first problem set again, but this time reversing colors. If I really know the patterns, it shouldn't matter whether the board is rotated, reflected, and projected on the moon. I want to answer the problems fast from all perspectives. I'm going to incorporate this into my mini-circles. Once I am so good at a problem set that I get cocky, I'll switch up the board to bring myself down to Earth.

	# Circles	Percent Correct
Problem Set 1	13	98-99-100-100-100-100-100-100-100-100-99-100-99
Problem Set 2	2	90-93
Problem Set 3	0
Problem Set 4	0
Problem Set 5	0

NOTE: Circles done with CTB.