31 January 2010

Chess960 in Andorra

After posting the photo of Mr. Chess Classic Mainz, I started to wonder if there were any photos relevant to chess960. I found a few from a tournament in Andorra ('Uploaded on June 27, 2009').

Fischer Random © Flickr user Federació d'Escacs Valls d'Andorra (FEVA) under Creative Commons.

The large display shows the start position (SP879: QRKRNNBB). On the board in the foreground White has just played 1.g3, while the players on the next board are still setting up the pieces.

30 January 2010

Mr. Chess Classic Mainz

While working on the latest biweekly chess photo for my main blog (see Posts with label Photos++), I found a nice shot of the man behind Chess Classic Mainz (CCM).

Hans-Walter Schmitt © Flickr user Georgios Souleidis under Creative Commons.

Follow that Flickr link for a set of CCM 2009 photos, although none are for chess960. There is a photo of the winner of the 2009 CCM Chess960 Open at GM Grischuk, 'thinking on the starting position of chess960', but I can't copy it because it's marked 'All rights reserved'.

24 January 2010

GMs vs. 2100-2200 (II)

In GMs vs. 2100-2200 (I), I noted that the GMs were more consistent in their choice of first moves (see that original post for the background of my 'experiment') and wondered if it was a random result. I repeated the experiment on the games from the top ten boards of the first round of the 2008 Mainz Chess960 Open and produced the following chart.


It again appears that the GMs were more consistent in their choice of first move. Four of the five games with the GMs as White started 1.c4, which was not chosen by any of the lower rated players as White. Are the GMs using some kind of chess logic which escapes the others?

[It should be mentioned that the software I'm using to produce the diagram doesn't handle chess960 castling. That's why some variations end as soon as one side has castled.]

23 January 2010

GMs vs. 2100-2200 (I)

Here's a little experiment combining the idea from How Top Players Treat the Same Chess960 Position with Tartakower and Chess960, where I looked at an early round game in an Swiss-style open event between a strong player and a less accomplished player. For this present post I took the games from the top ten boards of the first round of the 2009 Mainz Chess960 Open (see CCM9: Nakamura, Grischuk, and Rybka; in each round of Mainz, all players start with the same position), where some of the world's top GMs faced opponents rated around 2100-2200. The GMs won all of the games.

The following chart shows the first moves by all players. In the notes I've given the names of the GMs, but named their opponents 'NN'. This is not through lack of respect for the lower rated players, but to keep the chart easier to read.


What does this prove? The GMs were more consistent in their choice of first moves, but I'm not sure if this is a random result. Experiment to be repeated.

17 January 2010

Strange Moves, Strange Game

Behind every move there should be an idea, behind every good move there's a good idea, and behind every great move there's a great idea. Sometimes no matter how much I look at a game between other players, I don't understand the idea behind certain moves and can't judge whether they are bad, good, or maybe even great moves.

In my previous post -- Tartakower and Chess960 -- I used a game between a co-winner and an also-ran of the SchemingMind 2008 Chess960 Dropout Tournament (see that post for explanation and links). In this post I'll use a game between two of the strongest players in the event. In the last round both players were on the brink of elimination, making the game a better-to-win, but can't-lose for both. The game started 1.e4 e5 2.Nc3, and Black played a move that I simply don't understand, 2...Qg5, arriving at the diagrammed position.


The game continued 3.h4 Qh6 4.Nb3 Nc6 5.f3 Be7. Note that Black's last move threatened the lightly protected Pawn on h4. White could have played 6.Bf2 or 6.g3, but tried instead the surprising 6.Qe2. After Black took the loose Pawn with 6...Bxh4+, White played the equally surprising 7.Kd1, forfeiting the right to castle.

Now the game entered a long phase where White attacked on the Queenside (to use the familiar terminology of traditional chess), while Black worked to unscramble the misplaced pieces on the Kingside. The full PGN game score, courtesy SchemingMind.com, is below.

[Event "2008 Chess960 Dropout Tournament"]
[Site "SchemingMind.com"]
[Date "2009.07.24"]
[Round "6"]
[White "Apocalypse"]
[Black "pjl1015"]
[Result "1-0"]
[Variant "fischerandom"]
[SetUp "1"]
[FEN "nnrqkbbr/pppppppp/8/8/8/8/PPPPPPPP/NNRQKBBR w KQkq - 0 1"]
[WhiteCountry "USA"]
[BlackCountry "USA"]
[WhiteElo "2593"]
[BlackElo "2461"]

1.e4 e5 2.Nc3 Qg5 3.h4 Qh6 4.Nb3 Nc6 5.f3 Be7 6.Qe2 Bxh4+ 7.Kd1 Qg5 8.Nc5 Rb8 9.Nd5 Bg3 10.Be3 Qg6 11.Qb5 b6 12.Nxd7 Kxd7 13.Qa4 Kd8 14.Bb5 Na5 15.b4 a6 16.Bxa6 Nc6 17.c4 Bf4 18.Bxf4 exf4 19.Nxf4 Qg5 20.Ne2 Nxb4 21.Qxb4 c5 22.Qa4 Nc7 23.Rc3 Ke7 24.Rd3 Kf8 25.Bb7 Rd8 26.Rd7 Rxd7 27.Qxd7 Qe7 28.Qc8+ Ne8 29.Bc6 f6 30.Nf4 Bf7 31.Nd5 Qe5 32. Ke2 g6 33.Bxe8 1-0

Chess960 games become very strange sometimes.

16 January 2010

Tartakower and Chess960

The SchemingMind 2008 Chess960 Dropout Tournament finished recently and unlike the two previous years (see Pyramids and Dropouts for background) there was no clear winner. Of the eight players who managed to stay under the elimination threshold until the sixth and last round, four reached the threshold without exceeding it and were declared co-winners.

It's always interesting to look at games where good players compete with each other, which I'll do in a future post. It's also interesting to look at the early rounds of an event where the best players are taking full points from their less accomplished opponents.

The following miniature, between one of the co-winners and the event's tournament director, was played in round two. White's first move was 1.Nb3. While there is nothing wrong with this move, which develops the corner Knight to its natural square, it does nothing to stake a claim in the center. The moves 1.d4 and 1.c4, restricting Black's responses, were stronger.

Black continued 1...c5, preparing the development of three pieces and establishing a toehold in the center. White's best move was probably 2.c4, avoiding the subsequent embarrassment of the developed Knight. Instead White played 2.d4, reaching the diagram and handing the initiative to Black.


The game continued with a Knight chase: 2...c4 3.Nc5 d6 4.Nxb7 Qb6 5.Na5 c3. The move 6.Bxc3 was White's last chance to avoid a rout, but he continued 6.b4 and the game ended a few moves later. Here is the full PGN game score, courtesy SchemingMind.com.

[Event "2008 Chess960 Dropout Tournament"]
[Site "SchemingMind.com"]
[Date "2008.04.06"]
[Round "2"]
[White "spoke"]
[Black "saxon"]
[Result "0-1"]
[Variant "fischerandom"]
[SetUp "1"]
[FEN "nbrqbnkr/pppppppp/8/8/8/8/PPPPPPPP/NBRQBNKR w KQkq - 0 1"]
[WhiteCountry "GER"]
[BlackCountry "GER"]
[WhiteElo "1634"]
[BlackElo "2373"]

1.Nb3 c5 2.d4 c4 3.Nc5 d6 4.Nxb7 Qb6 5.Na5 c3 6.b4 Qxb4 7.Nb3 a5 8.Ne3 e6 9.Qd3 Bb5 10.Qe4 Nb6 11.Qb7 Nfd7 0-1

Commenting later on White's 4.Nxb7, Black repeated a quip from Tartakower that was new to me: 'Never capture on b7, even when it's a good move!' Is it strange that it should also apply to chess960?

10 January 2010

Queens Lead the Attack

Continuing with Finding Games by Good Chess960 Players, I looked at some of the games played by Chess.com's highest rated chess960 players. The following position arose after two moves in a game between the current highest rated player (name 'John Doe') and his 'Best Win' (i.e. win over highest rated opponent).


It's a definite candidate for Extravagant Openings in Chess960.
Here's the PGN game score.

[Event "Chess960 Players Group vs American Group - Board 1"]
[Site "Chess.com"]
[Date "2009.09.24"]
[White "desperadoes"]
[Black "SeverusSnape"]
[Result "0-1"]
[WhiteElo "1942"]
[BlackElo "2173"]
[TimeControl "1 in 3 days"]
[Termination "SeverusSnape won by resignation"]
[Variant "Chess960"]
[SetUp "1"]
[FEN "nrbknqrb/pppppppp/8/8/8/8/PPPPPPPP/NRBKNQRB w KQkq - 0 1"]

1.e3 g5 2.Qb5 Qh6 3.Nf3 g4 4.Nd4 Qxh2 5.Rf1 Nd6 6.Qa5 a6 7.g3 b6 8.Qa4 c5 9.Ne2 Bb7 10.Bxb7 Rxb7 11.Qxa6 Nc7 12.Qa4 b5 13.Qa3 b4 14.Qa4 Ne4 15.Ke1 c4 16.d4 e5 17.Bd2 Rg6 18.Bxb4 Rf6 19.dxe5 Rxf2 20.Rxf2 Qxf2+ 21.Kd1 Bxe5 22.Be1 Qxe3 23.Qxc4 Rb6 24.c3 Rd6+ 25.Kc2 Rd2+ 26.Bxd2 Qxd2+ 27.Kb3 d5 28.Qc6 Qxe2 29.a3 Nd2+ 30.Ka2 Nxb1 31.Kxb1 Qf1+ 32.Ka2 Qc4+ 33.Qxc4 dxc4 34.Nc2 Bxg3 35.Ne3 h5 36.Nxc4 h4 37.Ne3 h3 38.Nxg4 f5 39.Nh6 f4 40.Nf7+ Ke7 41.Ne5 h2 42.Nc6+ Kd6 0-1

Compare the PGN against the other examples I listed in Chess960 PGN.

09 January 2010

Finding Games by Good Chess960 Players

Here's a useful tip from Study 960 Games. Although it's on the Chess.com/forum, the technique might apply to other online play sites.
Q: Is there any easy way to find 960 games to study? I thought I might look through some tournament games but there doesn't appear to be a way to search for just 960 games.

A: Go to view players with the highest 960 rating then look at their games.

I didn't know how to find the list of top rated players on Chess.com, but after a few minutes worked out the following: Results for players rated over "2000" and rated under "2800". There might be an easier way, but the query works nicely. The chart indicates that the average chess960 rating is about 1200.

02 January 2010

Google Goes Gaga

While working on today's post -- which is now tomorrow's post, since today's post is the one you're reading -- I ran a routine query looking for keywords on a domain:-

Results 1 - 10 of about 90 from chessforallages.blogspot.com for mainz.

After returning a few correct results, Google started to return rubbish...

...where the results had nothing to do with the query, but showed descriptions from other pages that do match the query. The same thing happened using the 'Search This Blog' widget on the Chess for All Ages blog. Ditto for other search terms. I can't remember a Google search ever performing so poorly.

What was the name of that Microsoft service? Bing! It's an awful name, conjuring up all sorts of associations that have nothing to do with web search, but I think I'll give it a try.


More weirdness: Today's date in YYYYMMDD format -- '20100102' -- is a numerical palindrome. When was the last time that happened?