02 March 2013

Carlsen and Chess960

Magnus Carlsen and chess960? As far as I know, the world no.1 has never shown any interest in Fischer's invention. The purpose of that headline was to introduce a portion of GM Bachar Kouatly's editorial in the March 2013 issue of Europe Echecs:-
The characteristic of a genius is to innovate and to take the opposing view against established dogma. Carlsen remakes chess today the same way as Fischer who wanted to avoid forced passage through the openings ['le passage forcé par les ouvertures']. The invention by the American genius of Fischer Random Chess or 960 was destined to measure the intrinsic strength of a player without reciting the opening by heart. His attempt to transform the game of chess by a renaissance was too brutal for a world that doesn't like change.

Carlsen succeeded the tour de force by playing at chess960 ['échecs 960'(!)] with our old rules. In effect, it is extremely rare to see him take the advantage in the opening, contrary to Kasparov who nearly always led the debate in this phase of the game. To the contrary, Carlsen tries to obtain a playable, even slightly inferior position in the first 15 to 20 moves, which goes against all of today's players who like to play with the initiative from the beginning.

He accepts to kowtow ['courber l'échine', literally: 'bend the spine'] in order to begin the real combat in the middlegame, and frequently to terminate it in endgames exhausting to all his opponents. Through his phenomenal understanding of the game he manages to abandon what we believe to be fundamental in a chess game.

[I left in a couple of GM Kouatly's original phrases because I'm not sure about my translation (in case there is any doubt, I'm not a professional translator) and mentioned 'échecs 960' because I couldn't remember encountering the phrase before.]

There is considerable food for thought in the brief passage from EE. I should point out that, unlike many mainstream chess resources, Europe Echecs is not anti-chess960. I referenced a few of their articles in one of my posts about the 2011 St.Louis event, Kings and Queens on Chess960. The statement that chess960 is 'too brutal for a world that doesn't like change' remains to be seen. It's still too early in any transition to make categorical statements like this.

As for the comparison between Fischer's idea and Carlsen's style, it comes down to the observation that geniuses know when and how to break the rules that the rest of us are struggling to assimilate. Kouatly seems to be saying that Carlsen relies on slightly inferior moves to unbalance his opponents. This would be a good topic to explore on my main blog, where I'm building a collection of Carlsen's games; the most recent post was Carlsen TMER PGN.

I've already addressed the idea of using inferior moves to escape the burden of opening theory; see, for example, Shall We Play Amar's Opening? While it has certain attractions for diehard adherents of traditional chess, it is far less rich in ideas than the universe of alternatives offered by chess960. The proof of this is the notion that it takes 'the first 15 to 20 moves' for Carlsen to find an interesting position. In chess960, this happens already on White's first move.

Although it would be a great boost for chess960 if a player of Carlsen's stature would openly support it, I have no illusions that this will happen soon. Professional players need to 'follow the money' and today there is no money in chess960. The transition from traditional chess will happen not from the top-down, but from the bottom-up. As club players realize that too much effort is required to tackle the enormous body of opening theory, which is a necessary prerequisite to finding new approaches in that theory, they will be attracted to the simplicity, elegance, and challenge of chess without opening books. As this happens, the money will start to trickle in and the professionals will follow.

23 February 2013

Problem with the SP or with the Engines?

Continuing with Proof of Concept with HarryO, in Waving a Yellow Flag I discovered that SP408 and its twin SP749 give lopsided results for White in engine-vs-engine play. Is this because of a problem with the SP or a problem with the engines?

HarryO and I set out to play the SP in the comments to a post on his blog: Non-Random Chess 960 Trial Game 6: SP408. He chose to play 1.d4, the same move played in 64 of the 112 CCRL games, and which had garnered a success rate of 75% for White. The resulting position is shown in the following diagram.


SP408 RBQNBNKR after 1.d4

Now it was my move. I noted in my first comment,

From CCRL I'm seeing 1.d4 with a WLD score of +44-12=8. The two most important variations are 1...c5 with +22-9=6 and 1...Ng6 with +20-2=0. Those are terrible stats for Black.

Later I added,

After 1.d4 c5 2.d5, the most popular move is 2...c4. It's not a bug. The engine is trying to prevent c2-c4, which creates a strong center for White. Note that White has also moved the same Pawn twice, but the second move is very strong because it limits the movement of the Black Knights. I don't think 1...c5 is playable.

On top of that, I don't like 1...Ng6. It commits the Knight to a less than optimal square and leaves White a free hand in the center. I appreciate that it prepares ...O-O and guards the weak e-Pawn, but neither of these objectives is a priority.

I finally decided to play 1...d5, a move which had not been tried in any of the 112 CCRL games. HarryO played the critical move 2.c4, against which I had prepared 2...Nde6. The move depends on the correctness of the tactical sequence 3.cxd5 Nxd4. We played through move 16, and although Black never achieved full equality, he was never in real danger of losing in the opening.

Looking again at the comments we made while making the moves, I am impressed by our general discoveries about chess960. After the game, HarryO expanded the anchor post to highlight some of the unusual variations that might have been played.

Our next game started with the twin of SP408: Non-Random Chess 960 Trial Game 7: SP749. It has the same sequence of pieces, but in reverse order -- only the castling considerations are different. The first two moves for each side mirrored the ideas discovered in SP408. White varied first, playing the equivalent of 3.e3. This put less pressure on Black and when we finally abandoned the trial on move 15, Black had achieved equality and was perhaps even somewhat better.

Getting back to the initial question -- Were the lopsided CCRL results 'because of a problem with the SP or a problem with the engines?' -- I'm convinced that it's a problem with the engines. A discussion of why they go wrong would be a good topic for a future post.

16 February 2013

Proof of Concept with HarryO

HarryO and I are playing our eighth chess960 opening now. We've played so many moves over the last six months that I've lost track of what I've covered on this blog and what remains open. Here's a summary of the games to date.

We started off with a test of Non-random Fischer Random, where the players actively choose the start position (SP) themselves, rather than rely on a random process. We played two games, taking a different color in each game. Links to HarryO's posts, where the moves and running commentary can be found, are marked HO; links to my posts are marked MW.

Although the non-random trials worked well enough, we decided to tackle several specific SPs that appear to push the chess960 concept to its limits. Our methodology involves playing enough moves to determine that Black is not overwhelmed in the opening and emerges with a playable game.

Toward the end of last year, I wrote a post that flagged other problematic SPs: Waving a Yellow Flag. HarryO proposed tackling those and we were off again.

Looks like I still need to cover SP408 and SP749. To date we haven't found any SPs that are hopeless for Black, but that doesn't mean they are easy to play.

09 February 2013

Of Knights and Start Squares

After finishing last week's post on win/draw percentages for the pieces on different start squares -- Pieces, Start Squares, and Scores -- I created more tables for pairs of similar pieces. For example, what is the win/draw percentage for Bishops starting on the a- and b- files versus starting on the a- and d- files? While I was doing this, I discovered another geometry that I had bever noticed before. I'll introduce it with a question.

How many chess960 positions are there where the Knights start on the a- and b- files? Of the six remaining squares, three are available for one Bishop and three are available for the other. Placing the Bishops leaves four squares for the Queen. Putting this together means there are 3 x 3 x 4 possible positions, for a total of 36 positions.

How many positions are there where the Knights start on the a- and c-files? If you answered, 'also 36 of course!', then you're wrong. When the Knights start on a- and c-, they are both starting on the same color square. This leaves four squares for one Bishop and two squares for the other. Since 4 x 2 x 4 = 32, there are fewer chess960 positions than for the Knights' a- and b- configuration. The same is true for all start patterns with the Knights on the same color.

I thought I had already seen all of the start position anomalies, but this was a new one for me. I wonder if there are any others that I've overlooked.

02 February 2013

Pieces, Start Squares, and Scores

It seemed like a good idea at the time. The CCRL data (see the right sidebar under 'Resources' for a link), last captured on this blog for Waving a Yellow Flag, is very straightforward: start position (SP), score for White (%), draws (%). Why not calculate the average % for different pieces starting on different files?

The following table shows the results. For example, Bishop starting on the a-file scores 51.55% for White overall and draws in 20.51% of the games.

Although the numbers look like statistical noise, a few patterns might be significant. According to the table, Bishops starting in the corner win less for White than when starting on other squares. Bishops on b/g-files draw less. Knights in the corner win more. Queens in the corner draw less.

Why only 'a good idea at the time'? I expected to see something more convincing.

26 January 2013

Castling Diversity in Chess960

After posting First Move Diversity in Chess960, I wondered if I could develop any sort of similar statistics for castling -- how many times did players castle to the same side, the opposite side, that sort of thing. What should have been a simple exercise turned out to be more difficult than I expected as I grappled with limitations in the database software I used. Excuses aside, here are the results I calculated on the same 4340 games collected for the 'First Move Diversity' post. First, here are the number of players who castled in each game:-
  • 0: 1516 (no castling)
  • 1: 1607 (only one player castled)
  • 2: 1217 (both players castled)

These numbers are surprising. Of the 8680 opportunities to castle, more than 50% of the time a player did not execute the option. I went back to recheck my calculations and couldn't find any errors on my part, so the reason must be elsewhere. The 'First Move Diversity' post mentioned that there were 45 games abandoned before the first move was made. I discovered that there were nearly 150 more games that had been abandoned after only a handful of moves had been made. Indeed, short games account for many of the games where no castling occurred. How many, I can't say for sure.

Of the 1607 games where only one player castled, here are the counts for castling to one side or the other:-

  • O-O:     883
  • O-O-O: 724

Of the 1217 games where both players castled, here are the relevant counts for castling:-

  • Both O-O:     659
  • Both O-O-O: 313
  • Both sides:    245 (both O-O & O-O-O)

It would be useful to compare these chess960 results with similar counts for traditional chess, but I didn't have time to do this. My gut feeling is that castling O-O-O occurs far more frequently in chess960 than in traditional chess.

It might also be useful to examine the castling patterns for the 56 different start positions of the King and Rooks (see Castling Patterns Visualized for a discussion of the basic patterns). Unfortunately, a sample of 4340 games probably isn't enough to produce valid results. I still might look at it if I find the time.

19 January 2013

First Move Diversity in Chess960

When I start playing chess960 on a new site like The Lechenicher SchachServer (LSS), one of the first things I do is to check for a chess960 game archive. For LSS I found
Database with all LSS Chess960 tournaments finished up to and including December 2011. 4340 games.

After downloading the file, I imported it into a standard database (not a chess database) for further analysis. Taking averages into account, I expect a sample of 4340 chess960 games to have about 4-5 games per start position (SP). For the LSS games, I counted 858 SPs, meaning a little more than 10% of the valid chess960 SPs haven't been played in a single game.

Of the SPs that have been played, I was curious to know how many different first moves have been tried on average, so I produced the following table. It says that 55 SPs have seen only one first move, 228 SPs have seen two different first moves, and two SPs have seen nine different first moves.

1: 55
2: 228
3: 189
4: 198
5: 115
6: 51
7: 17
8: 3
9: 2

Not shown in the table are 45 games that had no first move, because the games were abandoned before any move was made. As you would expect, the 55 SPs with only one first move were mainly SPs which had fewer games on LSS. What about the two SPs with nine different first moves? Here they are:-

NNBRKBQR 9
RQKNBNRB 9

[Apologies for not adding the SP numbers, like 'SP001' or 'SP534'. Although these are calculated for all LSS games and recorded as a comment before the first move, the number is missing for many games. I could have worked around this, but it wasn't important for this current exercise.]

The second SP in the list ('RQKNBNRB') is in a class by itself. When I look at the counts of games having a particular SP, this is what I see for the first five in descending order (the counts are all even because LSS events are composed of two-game mini-matches where the opponents play both White and Black using the same SP):-

RQKNBNRB 42
NRNKRQBB 18
NRKQRNBB 16
RKNRQBBN 14
BRNBKNRQ 14

In other words, 'RQKNBNRB' was used in 42 of the 4340 games on the database. Why it was used more than twice as many times as any other SP is a mystery. The other SP with nine different first moves ('NNBRKBQR') is more normal. It is shown in the following diagram.


SP070 NNBRKBQR

Here are the nine first moves that were played on LSS, along with a count of the number of times they were tried (e.g. three games for 1.d4):-

3 : 1.d4
2 : 1.e4
1 : 1.b3
1 : 1.c4
1 : 1.f3
1 : 1.g3
1 : 1.g4
1 : 1.Nb3
1 : 1.Nc3

Where do I go with this data from here? I can't say for now, but I'm sure something will come to mind.