Chess960: Instructive or just fun?

10/29/2016 – Bobby Fischer came up with the idea of Chess960 or Fischer Random Chess, where the pieces are placed in a random order on the last rank. Top grandmasters have dipped their feet in the ocean of Chess960, but the format hasn't really caught on. In this article Ganesh Dorairaj asks a very pertinent question (and seeks your feedback): is the study of Fischer Random games played by top grandmasters useful to improve our chess?

ChessBase 16 - Mega package Edition 2022 ChessBase 16 - Mega package Edition 2022

Your key to fresh ideas, precise analyses and targeted training!
Everyone uses ChessBase, from the World Champion to the amateur next door. It is the program of choice for anyone who loves the game and wants to know more about it. Start your personal success story with ChessBase and enjoy the game even more.


Chess960: Instructive or just fun?

By Ganesh Dorairaj
Hikaru Nakamura vs Levon Aronian, Mainz 2009
Chess960, starting position number 666

[Event "Chess 960 Mainz 2009"] [Site "?"] [Date "2009.??.??"] [Round "?"] [White "Nakamura, Hikaru"] [Black "Aronian, Levon"] [Variant "chess 960"] [SetUp "1"] [FEN "rnkrbbnq/pppppppp/8/8/8/8/PPPPPPPP/RNKRBBNQ w DAda - 0 0"] [PlyCount "23"] {Chess 960-Position 666} 1. e4 e5 2. Nf3 g6 3. Nc3 c6 4. g3 d6 5. d4 Bd7 6. Rd3 Na6 7. a3 Bh6+ 8. Bd2 Bg7 9. Be3 exd4 10. Bxd4 Nc5 11. Rd2 Nf6 12. Ng5 Rf8 { And now both sides castled long. Check out in the gif below how castling is done in Chess960. The king and the rook end up in the same position after castling like in a normal game of chess.} 


That's how castling is done in Chess960. Check the replayer below for the remainder of the game.

[Event "Chess 960 Mainz 2009"] [Site "?"] [Date "2009.??.??"] [Round "?"] [White "Nakamura, Hikaru"] [Black "Aronian, Levon"] [Result "1-0"] [Variant "chess 960"] [SetUp "1"] [FEN "2kr1r1q/pp1b1pbp/2pp1np1/2n3N1/3BP3/P1N3P1/1PPR1P1P/2KR1B1Q w - - 0 13"] [PlyCount "17"] 13. e5 Ng4 14. f4 Ne6 15. Nxe6 Bxe6 16. h3 dxe5 17. Bxa7 Nf6 $6 18. Ba6 $3 e4 19. Qg1 Rd7 20. Bxb7+ Kxb7 21. Qb6+ 1-0

I stumbled upon this game recently. Going over it, I was really surprised as to how a genius like Aronian can go down in less than 25 moves. As a chess trainer for more than 15 years I have repeated the basic principles of chess endlessly, like develop your pieces quickly, control the center, don't open the position when underdeveloped etc.

Looking at this game should give a comforting thought for amateur chess players and woodpushers. I am writing this not to hurt anybody, and definitely not Aronian. Even the greats are not secure in unfamiliar territory. All I want to do is check whether studying chess960 games by these great players can help us to become stronger. Let's go to the fun part.

On move three Aronian played 3...c6, (Fischer would call it a lemon!) violating "don't make unnecessary pawn moves early in the game." Now that's something the Armenian genius would have learnt at the age of five! Developing the knight to f6 or c6 would have been pretty normal. The computer suggest 3...f5! as the strongest move.

The next error in the game was on move six by Nakamura. 6.Rd3 which once again goes against the boring traditional advice "bring in the heavy pieces only after the minor pieces are developed." He did bring his queen out early against Sasikiran (1.e4 e5 2.Qh5). Not a quick learner, eh?!

Aronian's 11...Nf6 was a pretty poor move. Levon, who was already in a difficult position, had to go for 11...f5! However, the move 11...Nf6 showed a complete disregard for this pawn break and the fight for space. Bad chess culture as Garry (Kasparov) would put it!

On move 13, Nakamura followed the basic principles and brought his king to safety with 0-0-0. The gif image above shows you how 0-0-0 is done in Chess960! However, he missed the most obvious and strong move, which should be pretty easy for a player of Nakamura's calibre...

...13.e5! Attacking the knight on f6 and undermining the guy on c5. After 13...Ng4 14.Bxc5 dxc5 15.f4, White has a huge advantage.

Aronian's 18...Nf6 was the last nail in his own coffin. 18...exf4 was a much better try, complicating the position. The main point being that the knight gets the e5 square to defend his king. As we shall see, in the main game, this option was not available. And 19.hxg4 is met with ...Bxc3 with an unclear position.

Nakamura (White) finished the game with a nice move here. What did he play?

It all boils down to king safety. Nakamura uncorked the strong 19...Ba6!! and the game was just over.

With this victory Nakamura won the title of the World Chess960 Rapid champion in 2009

After having looked at the above game and analyzing it, I have an important question to ask: Does analysing chess 960 games played by grandmasters help us to understand the game better? The jury is still out on this and I look forward to your views in the comments section below.

How to save a chess960 game?

by Sagar Shah

When Ganesh mailed me this article, he sent the moves and analysis in Word format. The challenge for me was to put the game in a ChessBase board and then save it. I tried to find a "New Chess960 game" in ChessBase 13, but was unable to locate such a feature. Then I clicked on the setup position option (clicking "S" when the chess board is open gives you the setup position board). I googled the position number 666 and got to know how the pieces are placed on the last rank. Everything was going smoothly with this method, until I reached the move of castling. As the position was set up, the option of castling was impossible. 

When one family member isn't able to solve the problem, you have to go to the other! ChessBase 13 couldn't help me to save my Chess960 game, but Fritz 15 came to my rescue!

On the board screen, click on "home", then "New Game" followed by "Chess 960"

You will be asked to input the position number that you would like to have. I needed the position number 666. If you want a random one, click on "Draw Lots".

And there you have it! The position is ready and you can either play against the computer, or click on "Infinite Analysis", put the moves in, and save the game. Once the game is saved you can open it, in ChessBase 13 as well.

About the Author

Ganesh with Hou Yifan

Ganesh Dorairaj is an ardent lover of the game and extremely passionate about chess. He has a coaching experience of nearly 16 years and is also a trainer at the famous Chess Gurukul Academy in Chennai. He would be interested to train people who share a similar passion towards the game. You can reach him on his email:


Rules for reader comments


Not registered yet? Register

lordnala lordnala 3/19/2018 08:12
I can't get Fritz 15 to load and engine and play Chess 960 positions. Am I doing something wrong or is this a bug/limitation?
genem genem 11/2/2016 07:38
If chess were just being invented this year for the first time, and we were all debating which of the 960 start setups to use, the traditional position would have its advocates and its detractors.
Advocates would say the symmetry of pieces between the a-wing columns a,b,c and the h-wing columns f,g,h is desirable, partly for aesthetic reasons.
Detractors would say the symmetry narrows the range and variety of ideas for opening and middle-game play, for such things has how to fight for control of the center.
Specifically, I dislike the over-abundance of four Bishop-to-Knight-5 pinning moves that are emphasized by the one traditional position. The B-N5 moves crowd out a lot of other ideas that are available from other chess960 start setups. Wouldn't two B-N5 moves be enough, so that other ideas could be allowed to breathe?
ulyssesganesh ulyssesganesh 11/2/2016 05:04
dear just passingby, why don't you come out with your own articles for chessbase.....we would love to read them!
rajanss rajanss 10/31/2016 02:42
Very nice article!! Chess 960 is both fun and instructive. Looking forward to more articles like this.
Just Passing By Just Passing By 10/31/2016 09:49
With all due respect to Mr. Friedel and in general, which has been a fantastic source of chess news for years, the direction that has taken recently is quite worrysome.

I enjoy articles by Mr. Shah, he is a good writer, there is nothing to argue about there. But, looking from the outsiders perspective, you get a feeling that is becoming an "outpost" for all-things-Indian - majority of the tournament reports are mainly focusing on the Indian players, there are consistently promotional articles about various tournaments in India, there are suddenly reports by other writers, i.e. not Mr. Shah, which follow the same pattern of focusing heavily on India, and then suddenly there is this - an article literally about nothing, with a coaching ad at the bottom of it!

Shouldn't Chessbase India be the place for all that, with only the most important articles also shared over here, in the global news site?

Again, do not get me wrong - articles, even if there is an Indian focus, are still very interesting to read. However, there is simply too much of that focus recently, and the fact that articles like this one, which are literally quite useless and written by who-knows-whom, with a coaching ad(!!!) hidden in them, are allowed, is quite sad.
Nostalgiac1972 Nostalgiac1972 10/31/2016 06:55
FRC may not be suitable for learning anything that has to deal with a standard chess opening position, but certainly it can be of great help, right from its opening, to improve by discovering the dynamics and general and particular themes of middle game positions. Actually after nine or ten moves in an FRC chess game, we tend to have a fairly complex middle game position where no many trades have taken place. Therefore, it seems useful for those who would like to compensate their moderate knowledge of openings by relying on extraordinarily keen and creative imagination in the middlegame.
yesenadam yesenadam 10/31/2016 05:28
I keep reading Naka 'dominated' Carlsen in 960. I didn't see that whole match, but I saw the 960 game where Carlsen was winning by a mile and seemed to get confused at the end, throwing away his queen, maybe confused about the time control, but the mistake had nothing to do with 960 apparently. So I'm right to think that apart from that single game, it would have been 1.5-1.5?
Mawin Mawin 10/31/2016 04:06
An alternative to Chess960 is to allow players to relocate (swap) pieces on the first row before play begins:

The various relocation methods allow the players optionally to relocate king and/or queen before the play begins, whilst retaining the castling rights (the rooks remain in their position). The players may abstain from this should they prefer the standard setup. The resultant positions deviate marginally from the standard position and would be experienced as natural by most chess players. The foremost difference is that the player can himself decide the setup of his own pieces, whereas Chess960 is wholly randomized. The positions are "natural" in appearance, with rooks always on the corner squares (only in "Fischer Placement Chess" one rook can be relocated) and the bishops always on different colours on the four central ranks (i.e. never on the b or g file). Hence the bishops and rooks must be developed before they can have any impact, just as in standard chess.

DaveC DaveC 10/31/2016 01:56
"Chess18" is a variant that I had proposed in a letter in "Chess Life". In Chess18, the king and rooks are placed on the standard squares, thus maintaining standard castling, but the placement of the other 5 pieces is randomized with the restriction that bishops cannot be placed on the same color. This yields 18 possible starting positions. For me, the bizarre castling rules ruin Chess960. When I play a blitz Chess960 game, I frequently find myself unsure about whether I'm able to castle or not and what the position is going to look like after I castle especially when the king jumps from one side of the board to the other!
FiddleSticksNZ FiddleSticksNZ 10/30/2016 11:51
I find that FRC has enabled me to play a greater range of openings in standard chess. I rely on my better understanding of general opening principles that you have to employ in FRC. It is surprising that many players in FRC lack this knowledge even though they have high Fide ratings.
hpaul hpaul 10/30/2016 05:15
Time is precious, time is scarce. Amateurs as well as pros have only so much time to spend on the game of chess. If our goal is to win games played in standard tournaments we'll do best to spend that limited time on the kind of positions that we expect to find in our tournament games. On the other hand, I wouldn't mind seeing FR becoming the standard form of chess. One way to avoid the shock of the surprise setup is to give the players access to the board 20 or 30 minutes ahead of the clock being started, to familiarize themselves with the setup.
genem genem 10/30/2016 01:19
A better question is - "Does chess960 have anything interesting to teach us about chess?". To that question the answer is "Yes".

In chapter 4 of his 1941 book titled - "Chess the Easy Way",
Reuben Fine wrote what he believes to be the 9 opening *principles of chess*. A study of chess960 reveals that only some of Fine's 9 are genuine principles, in that they apply well to all 960 start setups. But the others are exposed as mere esoteric tactical truths about one particular start setup. And Fine's lack of exposure beyond one start setup probably kept him from detecting another principle not mentioned among his 9.

From the grandmaster chess960 games played in Mainz a decade ago, we learned that even grandmasters cannot play good quality chess when the start setup is announced only moments before the game begins. Bobby Fischer's goal was to eliminate study of any specific start setup. But I feel there is a better goal.


Over half of the chess960 setups are kinda flawed. From the many fine setups, FIDE could announce one per decade - to be the one chess960 setup to be used for that rated chess960 that whole decade; Discard the 'Random' from Fischer Random Chess. This would give the chess world exposure to entirely novel opening systems full of novel ideas.
And with Fritz & ChessBase & Let's Check, the whole world of amateurs could contribute to the creation from scratch of the opening theory and variations for the one chess960 setup. That would be fascinating to watch.
This is the only remaining large vein of chess knowledge that humanity has not yet studied.
(Meanwhile, traditional chess tournaments continue also as normal.)

The Mainz chess960 games also taught us that chess960 offers no cure for the preposterously high draw rate of 60% among elite grandmasters.

Three chess books have been written about chess960:
* Shall We Play Fischerandom Chess?
* Play Stronger Chess by Examining Chess960
* Chess960 / FRC: chess revolutionary future (title translated from German)

And Mark Weeks made weekly posts to his chess960 blog for years, at:
htt p:/ /chess960frc .blogspot .com/
Queenslander Queenslander 10/30/2016 12:51
The chances of learning much about playing Chess from blitz and bullet games of Fischer Random 'chess' are surely not very high.
fons fons 10/29/2016 09:13
@ Karbuncle: 2½ out of 3 is not a big enough sample size to make any definite conclusions. Especially as Nakamura had 2 whites, which is an even bigger advantage in chess960 than it is in regular. And on top of that they were blitz and bullet games.

To be fair the difference between the two is probably smaller in wild tactical complications, but even then it's hard to say whether that's due to tactical skill or just the increase of randomness. Carlsen is no slouch when it comes to tactics, his positional style notwithstanding.
Cyric Renner Cyric Renner 10/29/2016 06:44
I do not why you have to restrict yourself to examining Grandmaster games. If you enjoy Chess960, play it. I enjoy Chess960 over the classical game. I would not play it simply to improve my play in the classical game, but rather I feel it is superior to the classical game and I play it as a result of this.

It is unfortunate that it is not gained more popularity. To me success in Chess960 requires more natural talent then the classical game. It kills the opening, so memorizing lines will not help you.

Top players that have memorized all their openings do have that advantage and that could be one reason why Chess960 has not replaced the classical game yet. Also, I am not sure Chess960 is always conducive to fast play, bullet and blitz etc. You can easily lose a rook or be dead lost with one careless move. Chess can be that way too, but in Chess960 a bit of time in the opening is sometimes needed to see what is going on.

I agree Docamura stomped Carlsen in the Chess960. I do not know how much Chess960 Carlsen has played, but he looked like a fish out of water. It would be interesting if someone would organize a Chess960 match between Carlsen and Docamura. I think I would have to favor Docamura in the match.
horius horius 10/29/2016 05:29
Chess960 makes normal chess look boring as hell
Kurt Utzinger Kurt Utzinger 10/29/2016 05:07
I can't imagine that the study of Fischer Random games played by top grandmaster is useful to improve our chess. The well known pattern are so different that one may say that Chess960 has - at least in the opening phase - almost nothing to do with (classical) chess, it's like another game.
Mr TambourineMan Mr TambourineMan 10/29/2016 01:32
Just also, want to say that I agree with glider1! The way I see it it would be that higher rated players, and big tournaments (in category class) should play FRC and lower once should try to master FRC518 first!
Mr TambourineMan Mr TambourineMan 10/29/2016 01:26
Is the study of Fischer Random games played by top grandmasters useful to improve our chess? Yes. And FRC is chess, and also an accepted chess variant by FIDE that for once tok the right decision! Im looking forward to the time then we will have a real World Championship in FRC with classical time controll. Its just a question of time!
Karbuncle Karbuncle 10/29/2016 12:05
Nakamura recently dominated Carlsen in 960 blitz on It appeared Carlsen was more 'confused' if anything by this variant of chess, although he returned the favor by dominating the regular blitz chess games in the event.
glider1 glider1 10/29/2016 11:52
Fischer Random gives the player a keen sense of the general principals of pawn structure and piece placement when the start position is constantly changing. However it is only of very minimal benefit in Chess since Chess begins from a static start position, so memorization is always faster and more efficient than discovering general principals over the board. It is a pity that memorization has destroyed the beauty of thinking through the opening moves.