Greengard's ChessNinja.com

Nakamura 180 in Chess960

| Permalink | 66 comments

After starting off with two losses in the FiNet Chess960 championship in the Mainz Classic, US champ Hikaru Nakamura bounced back all the way into shared first place on the second day. That put him into today's final match of four games against defending champ Aronian. The Armenian lost his first two games today after his own clean 3-0 sweep on day one, but he won when it counted, beating Movsesian to seal his place the final. After exchanging wins in the first phase, the final is anyone's to take. Live games begin at 1830 local, 1230pm EDT.

[Nakamura is up 2-0 against Aronian with white in game three. -- And he just annihilated Aronian in 22 moves to take the Chess960 world title! Three-nil against Aronian, wow. (They play the fourth game anyway, which is nice for the fans. Chapeau to the Mainz folks for that. -- And we got our money's worth as well, a cute endgame bishop sac from Nakamura to force a draw. Final score 3.5-0.5. Movsesian beat Bologan for 3rd 2.5-1.5.)]

If Nakamura wins there's probably some horribly strained element of serendipity to be made about a US champion becoming the world champion in a chess variant spawned by the last US world champion. Maybe not. Speaking of world champions, shuffle chess may well be an indicator of pure tactical talent, but course where we really want to see Nakamura in Mainz is sitting across from Anand with all the pieces on their correct squares. That might happen next year if Nakamura manages to win the incredibly strong Ordix open, the participant list of which is now up. The top ten participants would make a tidy category 19 tournament! Rapid aces like Kasimjanov and Sargissian lurk below that group and there are another 20 players over 2600. Yow.

Other than just enjoying the fireworks and blunders, there are a few interesting things about watching shuffle chess. One is seeing how differently the games evolve from the same setup. This is particularly interesting once the big open starts and you have dozens of GM games going. Some are remarkably similar to each other as the players struggle to impose normal chess logic on the board while other players seem to be playing as much for chaos as anything else.

It's also notable how the players often rely on prophylaxis right from the start. Standard techniques of controlling a knight's outposts, for example, or blocking enemy diagonals, are valid regardless of the starting squares of the pieces. Humans just don't feel comfortable without such mechanisms. This is a secondary reason computers are so devastating against humans at shuffle chess, apart from the primary one that the games are pure tactics from the start. Comps don't waste time with common sense defensive moves.


I think the most meaningful match up between computer and human would be to leave each to their own. Do not give computers opening databases or game databases--let it work purely from blinding, brute tactics. Let the human rely on heuristics, theory, and experience. How different would the outcome be? Would top GMs still struggle to survive? Perhaps this has already been done?!

In which ever way computers can never be compared to humans as they suffer from different disadvantages.

Computers have the so called horizon effect as they dont see beyond N number of moves. Hence, if N+1 is the losing move of the computer, when until the Nth its winning, its completely blind and it will go for it only to realize later. But, unlike humans it has the knowledge of the upcoming N moves perfectly (unlike humans where we have to visualize the pieces in the right squares). So you can actually say computers CHEAT since internally in their representation, they are moving the pieces and evaluating positions.

Humans of course suffer from the visualization disadvantage.

A possible way to compare humans against computers is to give humans a separate analysis board, where he/she can move pieces (just like how the computer does it internally). Here I think humans will beat computers as they dont suffer the horizon effect. And there is no way we can get rid of the horizon effect for computers (unless we have true intelligence as opposed to AI) to have both at the same level.

Aronian - Nakamura game 3. White's king on c1, rook on a1. Aronian played 13.O-O-O, without the king moving because it was already on c1. This looks so strange. Nakamura did the same thing, king on c8, rook on a8. He played 13...O-O-O.

I was watching the game and had to shake my head a few times when I saw it. Never expected it.

Sorry, it was Nakamura - Aronian, not the other way around. I guess I am still stunned at seeing 13.O-O-O without the king moving.

Nakamura FTW!

Congratulations to Nakamura, who has already won the final against Aronian by going 3-0 (best of 4). Including games from the group stage he has 7 straight wins.

Do they still play the last game?

Whatever that means, Nakamura is totally obliterating Aronian. They'll have to sweep Aronian up off the floor. Stop the fight.

Aronian lost 5 of his last 6 games... seems to be intimidated by Naka

Wow, a complete wipeout. Looks like Aronian just ran out of gas against the aggressive Naka. Would love to see Naka play 960 with Anand, that would be a match!

"Do they still play the last game?"

I hope not.

I don't like 960, but it's awesome that Hikaru won this. It's doubly awesome that he demolished no one less than Aronian.

Luke - No one here on this blog is interested in what you have to say. You have become pedantic, boring and irrelevant. Go away.

For those (me 1st) that didn´t believe in the guy , Nakamura is having the year of his life.
Chapeau .

Nice recovery by Aronian to draw the 4th game. Not easy to do after being squished like a grape in the first 3 games. Good for him.

Interesting that the same rankings seem to apply, more or less, to 960, as normal chess. (Yes, Aronian is higher rated than Naka, etc- but generally it seems approximately true? i.e. 2700 beats 2600 etc, no really shocking surprises.) Maybe only I am a little surprised by that.

Wow! Incredible performance by Naka!! San Sebastian win must've felt vey good for him. This one perhaps would rank even higher!

Actually, Aronian gained a very advantageous position in the fourth game but blew it by letting Nakamura escape with a draw. Maybe the motivation just wasn't there for a game that didn't count.

What about this piece in Buenos Aires' Clarin?: www.clarin.com/diario/2009/07/30/um/m-01968636.htm

(July 30 is not fools' day in Argentina, just in case you were wondering.)

Every day is fools´ day in our country.

Guys, relax. Just look at Nakamura's performance in classical chess in 2008/2009 and everything seems to be clear - he is a very mediocre player, unlike other young players: Karjakin, Carlsen, etc. He lost this year several games to IM's and GM's with a rating 2300-2400, not even to mention those with 2500-2600. 2700+ - he lost it all. Just once this year he beat Carlsen. How can you call him an elite player? In Weik there'll be no mercy.

I think mediocre is ELO 1250 if I remember the stats right. Probably just about where you are.

GG -

You undermine your point by saying "very mediocre"

"...he is a very mediocre player..." (GG)

If you want to say that Nakamura is not yet an "elite" classical player, I'll agree with that. But not "very mediocre".

Naka remains unproven against the 2700+ in slow chess. That's where we'll see if his *ad hoc* childhood training can survive the Soviet School grads.

Lots of Westerners have deficiency in that regard: raw talent alone loses versus raw talent + lifelong training.

With one round to go, Kamsky is clear first in the Chess960 Open. If he wins in the end, both he and Nakamura will be among the four players deciding the "world championship" next year.

"USA!USA!USA!"!??! But I wouldn't take Chess960 and this tournament all too seriously.

It won't happen after all: Kamsky lost his last-round game and was caught by Grischuk.

does anyone know how they select a starting chess960 position by rolling dice?

They roll a comp

Thank goodness for Robert J. Fischer, or else we'd still be wondering if "ad hoc" Western training can in fact defeat the "Soviet school".

stringTheory: are you asking what they do at Mainz, or are you asking how to use dice to set up a Chess960 position? If it is the second one, I remember in the Wikipedia article on Chess960, it gives instructions for using dice to pick a position.

Just a thought about 960. It is well established that in normal chess, the starting position gives a very slight advantage to White. What if in certain 960 starting positions, it turns out that White is not justly slightly, but clearly better?

The game would be a bust if that were so.

"In your opinion, what is the main difference between Chess960 and 'normal' chess?

Svidler: Some of the opening positions are much better for White, and some are almost equal - so there is a certain amount of luck involved in the position you will get. And the absence of opening theory is an important factor, of course." http://www.chessbase.com/newsdetail.asp?newsid=1121

"so there is a certain amount of luck involved in the position you will get"

This is a powerful argument against taking 960 very seriously.

What about time in chess960?

Thinking from the 1st move instead of 10-25 moves of known theory leaves the player with less time for middlegame and endgame?

That could help players like Nakamura.

hey guys, it is back to normal chess in Mainz and after two rounds in the evening event, Aronian and Nepomniachtchi are leading with 2/2! What's going on with Vishy?

thanks cs.

i saw a pic which showed dice being used to pick starting position in chess960 for comps

info on using dice to determine starting pos:

"That could help players like Nakamura."

Yes, but ... (from the report on the tournament site about the second day of the preliminaries):
"Nakamura in particular was determined not to repeat the events from the previous day. As a result he invested much more time in the opening, which proved to be a good approach. Aronian was the first to suffer from the “new” Nakamura."

FRC aka Fischer Random Chess aka chess960 -- is really two ideas combined into one alternative implementation of abstract chess.

Break the exclusive monopoly of the traditional starting setup.
Let the fans see some chess that is not buried in 30 initial moves of repetitious opening theory. Let us see which of R.Fine's nine opening principles of chess are merely esoteric tactical details of the only starting setup that has yet been studied.

Randomly determine the setup a few minutes before the game starts.

I would applaud Hans-Walter Schmitt and ChessTigers.de if the Mainz FiNet chess960 tournament would shift to keeping Idea_1 but removing Idea_2.

Let the ches960 tournament announce that the next few years will use only the nontraditional setup of RNBBKNQR. Then let us enjoy the GROWTH of opening knowledge that could ensue.

As long as the Random aspect is retained, there will be no learning.

"so there is a certain amount of luck involved in the position you will get"

"This is a powerful argument against taking 960 very seriously."

I've seen proposals to try to fix this by letting "white" make the first move and then giving "black" the option to switch sides or to play on normally. The idea is that the player who moves first has an incentive to make a move that gives both sides equal chances.

Looks like Anand's run at Mainz is going to end tomorrow...

Aronian, Nepo - 2.5
Anand - 1
Naditsch - 0

>>'ve seen proposals to try to fix this by letting "white" make the first move and then giving "black" the option to switch sides >>
I have yet to see actual proof of this "problem". Why attempt to fix a problem you dont know exists?
I seriously doubt anyone will want to replace normal chess array with some other array as suggested by gene. It is obnoxious to try to force chess players and chess 960 players to play your game -- which no one cares about.

Keep fischer 960 as a separate game and chess as it is.

It would be easier to accept your opinion if you learned Chess before trying to change it.


stringTheory, Anand is not as far back today as Nakamura was after day one in the 960. Don't count him out yet.

"Svidler: Some of the opening positions are much better for White..."

Anybody know which ones are "much" better? I haven't seen one yet.

Chess 960 seems to avoid the draw death that Capa feared. If I read the results right there were only 2 draws in 20 games.

Hmm, after day one Nakamura was at least rather close (tied with two others) to the second qualifying spot, while Anand is now trailing by 1.5 points.
But still we should not count him out yet: all he has to do today is adapt his strategy from the WCh match against Kramnik and win with the black pieces. Back then, losing with white wasn't part of the plan - but arguably, avoiding a loss against Naiditsch is the easiest part of what Vishy has to do now.
On the other hand, what is (or would be) more surprising:
- Anand's reign coming to an end, or
- the very fact that it seemed bound to last "forever" ?

I prefer Seiko.

I prefer that the moron goes into the nearest alleyway, crawls inside a garbage dumpster, and sticks the watches into a dirty orifice.

Cute swindle by Naka v. Cvitan just now. I think Black could've saved himself by 42...Kf7 instead of the self-trapping ...Kh7, as played (and there were many other options beforehand for improvements; but that seemed to be the crucial last blunder).

42.-Kf7 would have allowed at least a draw by perpetual check, so maybe Cvitan's fighting spirit (avoiding a draw at any prize) turned out to be fatal for him.
On the other hand, white can still create mating threats after (42.-Kf7) 43. Rd7+ Kf8 [43.-Kg8 becomes the game continuation] 44. Rc1 Kd8 45.Rc7 followed by Ra7 (still with a perpetual in hand). I think black can save himself (Richard Fireman also wisely[!?] wrote "I think"), but maybe this looked too scary for Cvitan with little time left on the clock? Of course it is better than what he did ... .

BTW, Naiditsch plays also in the open!? That's eight games for him today (five in the afternoon, three in the evening), and ten (6+4) tomorrow. I wonder how this might affect his play against Anand, Aronian and Nepo - certainly tomorrow when he will probably face six GM's in the open.

The line I gave should of course go 44.-Ke8 45.Rg7: Kd8 46.Rc7 etc. ... .

What the heck happened to Nakamura in the Finet tournament?

He was 3/3 then this morning, now I can't see him in the standings, even the games disappeared from the live portal game list.

I guess you mean (and looked earlier at) the ORDIX Open [rapid chess]. Don't worry, he is still there, now with 4.5/5 - in the last round, he tested both his opponent and the clock but the game was drawn after 110 moves [including dozens when both players had less than a minute left].
The Finet tournament is Chess960, which Nakamura played one level higher.

Thanks Thomas, I was confused. I will follow Ordix Open. When there are so many of them going on, it is possible to lose track :)

Presently it's the GRENKELeasing Rapid World championship: Anand trying hard to come back, and Naiditsch at least providing some entertaining games (and possibly playing a decisive role as he beat Nepo).

As I said, Vishy tried hard but the queen ending was drawn (can they keep track of the 50 move rule in rapid chess?). Now Nepo needs a draw to qualify, and Aronian can give a draw to avoid playing Anand in the final!?

I thought Nepo was better in the rook ending against Aronian, but I guess he only wanted a draw.

Can you blame him? But looking at the entire game, he didn't "only want a draw" (from the very start) - however, of course he took the draw when it presented itself.

The good news I think is that Vishy will be back with a Vengence next year.. without the burden of wearing the crown for so long.. he would probably be harder to beat in a straight up match play though

yogibear wrote:
Chess 960 seems to avoid the draw death that Capa feared.

After the August 2005 chess960 tournament in Mainz, the data for that and all prior years accumulated showed that---
---The draw rate in Rapid chess960 is only modestly lower than for Rapid traditional chess1:

Rapid chess1.. draw rate: 23%
Rapid chess960 draw rate: 19%

For the ratio of wins between White/Black the accumulated numbers were:

Rapid chess1.. Wh/Bl win ratio: 1.27
Rapid chess960 Wh/Bl win ratio: 1.13


hcl wrote:
Naka remains unproven against the 2700+ in slow chess.

Probably true.
But in Naka's chess960 wins over Aronian this week, Naka gained each decisive advantage in the middlegame, AFTER the chess960 opening phase was finished.

The middlegame positions in chess960 are often subtly different than those in traditional chess. Yet they are fully valid chess positions.

I doubt Naka is ready to defeat Aronian in a traditional chess match played with long time-controls. So this wonderful Mainz chess960 experiment raises the interesting question of WHY did Naka win and win so convincingly over Aronian?

Nobody can prove any one answer, but...
One idea might be that the middlegame positional PATTERNS were unfamiliar to both players, thus negating the superior pattern recognition skills of Aronian.

A more mundane explanation might be that Naka is even stronger at fast chess than his Elo rating at slow chess would suggest.

Yes, a draw was the correct decision, even though at the end he probably had a win if he wanted to trade on h5. But, since he goes into the finals with a win or a draw, there is no reason to keep playing. He could even lose if he accidentally touched the wrong piece or his phone rang. Or, do they not care about the phones at this event?

Thank you for the info on Gene M – he posted an absurd viewer feedback article on chessbase.com (newsdetail.asp?newsid=4553) For the record, his rating is: 1206!
His flawed reasoning regarding draws and the “white advantage” in chess or chess 1 as he calls it is extremely annoying and reeks of obnoxiousness. Perhaps he should post his ideas to chessvariants.org the mother of all idiotic chess rule change proposals. I suspect though that these "ideas" have already been suggested so his 15 minutes of fame is up.

- And mr. Gene M, both Naka and Aronian are quite familiar with middlegame positional PATTERNS of fischer random or chess - they have played both for some time now. perhaps you should ask them.
And explain exactly what "subtle differences" you notice for middlegame positions in chess960 that naka and aronian are unaware of but you are.

GeneM's Reply to DailyDirt Tidbit:

In my ChessBase.com article =4553 I gave a chain of reasoning to support my arguments. All you are giving back as counter-reasoning is an adjective ("absurd"), and to make an empty unsupported claim my reasoning is "flawed".

The Chess3 rule set was devised to show what it would take to make chess less draw prone. As the article clearly states we do not want to actually play Chess3: -- "As an academic investigation, let us design new chess-like games that have a low draw rate; even though we do not intend to play these games.".

You labeled my reasonings as "extremely annoying". Web forums and those kinds of ChessBase.com articles are meant for an intelligent debate of ideas. At least you are consistent in that your post here in ChessNinja.com has nothing in it that could be construed as "annoying" by your criteria.
Frederic Friedel thought well enough of my article that he chose to publish it on ChessBase.com. Several readers replied that they got something positive out of it.
Apparently you would suppress such articles.

I notice you have not provided alternative explanations for the phenomena I have tried to theorize about constructively. If you have any nonnull thoughts on these aspects of chess, let us hear them.

DD Tidbit wrote: -- "explain exactly what "subtle differences" you notice for middlegame positions in chess960".
Again? I have already published extensively on that exact point, complete with data.

DD Tidbit wrote: -- "Both Naka and Aronian are quite familiar with middlegame positional PATTERNS of Fischer Random Chess".
Your assertion indicates to me that you have not deeply researched FRC middlegames, nor the patterns of transition that are possible from FRC openings to the middlegames.

Besides, Aronian and other masters have repeatedly said that tactics play a bigger role in FRC-chess960 (relative to positional strategies) than they do in traditional chess. Is that because by freak coincidence the traditional setup (S#518) happens to be the one out of all 960 setups where positional strategy can flourish? Obviously not. The real reason is lack of exposure to the detectably different patterns, and the lack of accumulated opening theory from those setups -- a sophistication that is necessary for top level positional strategy.

There simply is no one setup that facilitates all the good ideas that can be leveraged in every other setup. Many interesting ideas are hidden by exclusive devotion to only one setup. Maybe there are other benefits to exclusivity that outweigh all the interesting hidden ideas, or maybe not.

I am not hiding behind anonymity, but your are. The reasonings from an anonymous person can be important. But anonymity reduces the value of the poster's mere opinions.


A. Submit to chessvariants dot org. You can investigate or invent to your heart's content.
Flawed my friend because there is no problem, only in your mind. Funny you accuse me of using the very word you use to describe chess rules.

B. Annoying because you keep repeating the same notion with no hard evidence. Notice though that I actually provided the link here.
But readers must note that the feedback was very selectively pruned to favor more "ideas" .


But the top view there are 2 who were against it:
>> ... International Arbiter), Brussels, Belgium
There were already a lot of articles about the "draw problem" (there is no problem). I have sadly to say that Milener's article is the worse of all. ... >>

BTW - I believe Mig covered sometime ago as well with the solution of banning the draw offer. Your article was random musings jumping to various different rule changes/ chess variants you propose. And yes, you included a summary of ideas proposed by others at the end.

This is your words btw:
Chess-3 rules: my answer to the challenge

I rest my case - are you not inventing a new game? And do you know how much of these are already online?

C. Your so called phenomena is not a problem to anyone. Certainly not for you .. do you draw all your games? do you know what rapid chess is? blitz? Do you even understand what equally matched opponents mean? Introduce new imbalance? Why not flip a coin?

D. Sounds very general to me. And again you make the mistake of addressing my supposed lack of knowledge when I merely point out that you apparently feel you know MORE than Naka and Aronian. For the record though, I have played many games of chess960, so I know EXACTLY what you are talking about.. You have not said anything new to me. Chess960 can be very good for tactical training. But it favors the more "tactical" player which is not necessarily a good thing.

But at the level Naka and Aronian are playing who is more tactical? How can I know? It seems quite strange for a much lower rated player to make a judgment on their perceived weakness or strength.

E. My name is insignificant to this discussion. Yours IS, not because you are important ( I assure you, you are NOT), but because you are trying to make money off a chess variant, but have not demonstrated that you are a qualified expert in it - That of course can now be determined by anyone after reading YOUR views, not mine.

And lastly, I have not heard anyone who agreed with you for the countless times you post on the internet. The one review of your book on Amazon is mostly negative and reveals you add nothing of substance to chess960.
And one more thing - has it occurred to you that I may have been a potential buyer of your book completely disgusted when I realized the ideas the author represents.

Some time ago, I really needed to buy a house for my corporation but I didn't earn enough money and couldn't buy something. Thank God my fellow proposed to try to take the loan at reliable bank. Thus, I did that and used to be happy with my secured loan.

Twitter Updates

    Follow me on Twitter



    About this Entry

    This page contains a single entry by Mig published on July 30, 2009 6:20 AM.

    Biel 09 Final Round was the previous entry in this blog.

    Anand Era Ends In Mainz is the next entry in this blog.

    Find recent content on the main index or look in the archives to find all content.