Greengard's ChessNinja.com

Petroff Loses! Petroff Loses!

| Permalink | 78 comments

And Kramnik's Petroff at that. In Dortmund. And to Arkady Naiditsch. Unlikely, but true, and the German brought down the favorite's favorite defense for the only win of the day. As the saying goes, every time somebody beats the Petroff an angel gets its wings. Naiditsch used a nice novelty, 19.Qd2, to put on the pressure. This was notable in itself because usually it's Kramnik with the new moves, easy equality, and a short draw in the Petroff. But as Kasparov put it to me later, "the Germans always seem to find a way through the Russian defense!" (Russian Defense, geddit?) He also pointed out that somewhere out there, Anand or one of his seconds is banging his head against a wall, shouting "Naiditsch used our Qd2!" Kasparov and his team analyzed Qd2 in 1999, but with a4-b4 inserted, a line they gave up on because after 16.a4 Black can just castle. Naiditsch's idea is more promising and Kasparov said "of course Black must take the rook [on e5]." That would have led to just the sort of dangerous, open-king position White was hoping for, but the alternative was worse. Kramnik will always take the long-suffering technical defense over a king on the run if he has the choice, but here his defensive skills weren't up to the task. If it's possible at all to draw the Q+R vs 2R+N, he didn't come close. GM Har-Zvi and I were surprised at how quickly Naiditsch wrapped things up.

The win brought both Naiditsch and Kramnik back to an even score. The other games were rather weak tea, although Mamedyarov-Gustafsson was a typically sharp battle in the Anti-Moscow Semi-Slav line that has become the hot tabiya over the past two years. The fireworks ended abruptly with a repetition draw. Nepomniachtchi grabbed the e4 pawn against Ivanchuk's preferred Sozin and this was enough to convince the Ukrainian to offer a quick draw on move 19. van Wely tested Leko's Queen's Indian and got nowhere. The innovative Bc1-d2-f4-e3-d2-e1 deployment is something of a crime against humanity. They kept the tension and almost all the pieces on the board for nearly three hours and then repeated moves for a draw. Anyone else nostalgic for MTel and the anti-short-draw rules yet? Leko and Gustafsson lead on +1 after three rounds.

Round 4: Kramnik-Leko, Nepomniachtchi-van Wely, Gustafsson-Ivanchuk, Naiditsch-Mamedyarov.

From the entirely unrelated Cake or Death department: at the risk of laughter-induced labor, the very gravid Ms Mig and I went to see the great Eddie Izzard at Radio City Music Hall on Sunday. (The first time I'd been there since watching Garry give a big business speech last October.) Fab show; go see him if you have the chance. If you don't, go buy all of his DVDs immediately.



you ought to fix this site, I'm pretty sure it's the ninja shouting hey you buy something that's causing the issues. unprofessional companies attract cheap employees. gotta job?

yes, please fix it.

Can't see Kramnik playing the Petroff v Anand. What about the French?? French Rubinstein ?? Will he ditch the Slav as well?? A lot of work for both teams. London to Elista book gives great insight into the work involved in preparing openings for and during World Championship matches.

Since the line 19. Qd2 is probably leading to a draw (according to Chessbase), Kramnik just needs to learn a couple of new computer variations to make his Petroff stronger. He doesn't have to rework entire lines of the Petroff.

So that's good news for Kramnik and the Petroff. Bad news for the angels.

I expect Kramnik will use the Petroff against Anand. There is nothing he would like more than to neutralize all the games where he has black. This loss was probably a blessing for Kramnik as it pointed out a weakness to shore up.

I don't understand the "with a4-b4 inserted" comment. The combination works just fine after 16.a4 b4 17.Bf4 etc.

Oh. Never mind. :-)

Obviously this variation can be looked at by Vlad but Anand and his team must be giving Vlad's full Petroff repertoire the real treatment. It boils down to whether Vlad feels invulnerable there or not. If Vishy starts whites with 1.e4 I'll be very interested what comes back over the net. Vlad can always start with Petroff to see if Vishy can lay a glove on him but develop another e4 response as well (or vice versa start with other e4 response and keep petroff in reserve but dont think this is as good match strategy). This is why matches are so much more interesting (to me) than San Luis type tournaments.

I get no kick from champagne...(do be doo)...mere alcohol...gimme Kramnik-Leko instead! Sigh sorry I'm just coming down from the adrenaline rush..

Anand probably has something much better prepared than a questionable rook sack, at least I hope so. In all likelihood, the Anand-Kramnik match will see the testing and demise (one way or another) of the 5.Nc3 attack in the Petroff. What on earth else is Kramnik going to bring? Anand is not the guy you want to play the Berlin against. Positionally weird junk is Anand's thing.

I'd expect something offbeat and classical; maybe a different non-a6 Lopez, the Rubenstein French, or the Caro Kann.

Mig - I've seen Eddie Izzard too, performing in the UK some years ago. Fantastic show. I'd noticed some of his DVD's on your Amazon wish list before now! Glad you enjoyed the show.

PS - I'll have cake - if there's any left. :)

@gmc: I don't think the rook sac is questionnable. It is the perfect kind of novelty Anand has to find to win his match against Kramnik.

Kramnik does not like risky lines and the best moves against this novelty were all risky.

Somehow i dont think that Kramnik will play the Petroff. He has played Sicilan Schwesnikow before, but that opening will suit Anand very well i think. The French Defense and Caro Kannis out of the question IMO. Why not The Ruy Lopez. I think thats defenitely possible. Kramnik will surely play 1. D4. Anand will proberbly not play setups where Kramnik can play the Catalan. So i think Anand will play the Slav defense.
The seconds will be a key factor in my opinion. Anand will have Peter Heine Nielsen on his team.

I would expect Anand to test Kramnik in the Semi-Slav on the theory that theoretical preparation is Kramnik's weak point, and that's a highly theoretical (THE highly theoretical, outside of the Najdorf) line.

The Petroff has been losing a bit of gloss of late; this isn't the first supertournament loss its suffered this year. And no one wins with it. Any thoughts on why the French and Caro are out of the question? I know Anand plays the Caro-Kann himself, but Anand plays the Najdorf too and that's never stopped anybody trotting it out against him. There are still a few variations of the French with decent theoretical status, too (MacCutcheon, Rubinstein).

The reason I think Kramnik will stick with the Petroff is that I believe he feels he can give up a rare loss or two with the Petroff given how many whites he is likely to win. Heck, he may not lose any blacks; Anand will have to come up with something to crack the Petroff, and how many 'somethings' is he likely to manage that are actually winning against it? I would take Kramnik's whites any day against Anand's chances of getting many wins against the Petroff. If Kramnik switches to something else he is playing against his own strength, which is to make his blacks irrelevant and his whites the only thing that count.

Kramnik showed that he and his openings are vulnerable to a fighting 2600. I think, on balance, that's good news for Anand. Not Kramnik. It's BAD news for the guy who LOST.

"So that's good news for Kramnik and the Petroff."

"This loss was probably a blessing for Kramnik as it pointed out a weakness to shore up."

Any new openings you try for the first unless you are in a desparate situation is not a good idea I think. That way you will now have one opening less which is not a good news for Kramnik. Kramnik is exposed! The one-match myth is busted!!

I still believe that finding new winning novelties in the Petroff will be extremely difficult. Anand may find one or maybe not, but will that be enough to win the match? I don't think so. Even new ideas such as what we just saw are typically easily neutralized after they are played the first time. I just don't see Anand's team coming up with enough new winning ideas against the Petroff to beat Kramnik, thus no reason for Kramnik to switch away from it.

IT's very difficult to do opening research during a match. KRamnik would be crazy to go into the match with just the Petroff, Slav (Black) and Catalan as white. It would be a tough call to change all three.
I'd see the Reti as a strong possibility for white, In such a short match you only get a few gos at an opening. You cant wait ages to introduce the Scotch (a la Garry 1990).....
What will Anand play as white..... maybe d4 and ignore the petroff or substitute.

Anand hasn't been all that convincing against the Berlin in the past. He almost lost to Luke McShane in the Bundesliga, for example. It wouldn't surprise me if we see some more of that, although it's not such a perfect weapon as it was against GK.

Otherwise I'd expect some Ruy line (if it's not wall-to-wall Petroffs which is definitely a possibility).

Anand, if you read this, please go for the king gambit in your match against Kramnik!

xtra, dont you remember that Fishcher, annoyed after losing to Spassky's Kings Gambit, found its refutation?!

@d_tal. I remembered reading something like this (and found something similar on wikipedia):

In Fischer's "A Bust to the King's Gambit", he put forth this idea and brashly claimed that it refuted the King's Gambit, which was clearly an overstatement. Fischer concluded the article with the famously arrogant line, "Of course white can always play differently in which case he merely loses differently."

I'm not close to be an expert in king's gambit, but I believe all openings can be busted if you know all the subvariations. However, it is impossible to do for a human being. Kramnik has almost busted the Petroff, however he can always loose in a line he did not take the time to analyze.

will not present problems of a sophisticated and lasting enough nature to trouble even somebody weaker than Kramnik. There are some lines where equality for Black is pretty straightforward and forcing; the more hairy lines can easily be avoided by Black. The Petroff is slight less played out. I think one indication of the non viability of the KG is that Garry who played almost every viable Kings pawn opening in his career never touched it as white in a serious game.. I would of course love to be proved wrong!

darn, some stupid error. My full comment was:

I am a big fan of the KG, but I sadly believe it will not present problems of a sophisticated and lasting enough nature to trouble even somebody weaker than Kramnik. There are some lines where equality for Black is pretty straightforward and forcing; the more hairy lines can easily be avoided by Black. The Petroff is slight less played out. I think one indication of the non viability of the KG is that Garry who played almost every viable Kings pawn opening in his career never touched it as white in a serious game.. I would of course love to be proved wrong!

If Anand has to use stuff like the King's Gambit, (don't rule it out, remember he used the Scandinavian against Kasparov in their 1995 match and actually got an advantage) it may make sense using it in the early part of the match rather than use it as a surprise weapon later on. The idea would be to send Kramnik's seconds on a wild-goose chase finding "refutations". Other than that, i dont see much use for it.

dtal & jaideepblue. I agree that King's Gambit (or a similar surprise opening) should be used only in one game to force Kramnik to study it, then Anand should revert back to normal openings. This game should lead to multiple subvariations that Kramnik would have to study carefully.

What are Anand's choice:
1- Play something else than 1.e4.
2- Play against the Petroff.
3- Play 1.e4, but avoid the Petroff by playing an irregular 2nd move.
4- It does not matter, Kramnik will not play the Petroff.
5- Prepare multiple openings, try multiple stuff.
6- Try to win with black.

I guess Anand, will go for option 5. Anyway Anand cannot find 6 new ideas in the KG. Maybe just one new idea would be enough to force Kramnik to lose precious time on it.

First Principles: Question - What do you look for when you're playing against Kramnik? Answer: Either messy and tactical, or a forcing line where Kramnik is on the side seeking complications.

Things to avoid: Giving up the two bishops, long term positional weaknesses, allowing him to get to an ending out of theory.

For Anand as Black, you can discard the following: The Catalan, Nimzo, and Slav. That leaves the Semi-Slav, various Indian setups, the QG, and perhaps the Benoni and Benko, if you're insane. (Hint: You are not).

Aside from the QID, Kramnik has more experience on both sides of the Indian defenses, oddly enough. I would expect the QID to be a backup rather than frontline option for Anand, however, as he is not particularly better prepared than other top players in that line.

That leaves the QG and Semi Slav, and of the two the latter has a better theoretical reputation and provides concrete positions more frequently.

Anand as White will play e4 and presumably finally put the Petroff to bed. If he is unable to do so, he shouldn't be in this match.

Presumably Garry had something against the Petroff in London 2000, else he wouldn't have tried 1 e4. He should share his opening research with Anand!!!! Maybe the KG will see an outing, who knows?! Fun to speculate though! I wouldn't rule out a switch to 1 d4 and/or the Catalan for Anand as white either!

I was actually thinking of the Benko as a surprise option for Anand. He once used it against some French IM in a must-win situation in the 2001 KO. If he is reasonably well prepared, it may be worth a chance to put Kramnik in uncertain waters as white.

Place your bets!
1.Anand will play the QGA. And the Reti as black.
2.Anand will not break the Petroff. This match will be a death blow to e4. He will switch to d4! After his experience against Kasparov there is NO WAY he will come with only 1. e4 prepared!
3.Both players will use at least two completely surprising lines..and likely one will mimic the other's favourite opening.
4.Kramnik could compete quite well against Anand as black in the Sicilian. Anand does not own this opening!
5.Kramnik will win by a narrow margin, but not before losing at least once.
6.And finally if Kramnik is a good psychologist he will ply the Sicilian Dragon to rattle Anand!
You saw it here first!


could you add a topic for Anand vs Kramnik in advance. I'm sure the topic would be discussed until their match. The person that's closest to the opening selection wins a World Championship match debriefing with Mr. Garry Kasparov.

Kasparov's a nice guy n'all, but no way is he getting my briefs.

Anand may play against the Reti as Black but cannot choose the opening. I doubt he would play an "independent" Reti line against Kramnik; he'll just play Semi-Slav lines and let Kramnik either transpose or concede equality at move seven.

Anand has played 1. e4 continually throughout his career and is exceedingly unlikely to vary in the most important match he has played since 1995, especially against a player who has seldom played 1. e4 as white and faltered against it relatively often as Black. If he brings anything else, it will be desperation-only. 2. Bc4 might be seen as a way to avoid Petroff theory later in the match. You can get Ruy-Lite type positions from the Bishop's.

What a shame that the World Championship is likely to be decided more by the quality of someone's (or someone's team's) prepared analysis than by the quality of chess ideas found and played over the board.

A divided Reti?? "independent" Reti and a "dependent" Reti?? :)

I think Anand will play 1.d4 mainly!! LEko played 1.d4 for first time in Brissago a la Fischer in Reyjavik with some success!! It is essential to surprise the opponent. Anyone that has ever sat an exam knows the difference between a predictable paper and an unexpected one. If you can prepare an opening X and do loads of work now you have a headstart on the opponent who will find it hard to catch up on prep during the match. Furthermore if Anand expends loads of work on say the Petroff and it doesnt come up it's analagous to studying stuff for an exam and it doesnt appear on the paper !! You cant study everything for either exam or world championship but the goal in both is to study smart. Very difficult to come up wuth a full new repertoire in the time available but there definitely will be surprises. Kramnik black either QGA or French Rubinstein as his surprise. Reti possible "surprise" as white.

Brian, let me guess: do you play French yourself, and it's one of your favorite openings?

Playjunior, I couldn't possibly discuss my opening repertoire on such a public forum :-)

Anand has played 1.d4 a few times over the years. The first time was in 1998 against Adams (here: http://www.chessgames.com/perl/chessgame?gid=1292534)

After the game, Anand jokingly said "I've wasted twenty years of my life playing 1.e4".

Despite that he has never shown any inclinations of moving away from king-pawn openings. There have been some sporadic attempts with 1.c4 but they haven't been very successful, and once he almost lost to Vallejo. His last outing with 1.d4, also against Adams was a loss in Linares 2005.

Based on this, I would think it is probable that Anand may deploy it once, just to rattle Kramnik but unlikely to repeat it.

Separate thread for the match please. The matter of seconds is yet to be discussed :)
My suggestion: Anand should approach Shirov and ask him to join his team.

French is the opening that suits Kramnik least. Petroff, Berlin, Ruy Lopez, even Caro-Cann might be an option, but French!?

The reason i suggested the Rubinstein is that it leads to very clear technical positions with a slight plus for white. You'd be wasting your time trying to squeeze a win from Vlad here !!

"Anand may play against the Reti as Black but cannot choose the opening."
Eh? Not to nit-pick, but yes he can:
Vlads fave Nf3 is followed by d5.
2.c4 is met by d4 or dc=Reti/or transpose to QGA.
d4 met by QGA.(other prediction).

When was the last time Kramnik played the Reti? Has he ever?

The Reti and the QDA are hardly the same thing. We haven't seen the Reti i a WC match since 1987 or 1990.

There are many transpositions between the closed Cataln, Reti and Queen's Indian. Where it becomes one and not the other can be a matter of semantics.

Yeah, true, Reverend.

Indeed, both Rubinstein French and Caro-Cannes both look like a good candidate as a weapon of choice for Kramnik. I think out of the two Kramnik should prefer Rubinstein, since there white doesn't have many different attacking schemes to choose from. The disadvantage with Rubinstein is that white may choose 3. e5. I doubt that Kramnik will enjoy position with the c8 bishop hemmed in.

In Berlin black is loosing far too many games lately, besides there is no surprise factor anymore. Berlin won't happen. Anand is VERY strong on the white side of closed Ruy Lopez. I think Kramnik will avoid closed Ruy Lopez/Marshall purely because of that factor. That leaves Petroff and some new opening (perhaps Rubinstein) as possible. Sicilian is out of the question. Nowadays Kramnik is too 1-dimensional to try it (besides black are having problems in Sveshnikov).

IMO, on the balance, Petroff is the easiest to defend in a match because it's a deeply studied opening with a rather narrow strip of lines to defend and no clear way even to a minimal plus. The downside for Petroff is that it's the most expected choice, so Anand will/should spend most of the time plotting against Petroff. I hope Kramnik ditches Petroff, since otherwise it would be so predictably boring and boringly predictable.

Kramnik is a very difficult match-up for Anand to play against in a match, as far as opening repertoire goes. Anand would have to pretty much refute the Petroff with white, while Kramnik with white will work with Catalan (which he used to great success including a nice win against Anand in Wijk 2007) and Semi-Slav, which definitely can pose problems to black, especially given Topalov's recent Nxf7 novelty and the fact that white gets very promising positions in that openings in general. And Semi-Slav just by its very nature lends itself to killer novelties much better compared to the Petroff. So, while Kramnik's "match repertoire" may be a liability in tournaments like Wijk, it is a great thing to have in a match.

Kramnik's loss to Naiditch is his 1st loss in the Petroff in what - 3 years? The Naiditch game won't refute the Petroff, it is basically a one game novelty, albeit a good one.

So, Anand is in a disadvantage as far as openings go. He is the best 1.e4 player in the world, but he happens to be matched against the defensive specialist who uses a pretty much bulletproof opening against 1.e4. So, Anand's best chance may be to dump 1.e4. But whether he will actually do it is another story. Leko worked very hard in preparation for the 2004 match versus Kramnik and had perhaps outprepared Kramnik, but Leko had 2 years to prepare for that match, while Anand had only one. And like, Anand is the best open games player in the world, Kramnik is the best closed games player in the world, so switching to 1.d4 and preparing well enough so that he can pose problems before Kramnik would be a huge task for Anand. I doubt he has the energy and determination to do it. I think he will just try to find some playable novelties in the Petroff, and will maybe try things like 2. Nc3 if he cannot do anything against the Petroff. But if Anand dumps 1.e4, it will be a huge hit to 1.e4, and a great symbolic acknowledgment of the superiority of 1.d4.

Kramnik may or may have a new "drawing weapon" prepared for Anand. He came up with the Berlin against Kasparov, and he used the Petroff against Leko, and maybe he had something new in his 3rd WC match against Topalov, but he never got a chance to use it since Topalov didn't feel like wasting a game with white on 1.e4. But the next great drawing weapon may be there and ready to go. I would guess it is either Ruy Lopez Marshall or Caro-Kan. But I think the most likely scenario is Kramnik sticking with the Petroff. I'd give 55% chance of Kramnik sticking with the Petroff, 25% chance of a Marshall, 10% chance of Caro-Kan and 10% chance for other options.

French Rubinstein? Are you people serious? Kramnik will play 1... h5 before he plays the French Rubinstein. I play he French Rubinstein myself and I can tell you it is a painful opening to play. You might say: "so is the Berlin", but the analogy doesn't work. The Berlin made perfect sense because Kasparov without queens on board wasn't the same as Kasparov WITH queens on board, so playing the Berlin made perfect sense. Kramnik was willing to be tortured in the Berlin endgame, because even though it was a very unpleasant endgame, he felt it was an endgame he was much more comfortable with defending than Kasparov was with squeezing. But playing French Rubinstein against Anand would be torturing yourself WITHOUT any rationale at all. Look up the stats: Anand is the best player in the world against the French Defense. He has a performance rating of 2976 in 16 games versus the French since 2000 (compare that to Kasparov's performance of 2851 on white side of the French in 7 games, other anti-French specialist Ponomariov's 2893 in 14 games and Morozevich with 2837 in 16 games in the same period). Anand is as dominant versus the French as no other player dominates against any other opening, with the exception of Kramnik's 3000-ish performance rating in the Catalan. And Anand doesn't just beat people in the French. Check out his games. He makes people like Bareev look silly in the Rubinstein. So of all the advice one could have given Kramnik about opening preparation, suggesting the French Rubinstein may just be the worst :)

Hmm, disagree with pretty much every single conclusion of Russianbear. Yes Kramnik's black repertoire will pose huge problems for Anand, but Anand's black repertoire will pose similarly difficult problems for Kramnik.

Black can avoid the Catalan, and Semi-Slav gives black excellent counter-chances both theoretically and practically and, most importantly, leads to position that suit Anand much more than Kramnik. As for Topalov's Nxf7 being a serious threat to Anand, that must be a joke right? First, I don't think white won a single high level game in that line except the original one. Secondly, I doubt that an idea of playing a speculative sac (i.e. the one that doesn't guarantee him a safe exit) has any appeal to Kramnik at all.

As for French Rubinshtein and Anand's great score against french... French Rubinstein leads to positions absolutely unlike the usual French(meaning classical or Vinaver or McCutcheon). So Anand's stats against the French are irrelevant. Anand's stats against Rubinstein (Megabase 2008, both players over 2600, including rapids, not including blitz and blinds) are +4 -2 = 6, performance 2715. Not that impressive really. Anands score against Bareev in Rubinstein is 0.5-0.5. If somebody looks silly, it's certainly not Bareev here.

The game that Anand should study most closely is Kramnik-Carlsen from Wijk aan Zee. I think positions such as those that arose in that game is what he would be striving to achieve. As for the indestructibility of the Petroff, I'm assuming that Anand is content to try to grind it out - his game has changed since his glory days and has become far more drier.

What a great game by Chucky, looked like he was determined to beat the Petroff with any and all means necessary. BTW Anand does not need to lokk at anything he wil find his own way.

And Vlad loses again in the last round, again w. the Petroff (Mig, it's a good thing it's your wife that's pregnant and not you, else this would no doubt prompt the baby to come out to celebrate)...seems like he came out of the opening ok, then drifted/got complacent; something he (of course) can't afford to do against Anand. It's not just his play and opening preparation that needs to be sharpened for the big match; it's his attitude.

...and that looks like the end of Petroff for Kramnik. Botvinnik once said that you should employ the opening only as long as it brings you success. Btw, Kramnik already lost a match with a Petroff to Shirov.

Great game by Chuky btw. These days Ivanchuk looks like an improved version of Kramnik: capable of converting paper thin advantage yet much more aggressive and versatile.

As my post said, "grinding" down the Petroff is perfectly acceptable - the disadvantage for Kramnik is that there is no risk for white in such positions.

"As for French Rubinshtein and Anand's great score against french... French Rubinstein leads to positions absolutely unlike the usual French(meaning classical or Vinaver or McCutcheon). So Anand's stats against the French are irrelevant. Anand's stats against Rubinstein (Megabase 2008, both players over 2600, including rapids, not including blitz and blinds) are +4 -2 = 6, performance 2715. Not that impressive really. Anands score against Bareev in Rubinstein is 0.5-0.5. "

Rubinstein does resemble the Burn variation a lot. In fact, the Anand-Bareev game I was thinking of was Burn variation, but the structure is roughly the same and Burn is arguably is superior to Rubinstein, as black often gets the two bishops for his troubles. Anand may not be as dominant against Rubinstein/Burn structure (where black gives up the center) as he is in other lines of the French, but I still think it is a big mistake to play something like that against Anand. Anyway, like I said, we are not likely to see Kramnik play it, so this doesn't really matter.

Admittedly, when doing the stats I overlooked the possible transposition after 1.e4 c5 2.d4 d5 3.Nc3 Nf6 4.Bg5 to the line that I would also consider Rubinstein French. Here Anand scored +2 = 6 -0 against ELO 2694 for the 2787 performance. That's of course a very good fare, but still well below Anand's lifetime 2829 performance with white (all my stats include rapids, but exclude blitz and blind).

I also looked at Anand's wins in Rubinstein, and a couple of these games he would have lost against the best play (including the game against Bareev that you mentioned).

Based on this little stat analysis, Kramnik's style and comparatively simple positions that ensue, I think that playing Rubinstein is a very decent idea for Kramnik.

As for the probability of Kramnik actually employing this line...I think it's reasonably high. The most expected choice is Petroff, where Anand has much more experience, likely will do most of his preparation and has a lifetime performance of 2825...

Has Kramnik ever played the French? In a serious game?

Interesting analysis.

1. Black can't choose the Reti because it begins 1.Nf3 and 2.c4 by White. I don't see Kramnik going with it because Anand can answer 1...Nf6, 2...c6 without showing anything. Anand can't force it, though, and Kramnik has no actual reason to avoid transposition by 3. d4.

2. In retrospect I think I agree the French probably won't be seen in Kramnik's repertoire. He prefers open play for minor pieces, even at the expense of development or strategic factors. Oddly enough, the Sveshnikov might be a possibility (two bishops compensate for positional disadvantage just as in the Berlin). Kramnik used to play it.

3. I think the opening situation favors Kramnik, but that Kramnik will have a difficult time actually playing up to Anand's level. His middlegame play has been dicey lately... The loss to Carlsen at Corus, two Petroff losses at Dortmund...

gmc wrote:
"In retrospect I think I agree the French probably won't be seen in Kramnik's repertoire. He prefers open play for minor pieces,...Sveshnikov might be a possibility(two bishops compensate for positional disadvantage just as in the Berlin)"

I guess it's high time for some openings 101. Rubinstein French is not your typical French. It leads to open play and black often gets two bishops at the expense of development. One possible line is 1. e4 e6 2. d4 d5 3. Nc3 de 4. Nxe4 Nf6 5. Bg5 Be7 6. Bxf6 Bxf6...

For opening surprises, and to make Anand's seconds work harder I guess throwing in a French somewhere can't hurt Kramnik.

I wouldn't say Rubinstein French is not a typical French. Rubinstein/Burn is as typical as just about anything else in the French. In Rubinstein, black gives up the center without getting much in return. That reason probably makes it theoretically inferior to more "typical" lines that don't give up the center, IMO.

On the Rubenstein:

I don't play the Rubenstein in serious games (giving up on my years of trundling through the Najdorf and Taimanov would be silly at this point), though I've employed it often enough in blitz and rapid games.

I don't think the opening is "strategically inferior" to the closed French lines. The Rubenstein is to the Sicilian what the Leningrad Dutch is to the King's Indian.

In the KID, Black often sets up and plays e7-e5, planning f7-f5 to contest light squares. The Leningrad Dutch starts out with the same setup, but the f pawn advanced, seeking to extend the e-pawn later.

Likewise the Rubenstein French is an attempt to gain an e-pawn vs. c-pawn position as in a Sicilian after d7-d5xe4. But instead of playing c5 initially, Black plays d7-d5 first, supported by the e-pawn, and only then attempts c7-c5 to clear away the second White central pawn. In either case, the goal is to eventually contest all the central squares and to come away with an extra central pawn. Like the Leningrad Dutch, though, the Rubenstein gives away a bit more positionally than it's cousin. If anything, I'd say that the Rubenstein/Burn's lines seem more first-glance-sensible, trading away the pawn that otherwise cramps Black at e5.

It's weird that I'm advocating the Rubinstein French, given that I've never played it in my life (not even in a blitz game), but I've faced it plenty as white and IMO it should suit Kramnik's style very well.

I disagree with Russianbear that Rubinstein is inferior to closed French and that black gives up the center without getting nothing in return. Black gets in return a very solid position, easy development, well placed pieces, and, sometimes, two bishops. It's a typical hypermodern strategy: white has more presence in the center, but black compensates it with piece pressure and possibility of pawn breaks (c5). No denying, white is more active, but it's quite difficult to convert this activity into anything substantial because of above mentioned features of black's position.

In comparison with closed french it's not inferior, it's just different. Essentially closed french is a fighting opening, where black takes serious risks (both strategical and tactical) to get unbalanced position with counterattacking chances. Closed french is an opening to play for a win, but it's a very risky way and black is crushed often. On the contrary, Rubinstein french is a wussy opening, an opening to play for a draw, but it's a good way to do it.

So far Rubinstein is not as popular as Petroff/Berlin/Marshall, but IMO it's just a fashion thing. I believe objectively probably it's not any worse than Petroff and definitely better than Berlin (now that everybody is prepared for it) and if Kramnik adopts it, he would fare well with it.

gmc, I don't think the French players are worried about the e4-e5 pawn push. In fact, if you survey the people who play the French, you are likely to find that people love to play against lines where white plays e4-e5 the most.

I definitely think Rubinstein/Burn are inferior to most other lines of the French. The only positive thing about Rubinstein is one can play it against both 3. Nc3 and 3.Nd2, so it cuts down on preparation. So a lot of people who are just starting to play the French play the Rubinstein when they haven't learned a line against the Tarrasch or the 3.Nc3 yet. That's how I came to play the Rubinstein myself. But make no mistake, it doesn't just look bad, it also IS bad :) Just check out chessgames.com. 3... dxe4 are some of the worst scoring lines of French. To give up the center for nothing before white even had a chance to commit any pieces anywhere so white can develop around it is just something that can't objectively be good for black. So no wonder white scores so much against it. IMO, only a person who is unfamiliar with dyamics of Rubinstein can say "Black gets in return a very solid position, easy development, well placed pieces". I am not sure Rubinstein is better than Berlin, and it is definitely inferior if we are to compare the Berlin against Kasparov to the Rubinstein versus Anand. Berlin is like a dream opening against Kasparov, while Rubinstein is hardly that if one faces Anand.

I don't seriously disagree with any of the above except that Black has "well placed pieces" in the Rubinstein French. My limited experience suggests that poor piece placement is precisely Black's problem. Black is forced to deal with a bad c8-bishop in an open position; developing it to b7 is slow and leaves light squared weaknesses, while other developments are artificial.

Black's goal is to trade everything in sight until he can get the c8-Bishop activated, and then commence playing with the extra central pawn by e6-e5, e5-e4, maybe f7-f5-f4. Acheiving this position is difficult, so Black wins few games.

I've played the dxe4 variation in online blitz games. I find a big problem in placing the queen. After white goes 0-0-0, black can't go for the traditional c5 as the queen is in the firing line.

The Rubenstein French is like the Sicilian in another way, too; it spots White a lead in development in return for a bit of space between the forces, a lack of weak points, and the extra-central-pawn potential.

When you do that, you tend to run into rook-opposite-queen problems. In the Sicilian, this is mitigated by the availability of c7 for the queen.

I believe in the Rubenstein the response is the same; develop the Queen off the d-file. I don't know the theory, but I've played ...c6, ...Qa5 and ...Qd5-a5. I remember looking at ...b6, ...Bb7, ...Qb8 in a game at one point, but I think I played something simpler.

Putting all the above factors together, it seems the answer is clear: Begin with 1...b6 and 2...Bb7.

Black gets far more value (in the form of useful flexibility) by deferring the central pawn push -- which in the English Defense could end up being either ...d5 or ...e5 or ...c5 or ...f5, depending on how the opponent develops -- than he gets by deferring the fianchetto until after he's already made the commitment of pushing the d-pawn and opening the center, as in the Rubinstein.

Well, we can debate the merits all day, but the problem with the English Defense is that it allows White to literally do whatever he wants in the center. It's like a QID where you don't try to stop e2-e4 (which is to say, a QID that's strategically lost).

I've played the English in rapid and blitz, much like the Rubinstein, and never really been anamoured of anything except the f7-f5, g7-g5-g4 attacks you get if White plays passively. White can just fill the center with pawns and close off Black's play by d4-d5 whenever he wants.

And if you play ...d5 yourself, why aren't you playing the French?

Russianbear wrote:
"But make no mistake, it doesn't just look bad, it also IS bad :) Just check out chessgames.com. 3... dxe4 are some of the worst scoring lines of French."

I don't know what are the stats on chessgames. It would be more helpful if you actually posted them thus giving your opinion some factual ground. However the stats of megabase 2008 do not corroborate your thesis.
I checked the games played no earlier than year 2000 with both players over 2500.
In position after 1. e4 e6 2. d4 d5 3. Nc3 de 4. Nxe4 black rated 2596 scores 43% for the performance of 2569.

In another key position after 1. e4 e6 2. d4 d5 3. Nc3 Nf6 4. Bg5 de 5. Nxe4 Be7 black rated 2614 scores 45% for the performance of 2577.

For comparison, in Vinaver after 1. e4 e6 2. d4 d5 3. Nc3 Bb4 4. e5 black rated 2577 scores 41% for the performance of 2536.

In the classical line after 1.e4 e6 2. d4 d5 3. Nc3 Nf6 4. Bg5 Be7 5. e5 black rated 2585 scores 44% for the performance of 2543.

For me this clearly shows that on a decent skill level all of the above lines are equivalent statistically.

Moreover, in Petroff after 1. e4 e5 2. Nf3 Nf6 3. Nxe5 d6 4. Nf3 Nxe4 black rated 2627 scores 42% for the performance of 2580. Not really any better than Rubinstein. Now I'm waiting for Russianbear's rationalization why Petroff "doesn't just look bad, it is bad :)".

According to chessgames.com after 1. e4 e6 2. d4 d5 3. Nc3, 3... dxe4 scores the worst out of major moves:

After 3...Bb4 white scores 39.9%, 26.9 for black.
After 3...Nf6 white scores 38.6%, 25.3 for black.
After 3...Be7 white scores 39.3%, 24.6% for black.
After 3...dxe4 white scores 46%, 14.9 for black.

In Steinitz (1. e4 e6 2. d4 d5 3. Nc3 Nf6 4.e5 ), white scores 39%, black 22%.

In Advanced (3.e5), white scores 39%, black 32.6%.

In Exchange (3. exd5), white scores 26.5%, black scores 29.5 %.

After 1. e4 e6 2. d4 d5 3. Nd2, 3... dxe4 again scores the worst out of 4-5 major moves:

3...c5: 30.4% for white, 20.3% for black
3...Nf6: 45.1% for white, 22.7% for black
3...dxe4: 44.6% for white, 14.7% for black
3...Be7: 33.2% for white, 28.3% for black
3...Nc6: 43.6% for white, 24.2% for black

even 3...a6 scores higher in 300+ games:
32.2% for white, 23.8% for black.

In Kings Indian attack, there are two main lines: 2. d3 and 2.Qe2

After 1.e4 e6 2. d3: white has 40.5%, black 30.9%
After 1.e4 e6 2. Qe2: white has 47.2%, black 28.1%

So, 3... dxe4 scores worse than the alternatives after both 3. Nd2 and 3.Nc3, and it scores worse compared to other major lines of the French, which is not surprising, since black gives up the center without getting anything in return.

First, these stats lack the crucial info: average ratings of white and black or, alternatively, the average rating of black and their performance. That way percentages are not directly comparable.

Second, I assume that calculation is based on all of the games on chessgames. Is that correct? If so, that also includes a lot of amateur games. Together with chessbase stats that just shows that amateurs should stay away from Rubinstein French (since GM results are respectable).

Finally, what is the cutoff year? It's pointless to count games played half a century ago since the theory has changed considerably, critical positions are different, etc.

To give you a taste of what's going on, according to the same megabase white scores 56% overall after 1. e4 e5 2. Nf3, while whopping 58% after 1.e4 e5 2. d4.

chessgames.com stats may not be a correct indication of what is going on in state-of-the-art theory. But I do think that over thousands of games, things average out, so white's average rating is going to be quite close to that of black in most of the main lines, so the scores for openings are likely to correlate with performance ratings. I would say year and rating cutoffs are useful, but are not necessary to get the feel of how various opening moves score this early in the game. I could see how it could matter if we go further, as some lines may have been refuted long ago, etc, but here we are looking at very early stages where the development of theory probably doesn't matter that much.

Chessgames.com also has a slightly better score for 1.e4 e5 2. d4, compared to 2.Nf3. But that may just be because it is possibly a slightly better move. After all, the Petroff, the Marshall, and even the main lines of Ruy Anti-Marshall do offer great chances to equalize for black, so 2.d4 at least has the advantage of bypassing all that. 2.d4 may well be a better suggestion for Anand than Rubinstein French would be for Kramnik.

"Chessgames.com also has a slightly better score for 1.e4 e5 2. d4, compared to 2.Nf3. But that may just be because it is possibly a slightly better move."

Unfortunately (or fortunately) the deal here is pretty simple: after 2. d4 it's white who fights for equality at high level. There can be no debate here: 2. Nf3 is superior to 2. d4 on GM level.

You probably have forgotten that 2 years ago Nepomniaschiy tried 2. d4 at Aeroflot. He got a promising (at least interesting) position in the first game which he drew. In the second game he quickly got into slightly worst position and lost after a long fight. In the third game he was blown off the board by Vallejo Pons and that was the end of Central opening for him. Here is the game:
No other strong GM tried this line ever since.

This is precisely the problem with all stats that are not filtered by rating: many lines that are perfectly good at amateur level are unplayable when the stakes get higher and vice versa.

Then the idea that over thousands of games ratings of white and black averages out is also flawed. Certain openings (such as Rubinstein French) are played to draw and are more likely to be chosen by underdogs willing to split a point. On the other hand, Vinaver French is a fighting choice and therefore would likely to be chosen by rating favorite willing to take a risk in order to win with black.

Year cutoff very early in the opening is a more subtle issue. Probably, in many cases it's not necessary, but there also bound to be some cases when it's indispensable. I'd rather do it just to be on the safe side. As long as you have enough games for the meaningful stats, there is no harm in doing it.

I visited this page first time to get info on people search and found it Very Good Job of acknowledgment and a marvelous source of info.........Thanks Admin! http://www.bestphonelookup.com

Twitter Updates

    Follow me on Twitter



    About this Entry

    This page contains a single entry by Mig published on July 2, 2008 2:28 AM.

    1st, 2nd, and 3rd Blood in Dortmund was the previous entry in this blog.

    Outsiders On Top In Dortmund is the next entry in this blog.

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