Were Tal, Korchnoi and Geller 2200 rated players?

by Albert Silver
10/22/2021 – This was the question that a friend, grandmaster Alex Yermolinsky, posed after being inquired about a comment by another grandmaster. It was meant as irony, but was the logical result of a series of questions and comments that led to him posing this. He had explained with hard data how one could even ask this. If you want to know how we got to this point, read on!

ChessBase 17 - Mega package - Edition 2024 ChessBase 17 - Mega package - Edition 2024

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.

More...

It all started on a rather ordinary day while studying one of the games in Ludek Pachman’s great trilogy, “Complete Chess Strategy”. I had been rereading it in an effort to ‘reboot’ my chess, so as to ground my game once more on solid concepts before trying to steer it forward.

In the chapter on blockading pawns, expertly taught by the Czech grandmaster, he brings up a game he played against Viktor Korchnoi in 1954. This is not done out of hubris but, as he aptly explains, it is one with which he will have a much deeper and more intimate knowledge of what took place and why. 

A modern classic on positional play, this is volume two of "Complete Chess Strategy"

It is a decidedly unusual opening in which things go sour for the author. Trying to explain where he went astray, he gives a position and short line with the conclusion, “would give Black a good game.” Normally I would not spend too much time on a sidenote, which is not crucial to the lesson, but it was so contrary to what I thought of it, I stopped and scratched my head. Black has a good game? Really?

 

Since I was playing through the game and notes on ChessBase, I could easily just fire up an engine with a quick click on the keyboard, but it is one principle I do not deviate from: no engines when studying. I need to develop my brain and skill, and not my dependency on engines. So I spent a full minute analyzing and my conclusion was “I think Pachman is off his rocker”. Delusional? Arrogance? Hardly. It was simply my impression based on what I saw. I could easily be dead wrong, and I accepted that.

I decided to message a GM friend. Not Yermo. Not yet. I explained my consternation, my confusion. I also explained my refusal to ask an engine. Without even bothering to wait for me to send the position, he exclaimed, “Let us be honest, those guys knew nothing about chess. Pachman was like what? 2200?” "Now you are being excessively harsh too. He was not 2200", I replied. "He was", my friend insisted. I copy-and-pasted some of Pachman's CV from Wikipedia into our chat, but he was not dissuaded.

Ludek Pachman was a hard-working grandmaster who wrote and published no fewer than 80 books throughout his career. He considered his book on strategy to be his best work.

Spending a lot of time arguing this was obviously pointless, and he was entitled to his opinion after all. Not being anything near a GM myself, I was on shaky ground to debate this with him anyhow. Still, this made me wonder: do modern grandmasters really look down their noses on the skill of their predecessors like that?

I now messaged the veteran grandmaster whose strong views never left any doubts where he stood, Alex Yermolinsky. I explained the story and why I was asking him this. He told me he had an opinion on the matter, and asked me to give him a few minutes while he prepared some material to answer me. Some minutes later he sent me a small database with eight games in it. A bit confused as there was no explanation accompanying them. Was I supposed to study this? “No study. Just see them.” I did and all became clear.

The first three games were played by Tal, Korchnoi, and Geller in 1955, all in different events, and they all reached this position with white:

 

With no knowledge of the theory of the position, nor any need to consult 3500-Elo programs, the move that immediately draws one’s attention is the straightforward 12. e5! Since the line 12… dxe5 13. fxe5 Nd5 14. Nxd5 winning a pawn seems irrefutable. The point, as you may have surmised, is that none of the three legends played it. Tal won his game as a result of a concerted effort at hara-kiri by his opponent, though not from any special play in the opening, while Korchnoi and Geller each got nothing and drew.

These are the first three games out of the eight. The next games are played in the 50s and 60s by players of far lesser note, all of whom played 12. e5! and all of whom won their respective bouts.

“Incredible, isn't it?”, he told me. “A 2200 of today would play 12. e5 with no hesitation. Shall we conclude that Tal, Korchnoi and Geller were 2200 strength?” It is a rhetorical question. “My point is, it's impossible to estimate the strength of a player by small sample.”

It is a valuable lesson, even when it is one we think we have learned and mastered. Does this mean we should cease to question and analyze moments of disagreement? Of course not, but nor should we be too quick to dole out judgement based on that person’s lapse.

A special thank you to GM Alex Yermolinsky for the generous lesson, and to my unnamed GM friend whose comments led to this discussion, lesson, and article.

You can see my personal on-the-fly analysis of the position in the video above at 21:03. The video also includes a missed tactic (4:28) as well as a successful solution of one of Dvoretsky's endgame study exercises ending after which I go a bit nuts (42:50)

Born in the US, he grew up in Paris, France, where he completed his Baccalaureat, and after college moved to Rio de Janeiro, Brazil. He had a peak rating of 2240 FIDE, and was a key designer of Chess Assistant 6. In 2010 he joined the ChessBase family as an editor and writer at ChessBase News. He is also a passionate photographer with work appearing in numerous publications, and the content creator of the YouTube channel, Chess & Tech.

Discuss

Rules for reader comments

 
 

Not registered yet? Register

oxygenes oxygenes 10/22/2021 04:09
Author could use engine at least for check public release material, if not for position analyse. Otherwise he could not produce variant 12.-dxe5, because 13.Bxf6 and "finito". :) And there can be psychologicaly understable 3 GMs view, where they do not want leave opticaly promising atacking position, goes to endgame, where white has some advantage, but it requres good technique to reach something tangible. 12.- Nd5 13.Nxd5 Qxd2 14.Rxd2 exd5 - 15.a4!?
Frits Fritschy Frits Fritschy 10/22/2021 03:59
I just looked at the databases, and seeing the (although seemingly insufficient) counterplay black gets, my guess is that any IM level player nowadays would take at least half an hour before playing 12. e5, and very probably even would follow Tal c.s., to reevaluate the position at home after the game with a strong engine.
Frits Fritschy Frits Fritschy 10/22/2021 03:40
AgainAgain,
That's what I thought, but let's continue. Without a computer or looking at the databases, I got to 12 e5 b4 13 exf6 bxc3 14 fxg7 cxd2 (14... Bxg7 15 Qxd6) 15 gxh8D Nb4 16 a3 Ba4 17 Rxd2 Bxc2+ 18 Ka1 and what now? Black is a rook behind and while his position looks threatening, he still has to consider Bh6.
However, I can't imagine a 2200 player nowadays would play 12 e5 instantly. An 1800 player might, happy to show his theoretical knowledge. But a 2200 player would smell a rat – every Sicilian player by now knows this position (I guess, I'm not one of them), so why is black playing this anyway?
And that probably shows the problem: Tal, Geller and Korchnoi were quite capable of calculating up till 18 Ka1 and much further than that. But they had to calculate all that. A 2200 player doesn't have to calculate, he just has to remember. Which makes the two incomparable.
By the way, if the other five games were won by white in the 1950's an 60's with 12 e5, why are these players of a 'far lesser note'? They played in the same era as the three top GM's, so following GM Yermolinsky's reasoning, they should have had no problem with a world champion and two candidates. My guess is that they just had to follow analysis given after the three games, like 2200 players nowadays, or didn't calculate beyond 18 Ka1 and were lucky that there is nothing there for black (if there isn't).
TheDock TheDock 10/22/2021 03:38
I recommend all to read Willy Hendricks book On the origin of moves. Close to the same conclusion.
HowardGutman HowardGutman 10/22/2021 03:35
The second example is a little trickier than presented. In response to e5, b4 seems the obvious retort and one needs to calculate exf6, bxc3, fxg7!, bx 7, with Q-d6 giving white the advantage. I think that's 2,400-2600 calculation (Fritz found this for me). The article is fun, but there are a number of games with the opposite conclusion.

I think the older games were a little easier to look at, because there were clear and logical plans executed.
Computers don't plan, and games today are more haphazard and complicated, like comparing Kasparov with Fischer.
JoniCee65 JoniCee65 10/22/2021 03:14
After 1.e5, if Black plays b4 (as suggested below) he gets blown to bits by 2. exf6 bxc3 3. fxg7. Or am I missing something?
Depsipeptide Depsipeptide 10/22/2021 02:25
Firstly, annotations are written with the view of hindsight: the commentator's views are influenced by the game outcome. I have not looked at Korchnoi v Pachman, but would guess he considers the position after Nd6 better than what actually happened and hence an improvement. Secondly, classical annotations without the benefit of engines often miss key details. You should not be surprised that they are sometimes wrong, but marvel at how often they remain true and accurate! As for the Sicilian position, I do not understand the point unless you are expecting Tal, Korchnoi and Geller to play perfectly in every game. I certainly disagree that 'A 2200 of today would play 12. e5 with no hesitation'. In the opening phase, many inaccuracies are unpunished because the players just want to get to a familiar middlegame without carefully looking for refutations of move order nuances.
PhishMaster PhishMaster 10/22/2021 02:11
But yes, Pachman is dead wrong about black being OK there. It is hard to defend the king. He was just trying to demonstrate the whole knight blockader idea, but what writer, pre-computers, has not taken liberties like that?
PhishMaster PhishMaster 10/22/2021 02:05
Anyone can play a 2200-level move, or a 1500-level move, better or worse. Of course, Kramnik was once mated in one, and just recently, Aronian thought for about 5 minutes, and played 18.Ba3?? allowing mate-in-one. You cannot judge someone's ability by a few moves, but by their volume of work.

As a player, who was over 2200, and who has read the games of Tal (in particular the book "Study Chess with Tal"), Geller, Kortchnoi, and even Pachman, their abilities were FAR beyond mine.

Your GM friend may be good at chess, but is clearly not a high-level thinker, and that is the nice way of putting it.
AgainAgain AgainAgain 10/22/2021 12:29
After e5, black plays b4. How can you even think about dxe5?
Previous12