Kasparov in St. Louis: a closer look

Advertising

Books, boards, sets: Chess Niggemann

Marco Baldauf Marco Baldauf, born 1990, has been playing since he was eight. In 2000 and 2002 he became German Junior Champion, in 2014 he became International Master. He plays for SF Berlin in the Bundesliga.

Discuss

Rules for reader comments

Resistance 9/5/2017 09:17

@ Petrarlsen ---

That you have decided that results (numbers) are the only thing upon which you'll base your analyses and statements regarding the issues being discussed here, doesn't mean in any way that you are (therefore) right. For you haven't proved that results (numbers) it's all that it is needed here to prove your point. Again, you've merely decided, yourself, that results will be all you'll be considering here. Yet, not because a person has decided that he/she is right, then he/she is right (--right, my friend?--). So, you need to prove your point first, but right up until now you haven't been able to. And there might be a reason for it: that your argument doesn't work.

You say: "if a player wins, and wins, and wins... he MUST necessarily play good chess". That is, evidently, an invalid inference. Roughly speaking, that a player wins game after game only means (necessarily) that he is playing better than his opponent(s), not that he "MUST" be playing good chess. For example, if we played a 6-game chess match, and I beat you by 6-0, it wouldn't necessarily mean that I played good chess: it would only mean that I played better than you (--that my little sister won a 6-game chess match against my cat by the score of 6-0, wouldn't necessarily mean that she played "good" chess either, nor that she will become the next Bobby Fischer (since Bobby also won a match once by a score of said magnitud). That you win at chess doesn't necessarily mean that you also play well--). Ultimately, results (numbers) have nothing to do with somebody's chess level, since they, as such, do not tell you about players' actual strengths nor capabilities: they can only tell you about who won, and for how much (--i.e., they can only tell you about players' relative strength to each other. For they are not the players' actual strengths nor capabilities: they're just numbers, like the one at my house's entrance door--).

Many people know that Garry Kasparov is a really strong player, not because he is (was) the number one player in the world for so many years, but because he beat those many people AT THE BOARD throughout all those years. If you never played Kasparov, on the other hand, you can only assume he is (was) that strong. However, since Kasparov beat Anand, and Anand beat my cousin (at a simul), and my cousin is much, much better than me at chess (he has beaten me pretty frequently), then I have no problem believing that he is (was) such a strong player (--because if he was able to beat someone that beat my cousin at a simul, and my cousin is able to beat me whenever he wants to, then it is pretty possible that that guy Kasparov is a very, very strong chess player--). Yet, you also have the players' games themselves: there you can see how strong a chess player really is, or was...

Magnus Carlsen did not beat the same Viswanathan Anand that won in Mexico, in 2007 (--Anand was clearly not the same guy--), though he, Anand, won both Candidates Tournaments (2007 and 2014). The 'ease', on the other hand, with which Magnus won both those matches (2013 & 2014) cannot be used to say that he stands at the same level than Garry Kasparov. By following his games, however, we can see that he, Magnus, is also very, very strong. A match against Garry, both having the same age, would've been a great chess confrontation...

Petrarlsen 9/2/2017 05:25

@ imdvb_8793 : Yes, I think too that, now, we have really exchanged all our important arguments !

And now, as you said, it's time for the World Cup !!

imdvb_8793 9/1/2017 05:14

"Classical, Rapid, and Blitz chess are "quite different things""

Believe me, I agree more than you think! I use all three forms of play in my training, more or less equally - at least I try (and, of course, I've played many tournaments in all of these time formats, over the years) -, and it's a COMPLETELY different thing when I play either - I feel like I'm training COMPLETELY different aspects of my play, using entirely different skills... Of course, it IS the same game, after all, so maybe 'completely' isn't the best word to use, but I'm exaggerating to make a point, because, as much as that's possible, given that it's still chess, I, personally, don't feel like I'm playing the same game when I play blitz compared to rapid, or when I play rapid compared to classical. (This might, however, I should probably point out, be slightly different for the absolute top players, who think FAR more quickly than us regular folk, and can get to the heart of the position a lot faster, as well as calculate the key lines, without fault, most of the time, much more easily. They might not feel the difference between the three formats quite as much as you or I do - but, of course, I can't speak to that, as I'm nowhere near that good.)

"I think that these differences of approach rather explain why, founding our argumentations on well-asserted facts, our conclusions are nonetheless different ; but perhaps you will disagree with this also !!"

No, I think I agree pretty much completely. :) This is all simply an illustration of how subjective/speculative the whole debate is, by its nature.

* Hey, we did it! Just in time! Since the World Cup starts tomorrow, had the discussion continued beyond that, I would have found it hard to set aside the time to properly answer on this topic (it takes me a while - I'm not a fast thinker, never have been, because I tend to question every word/phrasing, edit and re-edit, and so on), as I get very immersed in following these major competitions, via live feed+commentary, and not only, plus I, of course, have other things I try not to neglect at the same time, which ends up taking up 99% of my waking time, I've come to learn... So, in a way, this was my last "free day" before that. :) Perfect timing! I love it when that happens - all is right with the world...

imdvb_8793 9/1/2017 05:14

"In fact, I think, reflecting more generally on our "debate", that the crux of our differences of view is our differing conception of a "peak period"."

Definitely one of them, at the very least. :) The inflation thing might be another. And there are probably others that are less important, but, yes, the definition of "peak period" is very likely the biggest point we disagree on. Nothing wrong with that!

"I consider that it is the period when a player isn't making anymore significant progress, and hasn't yet begun to decline significantly."

Like I said, I think this is another reasonable way to define it, even though it doesn't coincide with my own. (Because I think form - and everything it entails, some examples of which I've already mentioned more than once - is much more important in chess than in the other fields you mentioned, for example, in determining the results of a - in the case of chess - player's "work" in any given tournament. But this, too, is debatable and far from set in stone as being fact, and I recognize that.)

"I don't think necessary to enter in all the details of their respective form at the moment of the tournament"

I understand - again, of course, this is all very imprecise stuff, and mostly speculation, by nature. Same as with establishing a definition for "peak" - there's no scientific and unquestionable way of doing it, that's demonstrably better than all others.

"at the beginning of Carlsen's period as an "absolute top-player", he had difficulties playing Anand."

This IS unquestionable. :)

"Also, if I remember well, Carlsen had a even better score, in this period, against the other "absolute top-players""

Quite possible. I'm not at all disputing the fact that Vishy was a deserved World #2 at that time. But I do think the others were extremely close, and, in terms of how they might have done in a match vs. Magnus, specifically, probably even ahead of him.

Petrarlsen 9/1/2017 02:45

@ imdvb_8793 (2/2) :

Just one detail (because I don't think it is really a point on which we disagree...) : about Carlsen vs. Anand in 2010 - 2011, my meaning was only to say that, at the beginning of Carlsen's period as an "absolute top-player", he had difficulties playing Anand. (It's true, by the way, that I didn't take into account Rapid and Blitz games, but, whether I'm right or wrong is, I think, rather a question of "tastes", but, in general, when considering the playing level of a player, if the discussion isn't specifically about Blitz or Rapid chess, I use essentially Classical chess results, as, in my opinion, Classical, Rapid, and Blitz chess are "quite different things", even if I nonetheless find Blitz and Rapid very interesting forms of chess.) And, after this, Carlsen began to beat regularly ALL the other "absolute top-players" (including Anand) without losing himself any games, until 2013 or 2014 (I remember that the moment when this "golden period" ceased was between the World Championship 2013 and the World Championship 2014 : Carlsen lost a game against Caruana - just a little time after Caruana climbed for the first time above 2800 -, and, from then on, his period of invincibility against the other "absolute top-players" ceased for good). Also, if I remember well, Carlsen had a even better score, in this period, against the other "absolute top-players" (...the criteria that I used, for the definition of an "absolute top-player", was a player who, either was a 2800+ GM - Aronian -, or a 2750+ GM who, cumulatively, was once above 2800, and had been either World Champion - Kramnik and Anand - or World n° 1 - Topalov...), Aronian, Kramnik, and Topalov, than against Anand himself.

So, to sum up all this, in my opinion, at the beginning of Carlsen's period as an "absolute top-player", Carlsen had some problems with Anand, but, after, he "find the recipe" to play against the other "absolute top-players", and started to have very lopsided scores against them, but rather a little less against Anand than against the other "absolute top-players" (the worse being, if I remember well, Topalov, who, in this period - 2011 to 2013, approximately -, had an absolutely catastrophic score against Carlsen).

Petrarlsen 9/1/2017 02:41

@ imdvb_8793 (1/2) :

Yes, I agree that this exchange of arguments could be endless, and that it isn't perhaps completely necessary to continue indefinitely in this direction !

In fact, I think, reflecting more generally on our "debate", that the crux of our differences of view is our differing conception of a "peak period".

You rather consider that the best is to define with precision, moment by moment, the exact moment when a player plays his best chess ; I consider that it is the period when a player isn't making anymore significant progress, and hasn't yet begun to decline significantly. And I think that our differing conclusions are the consequence of these differing conceptions of a "peak period".

(Approximately, this is how I see things : for a painter, composer, sculptor, writer, architect, engineer, mathematician, etc., in my opinion, everyone would consider sufficient to find the period between the moment when the person really fully mastered his field and the moment when this person will began to decline significantly, and I reason in the same way for chess.)

And I rather reason in the same way, for example, for the 2014 Candidates tournament : Anand's opponents where the "right" opponents (i.e. at the right level of strength), and I don't think necessary to enter in all the details of their respective form at the moment of the tournament (I would nonetheless admit such an approach for a match, because ONE isolated player - as a match opponent - can really be completely off form, because of a transitory illness, for example).

I think that these differences of approach rather explain why, founding our argumentations on well-asserted facts, our conclusions are nonetheless different ; but perhaps you will disagree with this also !!

imdvb_8793 8/31/2017 12:24

PART 3:

"Don't forget that, these last years, Carlsen lost MUCH more points than Anand (even when taking into account Anand's present level), and this can't be the result of his age !! For now, Anand seems to have lost some of his level (taking into account his present level), but is it really caused by his age ??"

True, we can't be sure. It is, however, the most logical explanation, in his case. In Carlsen's case, there are probably others, possibly ones we aren't privy to - anyway, it's not important what the explanation is, because the fluctuations themselves are, in my opinion, what proves that a player is past (or not yet at) his absolute peak strength.

"I wouldn't completely exclude to see at a point one more time Anand regaining the World n° 2 place"

I would bet heavy money it won't happen. I don't think it's impossible, logically speaking, and I doubt I can prove it in any way, but intuition (which is based on having seen patterns like this before in history) tells me it's 100% not going to happen. I could be wrong. :) He would need to find some ridiculous amounts of motivation somewhere. Who knows?!...

A quick point about the theory that all of Anand's 2014 Candidates opponents might have been off form - it doesn't actually have to be all of them. It's enough that the 2-3 of them (possibly fewer, but it's never more, even in such a strong tournament) that had the actual potential to challenge for 1st in the first place are off form (and they didn't even need to be TOO off form - when the favorites in a tournament are close enough to one another in level, simply average form is enough to prevent any one of them from winning), and then all of the weaker ones in the field can be on top form, they still won't win it - they weren't good enough to begin with. Kramnik, for example, clearly still had the potential to win a Candidates Tournament, and was definitely off form compared to, say, 2016. Aronian was also off form. Topalov had already begun his big decline, and Karjakin wasn't yet one of the people that had the potential to win, in my opinion. So, there were probably exactly three that did - Kramnik, Aronian, and, as it turns out, Vishy, and the first two were clearly off form (with Aronian, if Garry is to be believed - and I do tend to believe him, though I hope I'm wrong, because I'm a bit of an Aronian fan -, it might have even been something more than that, in fact, but we can't be sure, so we have to assume he did have the potential to win it), while Vishy was on form, and very much so. Hence, he was the clear winner. That's my amendment to Resistance's theory. :)

imdvb_8793 8/31/2017 12:23

PART 2:

"About what I mean, when I use some expressions as "peak period", etc. : my meaning is a period when a player has already attained his approximate best level, and hasn't either began to decline. It is quite true that a player's results are not the same from one month to another, but, widely speaking, in their "best years", players keep continuously (...as always, it is possible to find examples of the opposite, but, in my opinion, they are quite rare, and usually explained by some well-defined reason...) a "good" playing level, so I don't go into all the details of each month's results, for each player ; I don't think this is necessary..."

I understand. Indeed, I myself define a player's peak as simply the moment when they played their absolute best chess. As in, they had no distractions, no bad luck, their form was perfect, their mental state was perfect, etc. Like Fischer during his interzonal-candidates-title match run, or, let's say, more generally, in 1970-1972 (not, for example, from 1967 to 1972 - that's another sort of peak for him, and he was clearly very close to his absolute best in patches between 1967-1969 as well, but moments of poorer form also existed, he wasn't consistent), or Kasparov when he won both Wijk and Linares with huge scores (it was either 2000 or 2001, I believe - not only then, but that's when his domination was most impressive.) Fabi's 7/7 in one tournament is a little short to be a "peak" - it should probably extend to more than one tournament. The main point is that it's when they were playing their absolute best chess vs. the strongest possible opposition at the time. There can be several such comparable "peaks" for players throughout their careers, but they shouldn't, in my view, include the lesser results in-between, in my understanding of "peak". I'm not sure whether an official definition exists, and, if it does, I'll adjust my arguments accordingly. :) But, for now, that's what I understand by that term. (Which is, of course, as you point out, quite different from your definition.)

"And, personally, I consider that the level of player is more defined by his high points than by his low points..."

I concur, but the low points are important in order to be able to define the player's "range" of potential results in that period, let's call it, and Anand's range was, I believe (I could be wrong), quite a bit wider in 2013-15 than in 2007-08, for example. Meaning he was proven capable of weaker results in the latter stage of his career. And we have a lot of reasons to think the Carlsen match could have been a moment of poor form. (The reasons I gave in an earlier post, and maybe others that didn't come to my mind at the time.) Instead of just exceedingly great play by Carlsen. When a player (in this case, Anand) isn't consistent, it's hard to be sure which "version" showed up for this or that tournament/match. Form does vary WILDLY, in chess, as in many other sports (not in all of them) - look at Wesley So's 2017 compared to his 2016! No particular reason for it. Also, Carlsen in 2016-17 compared to before... The examples from history are also many, they're not the exception. The absolute best players in history, at their peaks (as I define them) did not have such huge fluctuations. Usually, not even if you define their peaks as longer periods of time, the way you prefer to do it. Anand in 2013-15 did have such fluctuations - or, if not, at least started having them again immediately after, whereas (again, unless memory fails me once more) in his heyday he didn't. Not quite as big.

imdvb_8793 8/31/2017 12:22

PART 1:

OK, I checked again. So, in 2010, Anand did beat Carlsen twice (to nil, plus, of course, draws) in classical. Carlsen was heavily dominating their rapid and blitz games at the same time. Carlsen had defeated him in 2009, in Linares, and was clearly over 2800 in 2010, indeed. However, after that, from 2011 to the date of their first match, it was the opposite - Carlsen won two, with many draws, and Anand didn't win any games vs. him (again, in classical.) So, yes, I suppose that's not heavy domination, but it is domination. Which, obviously, got much worse for the second match, due to the 3-0 in the first one. And the second match is really the one I'm most arguing was too easy a task for Carlsen (that a player that wasn't 0-5 in classical in the last 3 years vs. Carlsen, and that, of course, had the same potential as Anand, would've given him a tougher battle, losing by, at most, one point.) Probably the first one, too, to a lesser extent - peak Anand wouldn't have lost by quite 3-0. But, who knows?! Maybe I'm too optimistic about that. But the second match I'll stand firm on... :)

"So Anand could be as low as World n° 5 in 2008 ; on the opposite, he obtained the World n° 2 place in July 2015 : for me, this shows that he was at the same approximate level in these two periods"

Yes, but 8 points off 1st place. :) Which is, of course, very, very different. You see, this is where inflation becomes crucial to my argument: I'm saying a lot of the top players were older and declining in 2013-15, and the new generation were still too young to set Magnus the kind of problems they're setting him nowadays, back then, when he was dominating so heavily. So, in my opinion, Anand's 2nd place then, which coincided with Carlsen's peak (so far), hence the big difference compared to 1st, unlike in 2008, is due at least in part to this favorable context. This is the part of my argument that is dependent on inflation. (Possibly - I'm not entirely sure it's not possible to make the same argument even without using inflation. So I won't claim you can, but I do feel it might be possible.) If you don't believe in the existence of inflation, you'll point to the cold numbers and say this decline in the opposition (for both Carlsen and Anand) isn't real. If you do, you'll point to the same numbers, subtract inflation, and say that it is, like I do. :)

"as we have just seen, even in 2007 - 2008, Anand could fluctate from the World n° 1 place to the World n° 5 place"

Sure, but, again, small differences at the top back then (unlike in 2013-15), and much smaller fluctuations, too.

Petrarlsen 8/31/2017 12:08

@ Resistance :

- For me, the only things that really count, in the end, in chess, are results. If you play "beautiful" games, and systematically lose... there is a problem somewhere. And if a player wins, and wins, and wins... he MUST necessarily play good chess, even if you don't like his games (...one example is Carlsen ; some persons don't like his games, and say he is overrated, etc., because of this, but he wins ! and, for me, this shows that, necessarily, he does indeed play good games, generally speaking...).

As I already explained elsewhere on this page, I think that Anand adapted his style to his age. So, perhaps, for some persons, his games are now less "palatable" than before, but the results are (or, at least, where, until 2015) there, and, for me, there is only this that really counts...

And what do you make, for example, of his Tata Steel 2013 Black victory against Aronian ? Even 2800+ GMs don't play every day such spectacular and dominating games against absolute top-level opposition...

- As for your examples (Wijk aan Zee 1970 and 2015 ; Linares 1985 and 1999), I don't find them convincing, because you don't demonstrate that these tournament victories where REALLY comparable (as I did for the comparison between the World Championship tournament 2007 and the Candidates tournament 2014) ; winning "isn't the only thing" : Was the respective lineups of these tournaments comparable ? By which margin did the involved players won these tournament ? With the tiebreaks ? with a 0.5 points margin ? With a 1 point margin ? (etc.) Did the involved players won the tournament the last day ? One day in advance ? Two days in advance ? For example a last-day victory on tiebreaks hasn't at all the same meaning as a one day in advance victory with a full-point margin (as, for example, Anand's victory in the Candidates 2014) : the first is indeed a victory, but by the narrowest of margins ; the second expresses a certain degree of domination : the difference between those two examples of results is quite important.

- "And I never said that the players from the Candidates 2014 were in bad form: I said that none of them was regular enough to get a better final score than Anand."

The nuance is really more than tenuous, in my opinion : for me, if you say, about the 7 other players of the 2014 Candidates tournament, that "none of them was regular enough to get a better final score than Anand", this means that you consider that they were in bad form : if you have irregular results, it necessarily means that you are not in good form. Or how could you say about a player that he is in quite good form, but have currently very irregular results ?? Regularity is also a component of form !...

Petrarlsen 8/31/2017 12:07

@ imdvb_8793 (2/2) :

- About what I mean, when I use some expressions as "peak period", etc. : my meaning is a period when a player has already attained his approximate best level, and hasn't either began to decline. It is quite true that a player's results are not the same from one month to another, but, widely speaking, in their "best years", players keep continuously (...as always, it is possible to find examples of the opposite, but, in my opinion, they are quite rare, and usually explained by some well-defined reason...) a "good" playing level, so I don't go into all the details of each month's results, for each player ; I don't think this is necessary...

- "(...) the fact that he was fluctuating between 8th and 2nd (and, as early as January 2016, within less than two years, he was back down to 8th, which confirms my point) also shows he was inconsistent. (...)" : as we have just seen, even in 2007 - 2008, Anand could fluctate from the World n° 1 place to the World n° 5 place - as I said, he never was an absolute "Elo King" as Kasparov or Carlsen could be. So, perhaps he fluctuated a little more still, in 2013 - 2015, but I don't think this makes a big difference... And, personally, I consider that the level of player is more defined by his high points than by his low points...

- "To me it looks pretty clear which player was actually feeling the effects of age/loss of playing strength - quite logically, it was the older one, Vishy."

Don't forget that, these last years, Carlsen lost MUCH more points than Anand (even when taking into account Anand's present level), and this can't be the result of his age !! For now, Anand seems to have lost some of his level (taking into account his present level), but is it really caused by his age ?? One more time, Carlsen is losing much more points, and obviously not because of his age... As for me, I wouldn't completely exclude to see at a point one more time Anand regaining the World n° 2 place (...or even the World n° 1 place, if Carlsen continues to shed Elo points by full truckloads !). I don't say at all that this is probable, but, in my opinion, this isn't completely impossible either...

- "Unlike Carlsen and Anand's scores before their first and second matches, which heavily favored Carlsen (after the initial difficulties before he became a top player.)"

Not completely exact : I don't remember the precise numbers, but, approximately from 2010 to 2011 (perhaps from May 2010 to May 2011 ??), this while Carlsen was already permanently above 2800, I remember that, against Anand, he had really very bad results (something like 2 defeats for 0 victories, with 4 draws). I don't remember the exact numbers, but what I DO indeed remember is that his results against Anand at this time where really very surprisingly bad - and he was already above 2800 (and, if I remember well, even World n° 1 most of this period - I think Anand regained the World n° 1 place in this period, but only for a few months).

...The end for now !... (...and I will now answer Resistance...)

Petrarlsen 8/31/2017 12:06

@ imdvb_8793 (1/2) :

- "(...) you're one of the best people to have such a conversation with, at least here, on Chessbase (...)" Thanks ! It's reciprocal !

- "(...) and might soon start repeating ourselves (...)" This is indeed the "danger" ! I must say I always tend to answer (!), when I have something to answer... but you are perfectly right ; it isn't useful, if there isn't any new argument ! So, being what I am, I will still answer (!!), but I will also try not to restate always the same now well-known arguments, and to limit myself to more-or-less (...hopefully more on the "more" side than on the "less" side...) "new elements"...

- About the inflation and Anand's "new Elo peak", in 2015 : I think that in the exact way that I stated my arguments, inflation has indeed nothing to do with this, because my meaning was only to answer what you said : that Anand's victory in the Candidates was not a "one-shot", but was confirmed but a completely different sort of "victory" : his subsequent regaining of the World n° 2 place in the overall World Rankings.

It is quite true that there are quite different ways of being World n° 2, and that, in July 2015, Anand absolutely didn't DOMINATE ; he only "won it" ; he succeeded in "winning" the World n° 2 place (one example that comes to my mind, representing the opposite - a real domination : in March 2014 - for the Candidates Tournament -, Aronian had a 43 points advantage over the World n° 3, Kramnik : this is a very important advantage ; quite obviously, this was not at all the case for Anand in July 2015). But, for my demonstration, this is sufficient ; it demonstrates that the Candidates didn't represent an isolated success, and, furthermore, a World n° 2 place, for Anand (who never was an absolute "Elo King", as Kasparov or Carlsen can or could be, when they are in-form, for example) is quite compatible with a "peak-level" form : for example, in October 2008 - in a overall globally indisputably "good period" -, Anand was "only" World n° 5 behind Topalov, Morozevich, Ivanchuk, and - already - Carlsen. So Anand could be as low as World n° 5 in 2008 ; on the opposite, he obtained the World n° 2 place in July 2015 : for me, this shows that he was at the same approximate level in these two periods (his "high point", in 2015, being clearly above one of his "low points" from a period when no-one, in my opinion, could say that Anand had began to decline).

Resistance 8/30/2017 02:30

@ Petrarlsen ---

Your argument that Anand 2007 is of the same level than Anand 2014 is not as strong as you presume, since it rests mostly on the nominal aspect of the issue being addressed here (--i.e., it rests mostly on numbers: Elo, world rankings, final standings--), while leaving aside the key element of chess properly. Theoretically speaking (--i.e., without taking into consideration the games themselves--), it is possible that Anand 2007 played at the same level of Anand 2014. However, if that was the case, numbers would not be the ultimate criteria upon which to establish the truth of such a proposition, but rather the games themselves.

That Mark Taimanov won Wijk aa Zee 1970, and Magnus Carlsen Wijk 2015, doesn't necessarily mean that they're both players "of the same level", though they both won the same prestigious tournament (Wijk aan Zee); that Ljubomir Ljubojevic won Linares 1985, and Garry Kasparov Linares 1999, doesn't necessarily mean that they're both players "of the same level", though they both won the same prestigious tournament (Linares). Likewise, that there was a number 2 player in the world in 2007 and 2014 doesn't necessarily mean that both those players played at the same level, though they were both number 2 in the world.

Your argument rests on the assumption that Anand 2007 and Anand 2014 played at the same level, because both Anands won their respective Candidates Tournaments; that just simply does not (necessarily) follow. If it is the case that Anand played chess at the same level in 2007 and 2014, it is because he PLAYED CHESS at the same level in 2007 and 2014, not because he won both tournaments (--which is not the same thing--). If you take a look at the games, and not merely at the numbers, you'll see that the Anand from 2007 is not the same as the one from 2014, though in both cases he won the corresponding Candidates Tournaments (--up to 2014, let me remind you, Anand had won just one (!) -that's right-, just one Classical Chess tournament since Linares 2008! (Grenke, 2013)--). Magnus Carlsen DID NOT beat the same Anand that won in Mexico, in 2007, and he knows it, just as anyone who followed both those Classical World Chess Champioship Matches (2013 & 2014).

(And I never said that the players from the Candidates 2014 were in bad form: I said that none of them was regular enough to get a better final score than Anand.)

imdvb_8793 8/30/2017 09:59

Also, Kasparov and Kramnik's score in the lead up to the 2000 match was incredibly balanced, with many, many draws and wins/losses going back and forth an unbelievable number of times, nobody really managing more than 1-2 in a row. Unlike Carlsen and Anand's scores before their first and second matches, which heavily favored Carlsen (after the initial difficulties before he became a top player.) So you can't say Garry had anywhere near as much reason to "fear" Kramnik, plus he was about 7 years younger than Vishy when the latter lost his title. Maybe lack of motivation could be brought into the discussion, but, really, in Garry's case, I think saying something like that is probably a little naive. :) But, who knows, deep down, and to some extent... I could maybe accept this last one as sort of possible. In any case, there are way more reasons (that I can see) to suspect Anand of not playing at his best (or, perhaps, even anywhere near it) vs. Carlsen than Kasparov vs. Kramnik. (I know this is more relating to my discussion with maxharmonist, of course.)

I told you the arguments were endless!... :) I think both of our cases are fairly convincing - meaning our logic is sound, as much as it can be for such an unclear situation -, but not convincing enough that we can't rather easily find equally convincing counter-arguments. Which is the way things should be, really, when there's no clear, irrefutable evidence one way or the other. :) So, yeah, I guess we should probably stop at some point, or else we're liable to never finish this discussion... Not that I didn't enjoy it, of course! I like being challenged to actually express my arguments and form a coherent case to support my opinions! And you're one of the best people to have such a conversation with, at least here, on Chessbase - you actually take the time to support your opinions properly, give well thought out answers, and never degenerate into personal insults and such, like (as you and I well know) far too many people do online. I enjoy our discussions very much! But we're probably not going to get much further from here, as I feel we've most likely already exhausted the most compelling arguments for either side, and might soon start repeating ourselves, which isn't good for anybody. :) I'm not just spitballing - I've seen this happen in my (and other people's) discussions before. It's not a tragedy, of course, but it's not very productive, either. :D

imdvb_8793 8/30/2017 09:59

"And a 2nd place in the World Rankings means exactly the same thing in 2014, in 2007, or in 2010 : that in the "Elo Rankings Competition" you have managed to "beat" everyone except the World n° 1"

Agreed. However, due to inflation, it might not mean you're as strong of a World #2 as you were in 2007 or 2010. That's what inflation does - it makes it impossible to say a (for example) 2790-rated player in year n is as good as a 2790-rated player in year m, or that a #2 in year n is as strong/meaningful as a #2 in year m. Again, if you believe inflation exists.

"his regaining of the World n° 2 place proved that he could still raise his playing level above the level of everyone else on the planet besides Carlsen."

Correct, but, again, it doesn't prove he could raise his level compared to 2007 or 2010, but only that he was able to raise it compared to his low point(s) in-between 2010 and 2014-2015. (Because, of course, inflation isn't THAT fast...)

"And I still don't see at all what inflation has to do with that !!"

Asked and answered! :) A lot, I'd say. To me, it might be the crux of the problem.

"So, I think that it is directly possible to conclude that Kasparov necessarily considered that, at 32, he was himself in his own "optimal age zone"..."

I understand your reasoning, and it's sound, but the problem is being in the optimal age zone doesn't mean you'll also be at your peak throughout it. You'll play better some years, and worse other years, even within that age interval. Otherwise, people's Elo's would just stay the same for years, as long as they were at the optimal ages. They almost never do. They fluctuate quite a bit. So, even if he was at the optimal age at 32, just as much as at 36, that's just not evidence that he was also playing at his absolute best. His results and ratings in 1995 compared to 2000-2001, in fact, indicate he very likely wasn't.

"2) For Anand, at the moment of the 1995 match, he was World n° 4 ; in 2014, at the moment of his win in the Candidates Tournament, he was World n° 8, and afterwards, as we know, he climbed back to the World n° 2 place. So, his 1995 World n° 4 place seems rather to fit in the picture of an Anand more or less at his best level : such a place seems just "normal", average, for him, in a good period... "

Sure, but the fact that he was fluctuating between 8th and 2nd (and, as early as January 2016, within less than two years, he was back down to 8th, which confirms my point) also shows he was inconsistent. Which means he could have EASILY been in bad form precisely in the match(es) vs. Carlsen. (For whatever reason - "fear", meaning lack of confidence vs. that particular opponent, or tiredness due to age, or lack of motivation, etc.) Which would have made it easier for Carlsen. This argument holds less water, for instance, in the case of 2000 Kasparov - you can't really say it's likely he was in bad form vs. Kramnik, because he was the clear #1 and would continue to be, and his rating didn't fluctuate as much even taking the rating loss in the actual match into account. He soon started gaining points again (which is incredibly hard at the 2800+ level, as we know, especially if you're the only one there), and didn't really start losing points again until 2003-2004, more than two years later. By comparison, Anand was back to where he was before the second Carlsen match within a year, and then immediately started dropping quite a bit. Plus, as you also pointed out, he wasn't even anywhere near 2nd in the world before, unlike Garry, who was clear first well before and well after his 2000 defeat. To me it looks pretty clear which player was actually feeling the effects of age/loss of playing strength - quite logically, it was the older one, Vishy.

Petrarlsen 8/28/2017 10:34

@ imdvb_8793 :

A little complementary detail : when I note that there was a 7 months "plateau period" after the Candidates, and that Anand began to climb up the rating scale only after this period, what I mean is that, if he obtained once more the World n° 2 place, it had nothing to do with the Candidates, because he only began this "climb" 7 months after the Candidates... So these two elements (his victory in the Candidates and his regaining of the World n° 2 place) are completely independent, and, in my opinion, taken together, they show convincingly Anand's very high level in this period...

Petrarlsen 8/28/2017 10:25

@ imdvb_8793 :

- "Anyway, the (obvious, for someone who believes in that) explanation for why, in spite of the rating gains and such, I still don't think he was quite as dangerous as a very in-form 2006 Topalov is inflation - that's a whopping 8 years of inflation..."

With this, I strongly disagree ! Inflation cannot have anything to do with this ! My meaning was that, after the Candidates, Anand had a 7 months "plateau period", followed by a 9 months period of constant progression, culminating in Anand becoming once more the World n° 2 in the world rankings. Here, there is no comparison with the past : you have a period of progression - the past have nothing to do with that - and a 2nd place in the World Rankings. And a 2nd place in the World Rankings means exactly the same thing in 2014, in 2007, or in 2010 : that in the "Elo Rankings Competition" you have managed to "beat" everyone except the World n° 1, which means, for Anand, that he "beat" EVERYONE ELSE besides Carlsen at this moment, in terms of Elo rating, including all the "old stars", all the "young and brillant GMs", etc., etc. This is indeed a very fitting complement to his convincing victory in the Candidates : the Candidates demonstrated that, in a competition, Anand could directly beat anyone, when Carlsen wasn't participating ; his regaining of the World n° 2 place proved that he could still raise his playing level above the level of everyone else on the planet besides Carlsen. And I conclude that Anand potential, in this 2013 - 2014 (...in view of his 2015 World n° 2 place, I think it could be also possible to even include 2015...) period was intact !

And I still don't see at all what inflation has to do with that !!

- About Kasparov's and Anand's respective levels in 1995 :

1) For Kasparov, my reasoning is that, if he considered in 2004 that Gligorić's 33-36 "optimal age zone" had to be reevaluated in the direction of a younger "age zone", he couldn't forget his own situation ! And, necessarily, if Kasparov considered that 33-36 was "too old" for the "optimal age zone", in meant that he considered that 32 was either included in his own "optimal age zone", or even after it. So, I think that it is directly possible to conclude that Kasparov necessarily considered that, at 32, he was himself in his own "optimal age zone"... (Or it would mean that he "forget" his own situation, in his reflexions on the "optimal age zone", and this seems, in my opinion, a little hard to swallow !)

2) For Anand, at the moment of the 1995 match, he was World n° 4 ; in 2014, at the moment of his win in the Candidates Tournament, he was World n° 8, and afterwards, as we know, he climbed back to the World n° 2 place. So, his 1995 World n° 4 place seems rather to fit in the picture of an Anand more or less at his best level : such a place seems just "normal", average, for him, in a good period...

- About the comparison between Topalov in 2006 and Anand in 2013 - 2014 : In fact, I think that this question depends nearly exclusively on the difference in level, for Anand, from 2007 to 2014. (If I remember well, you didn't discussed the fact that Topalov was approximately at the same level in 2006 and in 2010.) And, as we discussed in lengths this question in other posts, this depends directly on whether our respective arguments on this theme are convincing or not, in my opinion...

imdvb_8793 8/28/2017 06:57

Besides, I've picked two results for both, not one - the wins vs. Kasparov and Topalov for Kramnik, and the wins vs. Anand for Carlsen. (We can make it three, if you want, by including Kramnik-Leko and Carlsen-Karjakin. Doesn't change anything. In fact, it might favor Kramnik, as Leko was probably closer to the very top in 2004 than Karjakin in 2016.) And I picked them fairly, I didn't exclude any of the match results I left out without reason. I gave solid arguments for why both Kramnik and Carlsen were far from their peak strength in all of those encounters.

imdvb_8793 8/28/2017 06:55

@maxharmonist: "The problem with picking one result in one event against one opponent to rank players is that it simply doesn't say much."

Is comparing players' results when they were clearly significantly weaker than at their peak (despite still being very strong, of course, but that's beside the point), and using them as proof of how they might have done against each other at their actual peaks, just for the sake of having a bigger sample, a better option, though? I think not - I think it's much worse, and renders the whole exercise irrelevant.

imdvb_8793 8/28/2017 06:54

"For me, it isn't really possible to compare a 6 - 1 in 21 games against Anand with a 2 - 0 in 15 games against Kasparov. Kasparov was, in my opinion, stronger than Anand, but a 6 - 1 in 21 games is also better than a 2 - 0 in 16 games (to lose 6 games, in such a limited amount of games, even with one win to compensate a little, in my opinion, is really a complete rout ; a 2 - 0 in 15 games is a clear defeat, but not a complete rout). So, in my opinion, these two results cannot really be compared."

Definitely disagree, like I said. :)

"I don't remember exactly Anand's results in this period, but, obviously, it isn't possible to become the World n° 2 with a 2816 rating while losing games after games ! "

OK, I checked with Mega Database. You're right, Anand did have a strong period in 2014-2015 - inconsistent, but strong, overall, with even a couple of classical tournament wins, and not a lot of defeats. But he definitely wasn't dominant - he was finishing on +2, +3 at best, and sometimes on minus scores, of course. Anyway, the (obvious, for someone who believes in that) explanation for why, in spite of the rating gains and such, I still don't think he was quite as dangerous as a very in-form 2006 Topalov is inflation - that's a whopping 8 years of inflation... (Though, with Anand's extra match experience... but, then again, the bad score vs. Carlsen... it all evens out, I'd say. And, so, it all remains very unclear, and we're still forced to decide subjectively. The arguments for and against one or the other are probably endless...)

"It is possible that Kasparov wasn't completely at his peak level in 1995, but, for the reasons that I stated before, I can't see how he could be very far from it ; in is also possible that Anand wasn't completely at his peak, but, at this level of domination, I don't think that a slight difference in level for any of these two players would have changed much the results... "

That's already two assumptions we're making in the same direction (meaning, to favor peak Carlsen over peak Kasparov), neither of which seems particularly justified from an objective point of view! :) These may or may not be small differences. You think they are. I think it's way unclear, and makes a comparison impossible (from this point of view,) because they're AT LEAST significant enough, logically speaking, to make possible very big differences in terms of actual match results/scores.

"I think that if it isn't possible to prove anything, it probably mean that there is not a considerable difference between the two"

I definitely don't think that's the case. There's no correlation there. (I'm sure there are far better analogies there - I just can't think of one right now, and I'm too lazy/tired to take enough time to solve that issue. :D) That's like saying if a strong engine can't see a significant difference in evaluation between two possible moves in a non-tablebase ending, it's impossible for one to be winning, and for the other to be losing. Clearly, that's not the case, since engines (like people) misevaluate difficult endings quite often.

"I think that in 2013 - 2014, Carlsen had a very high level of confidence."

Agreed! And it's true Kasparov never ever had such poor years as Carlsen has, to shake said confidence, so it's hard to compare, I suppose.

maxharmonist 8/28/2017 02:01

"Think what match results Garry had before losing to Kramnik!"

The problem with picking one result in one event against one opponent to rank players is that it simply doesn't say much. If you want to give Euwe his proper place you can't just say that he was the only one to beat Alekhine in a match, you also have to look at his (lost) matches against Capablanca, Keres, et al, as well as his tournament results. The one result in 1935 isn't enough to say that since it was a much better result than Capa's in the match against Alekhine, Euwe was a better chess player.

I think Kramnik is best evaluated as chess player compared to Carlsen, Anand etc, by looking at all his results and not just the one against Kasparov or the one against Shirov two years before. Kramnik has been a very strong player for 25 years, but he is no Carlsen (i.e. clearly best player in the world for a long time) and has never been. I wouldn't say he is far behind Anand, but the latter has beaten Kramnik in all World Championships and Candidates both played, was #1 for much longer time and contrary to Kramnik also sole #1, had a higher rating etc.

Petrarlsen 8/28/2017 11:42

@ imdvb_8793 (2/2) :

- "(...) even if you take into account the Candidates win, which was his only major win I can remember at the time - you can't keep using it as the one piece of evidence that Anand was still a top 2 player; it could be a bit of a fluke, as in his being very in-form and motivated in that tournament, but not anywhere else, including the matches vs. Carlsen. You know the old greats always have one such last result that reminds everyone how great they were, even though they can't really compete with the very best anymore, overall, which they prove again and again in other events!"

But such was precisely not at all the case for Anand ! Just after the Candidates Tournament, Anand had a 2785 rating, that he kept from April to October 2014, but he didn't stay there ! He regained Elo points continuously for the next 9 months until July 2015 (this for a total gain of 31 points, between October 2014 and July 2015, and a final - in July 2015 - very strong 2816 rating, while becoming once more the World n° 2, behind Carlsen). I don't remember exactly Anand's results in this period, but, obviously, it isn't possible to become the World n° 2 with a 2816 rating while losing games after games !

So, putting together the Candidates Tournament result and Anand's subsequent very important gain in terms of Elo points, I maintain that Anand's level in 2013 - 2014 wasn't significantly worse than his level in 2007 !

- "And, besides, you can't really tell me 4-1 in one match, albeit a bit longer is THAT much worse than 6-1 over two shorter matches." It wasn't what I meant. In my opinion, these two results are approximately comparable, and they place Kasparov and Carlsen in the same category, the category of champions who can, at their peak, crush completely and utterly even "normal" World Champions playing at their approximate peak level. About Kasparov's games against Anand after their 1995 match, I also agree with what you said in a former comment : "(...) this 3-11-0 score is still pretty much the same as than Carlsen's 6-14-1 in the two matches against Anand.". My point is just that I don't think that, at his peak, Kasparov was significantly stronger that Carlsen at his own peak. It is possible that Kasparov wasn't completely at his peak level in 1995, but, for the reasons that I stated before, I can't see how he could be very far from it ; in is also possible that Anand wasn't completely at his peak, but, at this level of domination, I don't think that a slight difference in level for any of these two players would have changed much the results...

- "Again, it's very hard to find an objective argument that proves one or the other position here." (about the comparison between Anand 2013 - 2014 and Topalov 2006) I think that if it isn't possible to prove anything, it probably mean that there is not a considerable difference between the two (which doesn't mean that a slight difference isn't possible...) ; otherwise, I think that it would probably be possible to attain a (at least relatively) proven result...

- "He just doesn't seem to have anywhere near the same drive, or even the same confidence, as Garry." I think that in 2013 - 2014, Carlsen had a very high level of confidence. In my opinion, it is exactly, as you said, the "drive" that makes the difference. Kasparov has the enormous drive that can permit a player to stay at the absolute top-level for years and years. I don't think that this is the case for Carlsen, and this is probably why he is now rather plummeting down the ratings scale...

Petrarlsen 8/28/2017 11:41

@ imdvb_8793 (1/2) :

- "Again, 6-1 vs. the same player (...) is, to me, clearly less impressive than 2-0 vs. Kasparov at his peak rating and +1 vs. Topalov also at his peak."

For me, it isn't really possible to compare a 6 - 1 in 21 games against Anand with a 2 - 0 in 15 games against Kasparov. Kasparov was, in my opinion, stronger than Anand, but a 6 - 1 in 21 games is also better than a 2 - 0 in 16 games (to lose 6 games, in such a limited amount of games, even with one win to compensate a little, in my opinion, is really a complete rout ; a 2 - 0 in 15 games is a clear defeat, but not a complete rout). So, in my opinion, these two results cannot really be compared.

As for Topalov, I still maintain that Anand's level in 2013 - 2014 was comparable to Topalov's level in 2006. And, even if he was a little weaker (in my opinion, it was the opposite, but, obviously, we cannot determine this with an absolute precision), I don't see how a "simple" + 1 could become the equivalent of Carlsen's very dominant 6 - 1 against Anand. But, with the "Toiletgate", the match results were quite possibly warped, and perhaps, if the match had followed a normal course, the score would have been significantly more in Kramnik favor. But we will never know exactly what could have happened in normal circumstances...

- "(...) the Carlsen-Anand matches, which I found very disappointing to watch, overall (...)" This is obviously a question of tastes, but, as for me, I found these matches extremely interesting, esssentially as a demonstration of Carlsen's style and mastery (in particular Carlsen's first victory in the first match, which I found particularly impressive).

imdvb_8793 8/28/2017 10:05

PART 3:

"So I don't think that it is obvious at all that Anand's resistance was weaker in 2013 - 2014 than Topalov's resistance in 2006."

You do make an interesting case here... Still, Anand in 2010 BARELY beat Topalov - such margins are, again, not very relevant. And, like I said before, unlike you, I don't think Anand 2013/14 was as strong as Anand 2010. (Partially because I believe in inflation, but also because of his age, his - for obvious reasons - lesser motivation levels, his "fear of Carlsen" due to their recent results, etc.) ESPECIALLY in match play. Again, it's very hard to find an objective argument that proves one or the other position here. :)

"I don't see at all (perhaps I will be wrong ; we'll see in the following years...) Carlsen dominating chess over a very long period, like Kasparov. I think that, intrinsically, Carlsen was an incredibly gifted player (at Kasparov's level, more or less...), but hadn't the mental force to maintain his domination over a long period."

On this one we agree pretty much completely. I've been saying it from the beginning. But I have come to think he IS pretty strong, mentally. He just doesn't seem to have anywhere near the same drive, or even the same confidence, as Garry. Time will tell... I've been in very bad predicting form in recent years, so I tend to be pessimistic and think Carlsen will just win his next 10 title matches, just to spite me. (I'm not saying this is logical - in that respect, just looking at the trends and arguments and his level of play, I feel like he should lose his title very, very soon -, just that this is the gut feeling I'm getting.) That kind of thing has been happening to me a lot, of late. (Plus, I'm naturally pessimistic.)

imdvb_8793 8/28/2017 10:04

PART 2:

"And, if I understood well your own posts, you rather thought yourself that 1995 was not to far from Kasparov's peak, when you said, in one of your commentaries of this same page : "Let's say Carlsen today vs. Kasparov 1995, instead! That seems like a reasonable age." Or perhaps I didn't understand quite well your meaning..."

Again - being just a little off your top form can have very serious effects, depending on the circumstances, etc. - plus, again, not a statistically significant difference between the two scores... See above! :) I probably should've stuck to my original idea and just said 2001 or even 2000, to avoid such side discussions - so what if he lost to Kramnik that year! Doesn't mean Carlsen would've beaten him too, or even that Kramnik wouldn't have lost a second match, started, let's say 2-3 months after the first one...

"So, in my opinion, if we take the STRENGTH of Carlsen's domination over the rest of the chess word, it was comparable with Kasparov's domination (in 2013 - 2014, approximately)"

In any case, if we go by this measure, Fischer is still miles ahead of anybody else in history, with his 125-point gap. (I think Lasker also comes close, at the turn of the century - he was 100+ points ahead of Pillsbury in late 1897 and early 1898, at least looking at Chessmetrics. Of course, Morphy, too... Capablanca was never quite that far ahead of Alekhine and even Lasker, nor was Steinitz, compared to his contemporaries - nor anybody else. Obviously, we're not talking in terms of absolute playing strength here. Just domination over contemporaries. Although, personally, I remain of the opinion that Fischer was at least as good as Kasparov in absolute terms, and even engine analysis confirms this in some studies - like I said, they contradict each other... naturally, since they use different sets of engines and different testing parameters. Which is one of the reasons they shouldn't be trusted unless they consistently show some tremendous difference in favor of one player over the other. Which they don't, at least not in the case of Fischer and the other absolute best players of all time.)

From my (very quick and sloppy) calculations, Carlsen's biggest gap was 69 points (April 2014.) Kasparov's seems to have been 82 (January 2000.) Certainly comparable, but it's also hard to argue Carlsen's domination was heavier, since there is that 13-point difference in favor of Garry...

imdvb_8793 8/28/2017 10:03

PART 1:

"For me, it isn't the fact that Carlsen won the two matches that is impressive, but his overall 6 - 1 result in these matches."

Again, 6-1 vs. the same player, who, of course, after losing the first match to a younger player that had ALREADY been dominating him even before that match, was never going to be a serious

threat in the second one (unless Carlsen played much worse) is, to me, clearly less impressive than 2-0 vs. Kasparov at his peak rating and +1 vs. Topalov also at his peak. (And I felt the

same way at the time of the Carlsen-Anand matches, which I found very disappointing to watch, overall; I'm not just adhering to this point of view now, because it helps my argument.) I emphasize the "clearly" part. Think what match results Garry had before losing to Kramnik! And how heavily he dominated the tournament circuit in those years, as opposed to Anand's tournament results in 2013/2014 (even if you take into account the Candidates win, which was his only major win I can remember at the time - you can't keep using it as the one piece of evidence that Anand was still a top 2 player; it could be a bit of a fluke, as in his being very in-form and motivated in that tournament, but not anywhere else, including the matches vs. Carlsen. You know the old greats always have one such last result that reminds everyone how great they were, even though they can't really compete with the very best anymore, overall, which they prove again and again in other events! Chess history is full of such examples - Lasker in 1934-36, Smyslov in 1984, etc.) Of course, like I said all along, these are mostly subjective evaluations, and there's no fool-proof, objective way of proving either side of the argument right (because it's unclear whether the strength of the opposition, the circumstances, etc., were similar, or much tougher for one or the other - I tend to think Garry's opponents were more dangerous, but it's hard to make a strong case either way), so we'll probably just have to agree to disagree on this point. :)

"In view of Kasparov's opinion on this subject, I don't quite see how he could himself have been really far from his peak at the age of 32..."

Surely you see there can be arguments and counter-arguments here. I was just looking at the ratings, which are fixed and say Kasparov wasn't at his peak in 1995. How close or far... we're really getting into heavily speculative territory! :) I mean, non-peak is non-peak. Who knows how it can affect your level?! Would anybody have thought even non-peak Carlsen would struggle SO badly vs. Karjakin? (Well, me, but I don't have such a high opinion of peak Carlsen, so I'm not an appropriate example.) A slight decline from your peak level can lead to much worse results, sometimes, not just a little worse. And, besides, you can't really tell me 4-1 in one match, albeit a bit longer is THAT much worse than 6-1 over two shorter matches. One more win for Garry and you get about the same percentage. That can just be random variance. Plus, again, their score later on, at Kasparov's peak, was even more convincingly in favor of Garry (I've already given the numbers.) To me, even more convincing than Carlsen's 6-1. But, again, there's no "objective truth" here, or at least no way for us to know it.

Petrarlsen 8/28/2017 05:40

@ imdvb_8793 :

"The higher overall percentage is irrelevant, since non-peak Anand clearly put up less serious resistance than peak Topalov and peak Garry." Without taking part in the discussion on the respective levels of Carlsen and Kramnik, yes, I fully agree that Anand in 2013 - 2014 was certainly much easier to beat than Kasparov in 2000, but I'm not at all convinced about Topalov in 2006. Following my reasoning, as Anand was - according to my arguments - approximately at the same level from 2007 to 2014, as Topalov was nearly at the exact same Elo level for the 2006 match (2813) and for the 2010 match (2812), and as Anand beat Topalov rather convincingly in 2010, I would reason like this : Anand was approximately at the same level in 2013 - 2014 and in 2010. He beat Topalov in 2010. Topalov was approximately at the same level in 2010 and in 2006. So Anand beating Topalov in 2010 would be rather the equivalent of the Anand of 2013 - 2014 beating the Topalov of 2006. So I don't think that it is obvious at all that Anand's resistance was weaker in 2013 - 2014 than Topalov's resistance in 2006. I would rather think that, reasoning on all these basis, it would have been the opposite, even if it would probably be by a small margin. (The reason being, in my opinion, that Anand is a better match player than Topalov - Topalov is a very strong player, but I don't think that matches are really his strong point.)

Petrarlsen 8/28/2017 05:07

@ imdvb_8793 : "And, if anything, it probably favors Kasparov." As for me, I don't see any sufficently significant difference between them to be meaningful...

As for Kasparov being near his peak level in 1995, I think that, at the age of 32, he couldn't be too far from it, in particular when we consider what Kasparov himself explained on this subject in "My Great Predecessors, Part III" (in the chapter on Svetozar Gligorić) : Kasparov noted that "Gligorić, on the basis of his own experience, considered the optimal age of a chess player to be 33-36.". Kasparov then added : "But today, with the appearance of powerful computers and the Internet, chess is rapidly growing younger." And, a little further in this chapter : "By contrast, the brain of a player older that thirty, tired out by the constant effort, gradually gets rid of the information that is overfilling it and increasingly, in the interests of self-preservation, suddenly switches off at the most inappropiate moment..." So, obviously, Kasparov considered that the optimal age, for a chess player, was, at the time he wrote that book (in 2004, just before Kasparov's retirement from competitive chess), under Gligorić's "33-36 zone". In view of Kasparov's opinion on this subject, I don't quite see how he could himself have been really far from his peak at the age of 32...

And, if I understood well your own posts, you rather thought yourself that 1995 was not to far from Kasparov's peak, when you said, in one of your commentaries of this same page : "Let's say Carlsen today vs. Kasparov 1995, instead! That seems like a reasonable age." Or perhaps I didn't understand quite well your meaning...

About the length of Carlsen's peak period, I think that Kasparov and him are, psychologically speaking, very different players. I don't see at all (perhaps I will be wrong ; we'll see in the following years...) Carlsen dominating chess over a very long period, like Kasparov. I think that, intrinsically, Carlsen was an incredibly gifted player (at Kasparov's level, more or less...), but hadn't the mental force to maintain his domination over a long period.

So, in my opinion, if we take the STRENGTH of Carlsen's domination over the rest of the chess word, it was comparable with Kasparov's domination (in 2013 - 2014, approximately), but, for the LENGTH of their respective domination, I think that Carlsen, at the end of his career, will be nowhere near the length of Kasparov domination. And I think that these two elements are completely different, and both quite meaningful, but differently meaningful...

Petrarlsen 8/28/2017 04:24

An interesting comparison, in my opinion, between Anand's games in the 2014 Candidates Tournament, and his games against Carlsen in the 2013 and 2014 matches :

In Anand's 6 games against the top-three players in the world besides Carlsen, the World n° 2, Aronian (who had a stellar 2830 rating, at the time), the World n° 3, Kramnik (who was below 2800, but, as an ex-World Champion and the World n° 3 was nonetheless one of the indisputable top-players of the moment), and the World n° 4, Topalov (same thing as Kramnik : he was below 2800 at this moment, but as an ex-World n° 1 and World n° 4 at the time of this tournament, he was also an indisputable top-player, in my opinion), he won (globally) 2 games, drew 4 games, and didn't lost any.

In Anand's 21 games against Carlsen, he won 1 game, drew 14 games, and lost 6 games.

So, in the 21 games of the two World Championships, Anand only won 1 game, while in only 6 games against the top-three players besides Carlsen, he managed to won 2 games. This while losing a whole series of games against Carlsen (so it wasn't even that he was playing in a more defensive style). And against Aronian, Kramnik, and Topalov, he didn't lost any games, so it seems that his 2 victories against them weren't even at the cost of a more risky play ; rather the opposite, in fact, seemingly...

I think that these elements are rather self-explanatory about Carlsen's domination in 2013 and 2014 : while Anand could himself be really dominant against the other top-players, he was completely outplayed by Carlsen in their two matches (...one more time, an overall 6 - 1 is an incredibly crushing score...).

Petrarlsen 8/28/2017 01:54

@ imdvb_8793 : "Peak Carlsen has the two match wins vs. an old Anand. Which is fair, but not THAT impressive in my book."

For me, it isn't the fact that Carlsen won the two matches that is impressive, but his overall 6 - 1 result in these matches. Before these two matches, I would NEVER have thought that Anand, still playing at a very high level and with an ENORMOUS match experience against all types of opponents (Kramnik, Topalov, and Gelfand represented quite of range of different playing styles) would lose against Carlsen with such a lopsided score... Until these matches, I thought that Carlsen was a very good player, but that he would probably encounter serious difficulties in match play. These matches clearly demonstrated the opposite : that, even in match, Carlsen could really be a "beast", and completely crush even the best players on the planet... Which was for me a big surprise...

And, yes, Anand wasn't anymore a young player, but he won the Candidates 2014 (against, in particular, the World n° 2, World n° 3, and World n° 4 players), and was World n° 2 behind Carlsen once more from July to September 2015, so he proved that, at this time, he could still compete with complete success against the younger top-players...

Petrarlsen 8/28/2017 01:35

@ Pieces in Motion : A top-player returns to chess 12 years after his retirement, coming back more or less 15 years after the age when many top grandmasters stop playing in top-level competitions, gain a new 2801 rating in Blitz, thus re-entering directly into the "absolute elite" circle, in Blitz, and you consider that his results where a "terrible disappointment" and an "embarrassment" !!! Are you sure that you are really speaking of Kasparov ??

"At least Fischer won his '92 match against Spassky." In fact, Kasparov did the same in Saint Louis. In 1995, he beat Anand for the World Championship, and, in 2017, he, once again, finished the Saint Louis tournament ahead of Anand, this while Anand is 6 years younger, still completely active, and playing chess continually at the "absolute top" level. In 1992, Spassky wasn't anymore playing at the "absolute top" level, and it musn't be forgotten that it was Spassky who was 6 years older than Fischer, and not the opposite, so it wasn't very surprising that he lost once more against Fischer. On the contrary, the fact that Kasparov finished this tournament ahead of Anand is in my opinion a real achievement, for Kasparov. In my opinion, it is even rather hard to understand how he could finish ahead of Anand, in these circumstances, with the inactivity and age handicap...

Petrarlsen 8/28/2017 01:09

@ Resistance :

About the Aronian - Anand game and its length (23 moves), I remember that Giri's victory over Carlsen in 2011, also at the Tata Steel tournament, was one move shorter (22 moves). But, even if it is a subjective argument, I think it must nonetheless be pointed out that this game was much more the result of a - quite surprising... - blunder by Carlsen than the consequence of a particularly superior play by Giri...

Petrarlsen 8/28/2017 12:44

@ Resistance :

As I explained before on this page, I think that there is a whole set of elements that shows that Anand's level was approximately the same in 2014 and in 2007. And I think that to say that ALL his seven opponents, in 2014, were in bad form, and that this is the principal reason why Anand won so convincingly this tournament "is going a bit far" ; perhaps his opponents were not really completely in top form, but they were seven ! Seven of the best players in the world ! (Including the World n° 2, World n° 3, and World n° 4, so three out of the four best players in the world - besides Anand - were participating ; only Carlsen was - obviously - missing ; in the 2007 World Championship tournament, only two out of the four best players in the world besides Anand were participating : Topalov - World n° 2 - and Ivanchuk - World n° 4 - were both missing.) Against one opponent (as in a match), it is possible to argue that one of the two players wasn't in top form, but in a tournament, against seven opponents, the form of the participants can only play a marginal role in my opinion... For me, it isn't possible that two such nearly indentical results (the World Championship 2007 and the Candidates tournament 2014) could be compatible with an important difference in terms of level between these two moments, for Anand. And, furthermore, the only significant difference rather goes in favor of the Anand of 2014 : he won this tournament one day in advance - this was not the case in 2007 -, and I think that this is one of the important indicators of domination in a chess competition (be it, by the way, a tournament or a match).

And, yes, Anand had ups and downs during the 2011 - 2015 period, but, nonetheless, taking into account the evolution of his ratings between 2007 and 2015, and, more importantly, the results of the 2014 Candidates Tournament, I think that his chess potential remained intact during this period ; he didn't always played his best chess, but was still capable of really impressive results, as in the 2014 Candidates Tournament (I remember also his Black victory against Aronian - who was 2802 at the time - at the Tata Steel tournament 2013 ; in my opinion, this game would have been completely worthy of the Anand of 2007 or 2008 ; quite an impressive game ; a "demonstration", and against a 2800+ GM, not against a "standard GM"... ; I don't know the answer, but an interesting question would be to know how many times a 2800+ GM lost a game in 23 moves or less in the 10 last years - and with White, furthermore - besides this Aronian - Anand game ; not many - if any ! -, probably...).

Petrarlsen 8/28/2017 12:34

@ lajosarpad : I think we agree on what you explained in your last post.

In fact, when I said that Anand what at the approximate same level in 2014 and in 2007, I meant this essentially for classical games. He certainly didn't lose too much in Rapid and in Blitz (until the last Saint Louis tournament, he was still in the "2750 - 2800 zone" in Classical, in Rapid, and in Blitz, so the difference wasn't extremely important), but he certainly wasn't anymore either as strong as before in Rapid and in Blitz. Is it because of his age ? (But Kasparov seems to be better in Blitz than in Rapid, and he is older still than Anand.) Is it because he isn't interested anymore in Rapid and in Blitz chess ? Is it due to a "personal factor" ? (perhaps also linked simultaneously to age and to a dwindling interest to these forms of chess...) I don't really know, and I must say that I am rather surprised that Anand seems to keep quite satisfyingly, all in all, a good level in Classical chess, and, on the other side, seems to really lose his level in Rapid and Blitz (...and it must be noted that he will be under 2750 in Blitz next month, I think for the first time...).

One more clarification : I think that Anand kept indeed his 2007 level until 2014, but probably not until now. I would choose as the last moment where I think that Anand was really still approximately at his top level November 2015 : it was the last month where Anand was above 2800 points, and, as, in the 2007 - 2014 period, there where many moments where Anand was below 2800 (including for the 2007 World Championship tournament and the 2014 Candidates tournament), I think that, at this moment, he hadn't really lost anything significant in terms of level. Afterwards, I'm not sure at all ; he is still a very good player (permanently in the "2750 - 2800 zone"), but it is quite possible that he is not quite as strong as before now. But I wouldn't be surprised either if Anand regained a 2800 rating, and his top-level ; age doesn't really seems to have the same effect on him than with some other top-players, so I think it is difficult to know for sure what will happen, for Anand, in the next years...

And I think, too, that Anand's style evolved through all these years ; it is probably also a way, for him, to adapt to age... But, in my opinion, what really count are results, and if, by proceeding this way, he manages to keep his level, it is a full success for him... I much prefer him to keep his level by adapting his style, rather than keeping his style and losing his level ; I find this attitude much more logical !... Perhaps it is also the reason why he continues playing at such a high level : perhaps he tried more to adapt to age than the other players of his generation, and he is rewarded for his efforts... But this only a hypothesis, and nothing more...

imdvb_8793 8/27/2017 07:19

Like I said, their non-peak match results are irrelevant for my argument. You can keep counting them, but then you should probably find somebody else to talk to about them, because that's not what I'm talking about at all. I only went into them a little bit to point out even with those included it's not so clear. You can simply ignore that paragraph, if you don't agree. I don't care, because those aren't really part of my main point. (Also, "considerably" older, really?! 18-19 is "considerably" older than 16-17?... And I've addressed the Shirov match, which is a separate thing, for many reasons I've already outlined.)

"Carlsen has this far won three title matches, one of them by scoring the highest percentage in a title match in 100 years (65%)."

This brings nothing new to the discussion. I already compared those results to peak Kramnik's match results, and I maintain they're not as impressive. The higher overall percentage is irrelevant, since non-peak Anand clearly put up less serious resistance than peak Topalov and peak Garry. If you really think 3-0 plus X draws vs. Anand is stronger than 2-0 plus more draws vs. Garry, just because it's more by strict percentages, there's not much left to discuss. (Not saying you're right or wrong, just that I'd immediately lose interest in that part of the discussion if that was the actual argument you put forth, because I have no respect for it. Not for you - I don't know you, so I have no opinion about you, personally. But such an argument I can't take seriously.)

"Kramnik won the World Cup but that was without playing any opponent in the top 20 and I don't think that says much."

Nobody's fault Carlsen has never played it since he became a top player (unless he did and I just don't rememeber it), so he could win it too and put this to rest. (Again, maybe this year...) Also, not Kramnik's fault that his potential top 20 opponents (I'm sure there were plenty) got knocked out by non-top 20 players before having a chance to play against him in that tournament. He beat the players that beat them, which is more than enough. He couldn't do more. It's hard enough to win it that way...

maxharmonist 8/27/2017 04:56

"Carlsen played one candidates match in 2007, when he was just slightly younger than Kramnik in 1994. He lost to Aronian (narrowly.) So, even their early results in matches don't look significantly better for one or the other. Also, Kramnik has won the World Cup"

But the match against Aronian was played when Carlsen was 16 years old, and he was the lowest ranked player of the 16 in the Candidates while Aronian was ranked first. Still Carlsen drew the classical part as well as the rapid part before losing in the blitz tiebreak. I wouldn't take that as an example of Carlsen being a bad "match player" (rather the opposite). The Candidates matches Kramnik lost when considerable older and the favourite (scoring 1-7 against Kamsky, Gelfand and Shirov) were worse results. Kramnik won the World Cup but that was without playing any opponent in the top 20 and I don't think that says much. Carlsen has this far won three title matches, one of them by scoring the highest percentage in a title match in 100 years (65%). I rank him well ahead of Kramnik as as a chess player, maybe 5-7 among the greatest ever already aged 26, with Kramnik somewhere around 12-15.

imdvb_8793 8/27/2017 02:40

PART 3 (reply to Petrarlsen's comments)

Indeed, these are all opinions and we could spend days discussing them. :) There are plenty of arguments for both sides, it's far from clear-cut.

One observation I have on the 2007-2014 comparison - if, as per this comparison, we assume Anand was only slightly stronger in 2014 than in 2007 (if at all), I'd like to point out that between 1999 and 2005, when Garry retired, Vishy didn't win a single game against him in any form of chess competition, and lost seven. Many draws. If we only include their classical results in this period, it's 3-0 and 11 draws, 2-3 of them quite short, many under 40 moves (a kind of draw which, by the way, in a short match for the title, would be at least slightly less likely due to one or the other being behind and having to push.) Even if we consider all of these draws 100% relevant, this 3-11-0 score is still pretty much the same as than Carlsen's 6-14-1 in the two matches against Anand. (Percentage-wise, Carlsen's is 1-2% ahead, but, again, the short draws issue skews things somewhat, I'd say, plus, if we include rapids, I think Garry's score pulls ahead.) Vishy's ratings in this interval hovered between 2760 and 2800, never over 2800, usually around 2780+, and just under 2760 for a short period of time. Even if we consider there's absolutele NO inflation, there's still no difference between his ratings then and in the 2007-2014 period that is big enough that it can't be explained away through random variance or, for example, the absence of a 2800+ monster like Garry to suck ratings away in the latter period... So, I'd say, a purely statistical comparison doesn't seem to yield a conclusion on the 'peak Kasparov vs. peak Carlsen' debate, and it still all depends on whether you believe in the reality of inflation or not. (I famously and strongly do.) And, if anything, it probably favors Kasparov.

"In 1995, Kasparov wasn't probably very far from his peak. "

Well, technically, if you don't believe in inflation (which, again, I do), you can't claim that, because his peak was 2849/2851 in 2000-2001, and in 1995 he got nowhere near that, hovering around 2800, usually under, and never higher than 2815. Even if you do believe it, inflation wasn't quite 50 points between 1995 and 2000...

However, it's true that Vishy was only around 2720 in 1994-1995, and only started his big jump towards 2800 in late '96. Again, none of this makes a convincing statistical case, either way.

"Kasparov's peak was VERY long, while, for the moment, Carlsen was only at his "peak level" for a rather short period of time"

Indeed, this is very significant. Makes Carlsen's peak look somewhat circumstantial/form-based. Not a good comparison for him. But he still has time to prove that wrong, of course. So far, he's not doing a very good job of it.

imdvb_8793 8/27/2017 02:39

PART 2

Peak Carlsen has the two match wins vs. an old Anand. Which is fair, but not THAT impressive in my book. We'll say the Karjakin match wasn't peak Carlsen. Probably true. But then we probably have to exclude the Leko match for Kramnik as well?!... Seems fair. In any case, peak Kramnik has the win vs. Kasparov, which is huge and, again, I doubt something Carlsen could achive, especially on his first attempt. We'll have to agree to disagree, I suppose. The loss against Shirov was two years earlier than the Garry match, when Kramnik was still only 23 and playing, unlike Carlsen vs. Anand, a highly uncomfortable opponent (their score was 11-7 to Shirov, plus draws - compare that to Carlsen's score vs. Anand between 2009 and 2013) - and one he actually started dominating after that! Post-1998, the score between Kramnik and Shirov became 14-5 plus draws, until 2008 (which is as far as I went with the count.) In 2000 and 2001, for example, it was 4-0 plus draws. So, clearly, there is a marked difference between the (match) player Kramnik was before 1999, which is where most of his match losses you give come from, and the one he was after 2000.

And peak Kramnik also has the win vs. Topalov (peak Topalov, by the way, there's absolutely no question about this) - it was a win, let's face it! He forfeited a game and still tied the match, so it was really +1 in classical, plus the rapid win. Plus the psychological pressure put on by Toiletgate... (Carlsen had no such pressure in any of his matches.) What other matches did he play between 2000 and 2007? None that I'm aware of, apart from the Leko one, which we're writing off the same as the Karjakin match for Carlsen. So, a 2-0 win vs. Kasparov, and a +1 win vs. Topalov (more, if you count the rapid results as well.) Both most definitely at their peaks, which ratings, if nothing else, prove. To me that's definitely more impressive than two wins against the same, ageing player, that you've been dominating psychologically for a while. However great of a player he was.

So, yeah, I think I have a pretty solid case that peak Kramnik in match play might just be better (not by much, again) than peak Carlsen in match play. We'll never know, unfortunately...

imdvb_8793 8/27/2017 02:38

PART 1 (apparently, it was too long...)

"I'm curious what you base that assessment on, given that Kramnik was never sole first on the rating list while Carlsen has been first for more than seven years this far and is still only 26 years old. Also, if you say that Kramnik's tournament results were not comparable to the other top players, what do you think of his match results? He is the only World Champion to have a minus score in competitive matches, with losses to Anand, Gelfand, Kamsky, Shirov, and drawn in classical against Leko, Topalov, Grischuk, Radjabov. Then of course the win against Kasparov, but that was a one off given his other match results."

Like I said about the other thing, on personal evaluations of the various pieces of evidence (and of the evidence to the contrary.) Which is all we have. In any case, perhaps I should have made this clear: I 100% meant as a match player, indeed. I'm not saying peak Kramnik would do anywhere near as well as peak Carlsen in tournaments (which is where 95% of a player's rating gains come from, so he could never outrate Carlsen had they both had their peaks at the same time), but in a direct match I think peak Kramnik would be a favorite.

The problem I see with your argument is that when you count his wins and losses in matches, you're not counting them for his peak years only. Far from it, actually. Kramnik played in candidates matches (two different cycles) as early as 1994, if I'm not mistaken. He was 18 or 19. (He also lost some of those matches you listed after having passed his peak.) He actually beat a guy (Yudasin) in one of them, before narrowly losing to Gelfand and losing clearly to Kamsky. Carlsen played one candidates match in 2007, when he was just slightly younger than Kramnik in 1994. He lost to Aronian (narrowly.) So, even their early results in matches don't look significantly better for one or the other. Also, Kramnik has won the World Cup, not having played it too often, if memory serves. Carlsen has played even less, and his best finish was a semifinal - again, in 2007. If he wins the World Cup this year (which he might), we can renew this discussion, of course. For now, it's hard to compare non-peak Carlsen to non-peak Kramnik because their playing frequencies in various formats and at various ages and stages of their careers have been very different. But, again, this is all irrelevant, because I'm only talking about a comparison between peak Kramnik and peak Carlsen.

Pieces in Motion 8/27/2017 12:31

Kasparov was a terrible disappointment in the tournament and considering all the hype, an embarrassment. One would have expected him to do better as he played impressively well in last year's St. Louis speed Chess tourney, even dominating the first half of the tournament. This year's tournament certainly tarnishes his legacy. At least Fischer won his '92 match against Spassky.

News

ChessBase 18 Premium-Package

Winning starts with what you know – ChessBase 18

€499.90

Countering the Ruy Lopez with the fighting Modern Steinitz

The Modern Steinitz (1 e4 e5 2 Nf3 Nc6 3 Bb5 a6 4 Ba4 d6!?) is an uncompromising counterattacking weapon that lets Black put White under pressure from the very start.

€49.90

ChessBase Magazine Extra 223

Videos by Leon Mendonca: French and Nico Zwirs: King's Indian. “Lucky bag” with 50 analyses by Ganguly, Giri, Praggnanandhaa and many more + two video analyses by Josefine Heinemann (in German)

€14.90

Reti Opening Powerbook 2025

The Reti Opening Powerbook 2025 has a tree structure based on a mixture of over 295,000 computer chess games, played in the engine room of playchess.com and the best games played by humans (137,000).

€9.90

Reti Opening Powerbase 2025

Reti Opening Powerbase 2025 is a database and contains 10989 high-class games from the Mega 2025 or the Correspondence Database 2024, 1069 of which are annotated.

€9.90

The Surprising Black Knights' Tango - Unleash Winning Strategies as Black!

After 1.d4 Nf6 2.c4 Nc6!?, Black takes the initiative, luring White into overextending their central pawns, only to dismantle them with precise counterplay. The Tango is not just an opening – it’s a weapon, designed for players who want to win as Black

€34.90

Catalan Powerbook 2025

Catalan Powerbook based on over 470,000 high-class games, most of them played by engines.

€9.90

Catalan Powerbase 2025

Catalan Powerbase 2025 is a database and contains 9158 high-class games from the Mega 2025 or the Correspondence Database 2024, 936 of which are annotated.

€9.90

SHOP

SHOP

Kasparov in St. Louis: a closer look

ONLINE SHOP

The flexible Taimanov Sicilian

Beyond the numbers

Theoretically on the highest level

Flexibility

Aggression

Missed chances

Time-trouble

Kasparov's masterpiece

All games by Kasparov

Links

Discuss

ChessBase 18 Premium-Package

Countering the Ruy Lopez with the fighting Modern Steinitz

ChessBase Magazine Extra 223

Reti Opening Powerbook 2025

Reti Opening Powerbase 2025

The Surprising Black Knights' Tango - Unleash Winning Strategies as Black!

Catalan Powerbook 2025

Catalan Powerbase 2025

Pop-up for detailed settings