Caro Kann counter

@ColossusChess said in #10:

It's not new information for me at all and such statistics are useless.

The Caro-Kann is not the most sucessful opening for example above 2200. Weaker players tend to struggle since they don't know any theory but that doesn't matter for me.

You wrote a comment with an answer in such a way that it was not sent to me. And it should not have been sent, since you wrote it without an addressee. I saw it here by chance, visiting the topic again (and I might never visit it again). I will not express doubts about the information not being new, although I doubt it. Since very few people know that c6 is the most successful response to e4 according to Lichess statistics (and in general, black is in the plus over white). I have not read or heard this anywhere (from anyone), but I personally saw it in Lichess statistics. And even if you specifically knew this, it does not cancel the fact that the overwhelming majority of those who read my comment did not know this before. That is, in any case, my comment deserved a like, not a dislike. And no, such statistics are not at all useless. In fact, they are very useful. They show the most successful response to e4 for the totality of all players on Lichess. It is even more correct to classify this into the category very-very useful. I know that if you look at the games of professional chess players, this answer is no longer in first place. Did I say otherwise? No! But by the way, it is still one of the most successful there. It is even more successful there, for example, than e5 and e6. Think about it, neither among all players, nor among professional chess players, is the answer to e4 with e5 more successful than c6! Lots of people play e5. And that is how things are. And almost all other answers to e4 are also less successful there and there than c6. And we are talking about all players here. And about all players - it is the Lichess statistics for all players that is indicative (and not for professional chess players). And it is not only very weak chess players who do not know the theory who experience difficulties in Caro-Kann. Lots of chess players who are not weak and know the Caro-Kann variations they have chosen experience difficulties in them. Which is clearly visible (confirmed) by the Lichess statistics for professional chess players. The answer c6 is also among the most successful answers to e4 (consistently ahead of the answer e5 and many other popular answers).

@ColossusChess said in #10:

It's not new information for me at all and such statistics are useless.

The Caro-Kann is not the most sucessful opening for example above 2200. Weaker players tend to struggle since they don't know any theory but that doesn't matter for me.

Answer c6 to e4:

• among all chess players on Lichess - in first place in terms of success for black;
• among professional chess players - in third place in terms of success for black.
  Everyone can see for themselves:
  https://lichess.org/analysis#1
  Among both of those chess players, it is more successful for black than, for example, e5. Which many, many people play. And than almost all other moves. And this is useless in your opinion, doesn't it matter? Ha! You are very wrong! This is very important information! It deserves a like, not a dislike!

Lichess database is not the only database out there.
According to Chessbase Live database for example we have:

percentages for White:

1...c5 52% for White 2428 ELO avg. 530387 games.
1...e5 55% 2445 273219
1...e6 55% 2412 144054
1...c6 53% 2429 105781
1...d6 55% 2411 43661
1...g6 51% 2424 35448
1...d5 56% 2397 24511
1...Nf6 55% 2405 20135
1...Nc6 54% 2394 6820
1...a6 53% 2411 707

And of course no one will abandon his opening because another opening scored 2% better on average. 1...e5 is the closest to 0.00 move according to the engines.

In blitz the Caro-Kann is especially hard to play on higher levels since white has a small edge in many lines and accurate defence is difficult to do with no time.

@ColossusChess said in #13:

Lichess database is not the only database out there.
According to Chessbase Live database for example we have:

percentages for White:

1...c5 52% for White 2428 ELO avg. 530387 games.
1...e5 55% 2445 273219
1...e6 55% 2412 144054
1...c6 53% 2429 105781
1...d6 55% 2411 43661
1...g6 51% 2424 35448
1...d5 56% 2397 24511
1...Nf6 55% 2405 20135
1...Nc6 54% 2394 6820
1...a6 53% 2411 707

And of course no one will abandon his opening because another opening scored 2% better on average. 1...e5 is the closest to 0.00 move according to the engines.

In blitz the Caro-Kann is especially hard to play on higher levels since white has a small edge in many lines and accurate defence is difficult to do with no time.

Are you answering like this on purpose so I wouldn't see? Don't you understand that you're answering incorrectly? You're writing into the void, not to a person! A person doesn't receive a message when you write into the void like that! And? Have you looked at your statistics? In your statistics, the answer c6 is in the same third place in terms of success for Black! After exactly the same moves. Plus, it doesn't include a bunch of other moves (which are in Lichess's statistics), in all of which the result is worse for Black than c6. And doesn't it bother you about e5, that there are a whole bunch of moves that are more successful for Black? What's the point of slightly better engine ratings if in practice people get worse results? And worse even than completely crooked moves! After all, White's opponents are much better prepared for e5! That's it!

@ColossusChess said in #13:

Lichess database is not the only database out there.
According to Chessbase Live database for example we have:

percentages for White:

1...c5 52% for White 2428 ELO avg. 530387 games.
1...e5 55% 2445 273219
1...e6 55% 2412 144054
1...c6 53% 2429 105781
1...d6 55% 2411 43661
1...g6 51% 2424 35448
1...d5 56% 2397 24511
1...Nf6 55% 2405 20135
1...Nc6 54% 2394 6820
1...a6 53% 2411 707

And of course no one will abandon his opening because another opening scored 2% better on average. 1...e5 is the closest to 0.00 move according to the engines.

In blitz the Caro-Kann is especially hard to play on higher levels since white has a small edge in many lines and accurate defence is difficult to do with no time.

And for some reason you are mistakenly writing about professional chess players again (but even for them the statistics with c6 are in third place). And what is being discussed is: for everyone, not for them. And for everyone - c6 is in first place. And this is despite the fact that you need to know much less theory to play c6 than your e5. As much as several times less. And the actual result is better!

@ColossusChess said in #13

And I draw your attention to the fact that the statistics discussed are for all chess players (in which c6 is in first place in terms of success for black), and not for chess professionals (in which c6 is in third place in terms of success for black). And I draw your attention to the fact that according to it (this statistics), black even beat white! Although with a minimal advantage (approximately 50.5%:49.5%), but black even won! Those playing Caro-Kann (black) even beat their opponents, who were playing white against them!

if i take all lichess ratings:

... e6 +1% (win for black vs. white)
... c5 0%
... e6 0%
... g6 -1%
... Nf6 -1%
... d6 -2%
... b6 -3%
... Nc6 -4%
... a6 -4%
... f5 -5%
... e5 -6%

So among all lichess players, e5 is 7% lower relative win rate for black than e6 (Caro-Kann). c5 (Sicilian) and e6 (French) are also good with 6% lower.

So, to me the main message isn't that Caro is best, although its interesting to know. Much more surprising is how much worse e5 is doing. It even does worse that b6, Nc6, and a6.

The results for intermediate lichess players 1400-1800 look similar, (albeit Caro is even 2% ahead of white).

... c6 +2%
... Nf6 +1%
... c5 0%
... e6 0%
... g6 -1%
... d5 -2%
... b6 -3%
...Nc6 -3%
... a6 -4%
... f5 -4%
... e5 -6%

For beginner players (1000-1200), it looks like this:

...c6 +2%
...e6 +1%
...c5 +1%
...g6 +1%
...b6 -1%
...d5 -3%
...d6 -3%
...a6 -5%
...Nf6 -6%
...e5 -6%
...Nc6 -7%

So, consistently in the lichess database e5 performs poorly for black.

What about masters?

...g6 -5%
... c5 -6%
...c6 -9%
...e5 -10%
...e6 -11%
...Nf6 -11%
...Nc6 -11%
...d6 -12%
...d5 -15%
...b6 -19%
...a6 -21%

So, for Masters, things change fairly dramatically
...g6 is best, at -5%, then ...c5 at -6%, and ...c6 at -9% and e5 at -10%. I.e. there is very little difference in relative win rate for e5 and c6 at the masters level.

Anyway, most of us are not masters and these numbers can shift dramatically depending on what line you play in a given opening. Play what you are familiar with and what your opponent doesn't know (if you have prior knowledge). Enjoy this 'analysis' with a grain of salt. As engines show new lines or fads come and go for amateurs, these numbers are sure to change. I guess Caro will always be a decent choice though if you are looking for a good opening for black.

@EmaciatedSpaniard said in #17:

if i take all lichess ratings:

... e6 +1% (win for black vs. white)
... c5 0%
... e6 0%
... g6 -1%
... Nf6 -1%
... d6 -2%
... b6 -3%
... Nc6 -4%
... a6 -4%
... f5 -5%
... e5 -6%

So among all lichess players, e5 is 7% lower relative win rate for black than e6 (Caro-Kann). c5 (Sicilian) and e6 (French) are also good with 6% lower.

So, to me the main message isn't that Caro is best, although its interesting to know. Much more surprising is how much worse e5 is doing. It even does worse that b6, Nc6, and a6.

The results for intermediate lichess players 1400-1800 look similar, (albeit Caro is even 2% ahead of white).

... c6 +2%
... Nf6 +1%
... c5 0%
... e6 0%
... g6 -1%
... d5 -2%
... b6 -3%
...Nc6 -3%
... a6 -4%
... f5 -4%
... e5 -6%

For beginner players (1000-1200), it looks like this:

...c6 +2%
...e6 +1%
...c5 +1%
...g6 +1%
...b6 -1%
...d5 -3%
...d6 -3%
...a6 -5%
...Nf6 -6%
...e5 -6%
...Nc6 -7%

So, consistently in the lichess database e5 performs poorly for black.

What about masters?

...g6 -5%
... c5 -6%
...c6 -9%
...e5 -10%
...e6 -11%
...Nf6 -11%
...Nc6 -11%
...d6 -12%
...d5 -15%
...b6 -19%
...a6 -21%

So, for Masters, things change fairly dramatically
...g6 is best, at -5%, then ...c5 at -6%, and ...c6 at -9% and e5 at -10%. I.e. there is very little difference in relative win rate for e5 and c6 at the masters level.

Anyway, most of us are not masters and these numbers can shift dramatically depending on what line you play in a given opening. Play what you are familiar with and what your opponent doesn't know (if you have prior knowledge). Enjoy this 'analysis' with a grain of salt. As engines show new lines or fads come and go for amateurs, these numbers are sure to change. I guess Caro will always be a decent choice though if you are looking for a good opening for black.

1. Yes. That's it. That's what I'm talking about. Caro-Kann is strong!
2. You have an important typo (better correct it, it's very important, right on the topic under discussion) at the very beginning. It should be c6, but it says e6. In the very first line, where it says black wins over white. And then in the text.
3. And against masters (national, FIDE, international) here on Lichess I've scored about 36% over the entire time (I understand that mostly below average masters play against me, but still about 36%). I beat different masters here regularly. I keep statistics of my games against masters for myself (I'm interested in this for myself), everything is calculated. They are not that far behind me (higher in level of play, but not colossally, just higher. I received the official title of candidate for master of sports in chess (but not international, but my country's) 25 years ago (when I was 15 years old). Then I easily confirmed it for about 7 years, and then I threw it in the trash by stopping playing in official tournaments, since I had completely different interests (and a little over a year ago I started playing again, but only here at Lichess for now). I still play at about that (my former) strength.

Those statistics mean nothing. The database contains a mix of bullet, blitz, rapid, classical, correspondence. The database contain all kind of players, from beginner to world champion. The database contains lopsided encounters like 2200 vs. 2700. The database contains recent games as well as old games.

For meaningful statistics you would have to filter say both players +2700, classical time control, 2024 or later.

It has been a long time the Caro-Kann was played in top level chess, like World Championship Matches, Candidates' Tournaments.

@tpr said in #19:

Those statistics mean nothing. The database contains a mix of bullet, blitz, rapid, classical, correspondence. The database contain all kind of players, from beginner to world champion. The database contains lopsided encounters like 2200 vs. 2700. The database contains recent games as well as old games.

For meaningful statistics you would have to filter say both players +2700, classical time control, 2024 or later.

It has been a long time the Caro-Kann was played in top level chess, like World Championship Matches, Candidates' Tournaments.

You are mistaken. In general, all the statistical collections that we have talk only about this. There is not a single one about the opposite. There is Lichess statistics for all chess players - in them, c6 is the most successful response to e4. There is Lichess statistics for professional chess players - in them, c6 is the third most successful response to e4. There is Lichess statistics for professional chess players, which ColossusChess took from another place - in them, c6 is the third most successful response to e4. After the same 1st and 2nd places. No other statistics are visible. Plus, EmaciatedSpaniard made selections above for different levels of chess players - everywhere, c6 is either the most successful response for black to e4, or one of the most successful. This is exactly what, according to you, does not exist, without which it is impossible to draw a conclusion. It does exist. Here is a selection above for you, a person made. Everywhere, c6 is either the most successful response for black to e4, or one of the most successful. And the sample is many times larger than enough for conclusions. Statistically reliable result. Obtained on hundreds of thousands of games. Caro-Kann is played by chess players of all levels (from beginners to world champions) against players of all levels (from beginners to world champions). Everything is distributed over all intervals sufficiently for statistical reliability. And there are a huge number of games in Caro-Kann in this statistics. Many times more than enough for reliability. The move c6 is actually the fourth most popular response to e4. And have you and ColossusChess colluded here to incorrectly respond to comments so that your comments are written into the void, and not addressed to a person (me)? Your comments written like this do not come as a message to a person (since they are written incorrectly, without specifying the addressee).