Terrible win rate with Caro Kann

Am I facing a real problem playing the Caro-Kann. I realise a lot of people high recommend it, but I'm finding I face the advance variation almost every time, which I struggle a bit to play againt - I have something like a winrate of 33% or lower with the Caro-Kann, but if I play Sicilian, hyper accelerated dragon, open, closed, I actually have a positive winrate.

What do you think is the best answer to the Caro-Kann advance for black? Maybe I should stick with what I win more with? At least for this elo?

In the Caro-Kann, Black is not trying to win. Black is trying to keep White from winning. If you want to win with Black, play the Sicilian.

I recently watched a course by Gothamchess - and started playing C5 against advance variation, and I've had much more success.

https://www.chess.com/forum/view/for-beginners/hi-what-is-the-best-defense-after-scaillin-defense

 Typically I would respond to Dxc5 with Nc6

AHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH

If you are having trouble with a specific opening, it tends to mean you do not understand the ideas in that opening well.  You have a couple options: study it more, or play an opening you understand better.  If you are looking to improve, I would recommend the former.  If you just want to pick up some wins, the latter.

Regarding how to study: you can start by playing through master-level games (if you can find annotated ones, even better) and see how they handle positions you are having trouble with.  Alternatively, you can invest $20-30 in Jovanka Houska's book on the Caro-Kann.  It covers the most popular lines you would see at the club level and she explains things very well.

I hesitated in recommending it, but then I looked at his rating.  Since she uses model games and explanations, it would be a decent starting point (moreso than Gotham's course).  That said, I completely agree that he has more important things to study than openings (which is why I did not mention Erwin L'Ami's Chessable course on it) ; however, reviewing annotated games would be helpful.

Yeah, one of the best things to do is use something like chessgames.com and get 50 games from somewhere around 1950-1970, and 50 modern games. Try to pull from world championship matches, candidates, interzonals, or other top level events.

Look at all the games at a fairly quick pace, less than 5 minutes a game. Write down a few simple notes.

Then go over them a few more times, but taking a little longer and analyzing a little more. If some of the games are not very good you can replace them.

After surveying 100 games you'll have a good feel for some of the basic ideas and variations. Of course books are useful, but if you haven't even seen 100 quality games in an opening then you probably have no clue about it.

Of course none of this is useful for "post-beginners" (as pfren calls them), and they have more important things to study (as pfren points out).

What is pfren's objection to Houska's book? Does he prefer Schandorff`s books? I have his most recent one, too.

So the question remains . . Caro-Kann, worth learning correctly or not

I suppose in order to improve generally, I'd say yes

The bit I don't understand is, I've heard so many players say the Caro leads to positions that "play themeslves" etc . . . and I just end up tying the board in knots mid game with really closed centres. I play the first part fine, but everything seems to fall apart after say move, 15+.

If I just play Sicilian for E4, or KID for D4, I really do find myself in positions that play themselves. But if white replies to Caro, with advance or fantasy, I just end up tying myself in knots.

For instance - check out this game:

Like I've said twice: Just tied up in knots.

The reason old games are good is you get to see high quality ideas. It doesn't matter if the theory is still relevant today. There are winning and losing ideas associated with certain piece placements, structures, endgames, and tactics. You can't understand modern GM games if you skip what came before them. Earlier ideas are easier to understand because it's the nature of humans that we discover easier things first and build on them later.

Plus, on a personal note, IMO it's usually better to play old lines with clear ideas than it is trying to mimic super GM or engine lines where memorization is the only way you can pretend you know what you're doing.

If you suck at an opening, then don't play it.

That's fair, but I'm hoping to get a better understanding of the game as to WHY it's not working

The basic flow of a game is bring your pieces off the back rank, castle, play a pawn break, then infiltrate to your opponent's side of the board.

In the caro you usually have this structure:

-

-

So your pawn break is either e5 or c5. If you can't find a way to play one of those breaks then your position is just worse... but you can't play them right away. Pawn breaks favor the side who is ahead in development, so first you'll get all your minor pieces off the back rank and castle. Then you'll play e5 or c5 at a moment when it doesn't lose material.

That covers what might go wrong for you regarding the opening... if you're doing those things but still losing then it probably has nothing to do with the opening. It probably has to do with basic tactics and calculation.

For example this game

(See the variation on move 5 for what I recommend)

-

-

This game by you used that sort of development. I think that's a big improvement.

-

Oh, you're white in that 2nd game, I didn't notice lol

Thank you very much though, that was very helpful.