Caro-Kann

1. @NoobBatter Botvinnik was a master of combining strategy with tactics. I think you shouldn't hate an opening just because a certain player played it.
2. @BlackSalt The problem with engines is it's lack of strategic knowledge and the lack of human understanding in defending: While a human would never want to play a position with doubled pawns just for some active pieces (In the sicilian Taimanov pin variation for instance), engines do quite fine defending such positions. The problem is: engines are great solving complicated positions and tactics, but are far back in strategic endgames or middlegames. They can't calculate strategic elements. I bet the ,,holy" depth of 40 can't show you the ideas or the pawn structure in a position (and Caro-Kann has very solid pawn structures). Also: Humans (at least the majority) don't like to play sharp positions where they have to defend; They want to be on the attacking site, because it's easier to play.

lichess.org/analysis/8/8/4K2k/R6P/8/P7/8/7r_b_-_-_0_1

This position is a draw and the tablebase can proof it. Engine still shows +2.

@hampy sicilian defense, najdorf for example, Nimzo-Larsen attack, Dutch Defense

@ModernChessIsBoring
That's a valid point, just because an engine can handle complicated setups doesn't mean I can.
Can't and argue against it, I couldn't even accurately play the endgame you posted, but I'm hoping as I get smashed I learn something.

A while ago I wondered if there was a sufficiently high evaluation from which I could deduce that a position is winning without looking at something else. There was a post about it, lichess.org/forum/general-chess-discussion/decisive-advantage?page=1.
I came to the realization that I couldn't practically do that with Stockfish, this is the example I contrived
lichess.org/analysis/standard/k7/8/P7/PK6/P7/P7/P7/2B5_w_KQkq_-

As for engines don't understand strategy I don't believe that, in my world strategy is a statistical heuristic, something which NN engines are good at.
I put your position in Sergio and the evaluation I got was +0.46 at glance doesn't seem like a resounding victory.

What I'm not saying:
Engines have solved chess and you will get the answer in a numerical form.

What I'm saying:
The notions that engine evaluation is a joke in the opening is not valid.
I used it to learn how to play against the caro.

@BlackSalt

I agree with your last point in that engines are valid to learn an opening, so long as you don't get stuck to just that opening. Just cause stockfish shows you the best move doesn't mean that a couple other moves would have been better. (Unless you are playing atomic).

@JoelHoge (In response to #5)

I like the Caro-Kann just because it seemed pretty unique for a time (when I was playing 1300s and 1400s). I also decided to get my first ever chess book as "Play the Caro-Kann", so I tried to take what I learned from there on to the chess board. (Don't worry, I also got Dvoretsky's Endgame Manual, not just an opening book).

So far, I have had great results with it. And I agree with the idea that my friend merely says that it is a terrible opening given the fact that I always beat him with it.

@BlackSalt
I think it depends on the position. When i thought about a d4 repertoire i came to the conclusion, that the Sämisch system against the nimzo indian is a very good choice, but the engine doesn't think so: 1. d4 Nf6 2. c4 e6 3. Nc3 Bb4 4. a3 and the Engine thinks Black already equalized, but the variation is much too complex to come to such abstract conclusions.
I can play both ways: slow and positional or sharp and complex, and i realised the complex positions were (sometimes) unsolvable for the engine (i made a normal move (with black) and the engine jumped from +2 to -1).

Your point about the engine can beat any Grandmaster, so we should listen to engine is false alwell: When a strong Grandmaster comes to your Chessclub and beats everyone with the Grob, you wouldn't start playing the Grob, you would look for a refutation.

I (as you can probably guess by my name) hate modern Grandmaster-Chess, because it's a 20 move computer preparation, with an instant draw as they reach an endgame position with equal material.

@BlackSalt
At the risk of continuing an off-topic discussion I will just add this: Everyone today who is serious about studying chess uses engines for assistance in their analysis, but that is beside the point. The whole discussion came about after you gave the figure 0.2 after 1. e4 c6 as if it were to have any bearing on the decision of what opening to play. Opening theory is a broadly charted area and has been developed for many many generations. We know what openings are playable (that is those that have withstood the test of time) and we (generally) choose which ones to play according to our tastes (what type of positions we want to play).

You mentioned that a computer can get to a playable middle game even without using an opening book which is completely irrelevant since you won't be able to copy a computer's method of generating moves. You can of course memorise the lines it spits out, which wouldn't however be better than memorising the early lines given by established theory.

When I study an opening I combine forming my own verbal explanations of strategic ideas, studying the development and history of the opening (important games played etc.), theoretical variations, and engine evaluations.

@ModernChessIsBoring
I hear you. I used to think along those lines as well but over time I have thankfully changed my mind. Chess is such a deep game that even though players can make 20-30 moves to get to a well-known position, there will still be a point in the game where both players are on their own where difficult decisions has to be made.
If it weren't for its enormous depth, chess would be dead a long time ago and none of the elite games would be decisive.

@MathematicChess
That is true. Mostly players start out playing 1. e4 e5 with both colours, so there is definitely an advantage to playing the Caro from that point of view as well. I stopped playing it after being annoyed by a variation in the advanced Caro-Kann (specifically the line called Short variation). Then I took up that line with White but I still haven't had that good results with it. The Caro remains an annoying line to face for me.

EDIT: Sorry for the double post.

Hey guys, I am the "creative" youngster from the CM Sarg0n example :)

So a 2300, I see.