Is there a refutation to the Nimzo-Indian 5...c5 6. f3 Saemisch Line?

My understanding is that the above was once a very mainstream opening, and both ChessBase and Chess.com's database show it having been played hundreds of times. However both databases also show Black with the better percentages in the d5, d6, and especially Nc6 responses. Even so it appears that high rated players are continuing to try their luck here. Can anyone give me an overview of this line's current state of affairs? I am particularly looking for a punishing line with the black pieces or, if it has actually been "busted", the game(s) in which this occurred.

Thanks and please forgive me if something about this question is particularly ignorant; I'm still very new to these types of set-ups.

LoL There is definitely no refutation for such a mainstream line. Here's a classical Lilienthal-Ragozin game (1935) where black won:

https://www.chess.com/games/view/25789

"Refutation" might be a strong word, but there must be a reason that black is getting the better results?

Just because Black's results are better doesn't mean there is a refutation.

The move 4.a3 is a line of the Nimzo-Indian, known as the Saemisch, but it's not as popular as 4.e3 or 4.Qc2, and one reason is that there is less room for error by White.  One slight slip-up and Black wins.  The 4.e3 and 4.Qc2 lines are far safer for White, but White's advantage requires patience in those lines.  Something many chess players don't have.

Here is another reason why the Nimzo-Indian is considered one of the strongest defenses to 1.d4.  No matter what Black plays, White can get a SLIGHT advantage because he goes first.  That goes for the Queen's Gambit Declined, Slav, Semi-Slav, Queen's Gambit Accepted, Nimzo-Indian, King's Indian, Grunfeld, Modern Benoni, etc.  However, the thing about the Nimzo-Indian that stands out is that White's advantage is in the form of the Bishop pair and other very subtle positional factors.  This requires a very slow and methodical approach to succeed.  Something like the Benoni, White can blow up the center with early f4 and e5 moves, and storm at the Black King.  Let's just say, hypothetically, that of all White wins in the Modern Benoni in the last 20 years in games involving 2 players over 2500, the average game is 31 moves.  And then let's say hypothetically that of all White wins in the Nimzo-Indian in the last 20 years in games involving 2 players over 2500, the average game is 44 moves.  Those extra 13 moves is 13 more chances for White to screw up, whether that be blunder into a loss, or simply make a couple of second best options and turning what was a small advantage into a dead equal game.

4.a3 tends to be played more by players looking for blood.  They don't care about pawn structure typically, which can be a long term issue if the attack fizzes out.  In 4.Qc2 and even 4.e3, the main lines don't involving doubling of pawns.  Sure after 4.e3, Black can play the immediate 4...Bxc3+, but it's not a good move at all and White has other trumps.  Outside of the Hubner, which would require White to play Bd3 and Nf3 after 4...c5, most lines involve a trade on c3 but then also a trade of pawns on c4 (or d5) and so White doesn't end up with the doubled pawns.

While this does deem 4.e3 and 4.Qc2 "more sound" than 4.a3, it doesn't mean there is a refutation to the entire 4.a3 system!

There have been some recent work, including repertoire books, on the f3 line in particular, with many dangerous ideas for White.  I’ve always found the simplest answer to be the following.

Hi there, I advice to study Ivan Sokolov's book winning chess middlegames as he explains the pros and cons of this type of structures. I played the two systems (4a3 and 4f3) most of my life. I also play sometimes the nimzo from the black side. To me the Saemish attack is a beautiful system to reach a very sharp middlegame. It is far from risk free and it requires lots of games to understand it. Honestly, I would never trust the comp in this postion neither the stats. The better player prevails in these positions.

Nah, in the Nimzo, Black already equalizes after Bb4. Whatever White does, Black has some counterplay. Not like in the QGD for example, Black is like a sitting duck waiting to be executed. That's why top players avoid the Nimzo as White unless they have very specific ideas against it.

Thank you all for all the helpful and high quality responses.

In case anyone else finds this thread looking for some fantastic play from these positions, here is Game 9 of Carlsen-Anand's Championship match in 2013.

https://www.youtube.com/watch?v=qa12XYa1Ols

Some incredible stuff. Lots of egg on my face given the title of this thread.

If you look at Nimzo's that have occured in World Championship matches you will find in the majority of games white creating a lot of problems. The typical chess com opening forum idea that if white allows a Nimzo he might as well pack up the peices and go home, is very peculiar.

I know this thread is extremely old and I hope it isn't poor judgement to resurrect it. Frankly I am also embarrassed, now that I have a tad bit more knowledge, that I thought such a fundamental chess variation was "refuted."

I just wanted to say that I ended up purchasing this book and it was an incredible recommendation. I am one of those types of average club players that has probably obtained 100+ chess books through my career and "Winning Chess Middlegames" by Sokolov has got to be among the greatest. As far as books on opening-through-middlegame that depend heavily on game examples, only Jesus De La Villa's Sicilian book for White can compare for impact on my own personal play. In case anybody stumbles on this thread in the future (I find myself looking at old threads more often than new ones) this was a perfect recommendation for somebody like myself trying to learn the pros and cons of basic Nimzo strategy, from either side of the board.

...I've played a bunch more games with both colors and still feel clueless about how to evaluate or how to approach the Saemisch, though