I have had a desire to switch from the semi-slav to the Gruenfeld for a while now, and I'm finally starting to actually do so. I'm comfortable (for now) attempting to self-teach myself the Gruenfeld itself, but since I've never used 1. ... Nf6 before, I'm realizing I have no idea what to play when my opponent doesn't oblige with 2. c4 and 3. Nc3.

I don't even know what other move orders from white I can still validly just attempt to transpose into a Gruenfeld from, and which ones genuinely qualify as "anti-Gruenfeld" systems. And among the ones that ARE anti-Gruenfelds, I don't know which of the myriad database replies make the most sense for me.

My core goal in switching to the Gruenfeld is just to expose myself to new types of positions and middle-game patterns that I haven't seen much before, in a generalized attempt to broaden my chess horizons. As much as possible, my goal against other d4 systems would be "consistency" - new types of position are all well and good, but one set of new positions at a time is better. I don't want to get any more confused than I have to by lots of different "new" things at once.

So what responses make the most sense for a Gruenfeld player looking for semi-consistent positions? I assume neo-Gruenfeld makes the most sense where possible. What about against the Trompowsky? 2. Nc3?

Keep in mind, in terms of what I'm likely to play against, that I'm a Class player, so not necessarily looking for deep lines, and probably won't face too many opponents with master-level theoretical knowledge. I just want some idea of a starting point for some recommended options, so I don't have to randomly pick lines out of a database.

Thanks!

Alternatively, suggest a specific player (who specializes in the Gruenfeld) who might make a good "hero", and I can do my own research on how that particular player responds to other lines.

I've never played the Gruenfeld as black, so take all of this with a grain of salt. But I do play d4 and see the Gruenfeld from time to time, so I'm at least a little familiar with it.

As far as a hero goes, I think Peter Svidler is your guy. Someone correct me if I'm wrong, but I think he's the foremost expert and practitioner of the Gruenfeld in 2700+ chess. He's had some really incredible and theoretically important games in it.

One word of caution, which you probably already know: the Gruenfeld is unbelievably complicated. In a lot of ways, it's the Sicilian of d4. White can play aggressively and sharply, or play more positionally if he chooses. There's an overwhelming amount of theory, but like you said, most opponents shouldn't be outrageously booked up.

Are there any lines in particular that are giving you trouble now, or are you just asking in general? Depending on what you want to do, in case of a move order like 1. d4 Nf6 2. Nf3, you could always play 2...d5 and transpose back into your familiar Slav structures.

Have you thought about playing the KID at all? You can play it against virtually anything, regardless of move order. A GM once told me that he really liked the KID structures because they teach you so much about chess. They force you to calculate, teach about counter-attacking from a worse structure, make you become familiar with different structures (the locked center, or a more open game if black elects to capture on d4 with a pawn if allowed).

I wish I could help more with Gruenfeld-specific knowledge. Hopefully there are some strong players on here that regularly use it.

Thanks! I'll definitely take a look at Svidler then.

As for the word of caution, yes, I am fully aware of the complications. That's actually a big part of why I'm choosing it. My idea is that exploring new openings now will further my chess development by exposing me to positions and structures I haven't seen much before. Looking back on how much I feel I learned when I started exploring the Sicilian (as an e4 player), I suspect that the extremely complicated dynamic of the Gruenfeld will probably make it rich in learning opportunities. Will I lose a few games along the way on pure theory? Probably. I'm willing to live with that though. As for a long term "permanent" repertoire, well, I'm not there yet. Explore new things now, Gruenfeld first, then other stuff later. See how I like it all. Probably will try the KID at some point as well, but the flexibility is actually somewhat counter to my current goals. If I can play it against anything, then I might not learn as much from it, because it won't stretch me quite as much.

As for lines that are bugging me right now, really I just want to know how to play against the Trompowsky. In searching for a hero, I've checked over half a dozen strong players, and each of them have played at least three different second moves on the black side of it - in anywhere from 6 to 11 games. So basically I can't find a ringing endorsement of any one response. It's rare enough that I don't want to sink too much time into exploring all the different possible ideas, I just want to pick one thing, spend a short amount of time on it, and say "if I happen to run into 2. Bg5, I'll just play that and see what happens." Too many options, in an opening that is too rare, is a frustrating combination, lol.

That it's too rare means that you won't be losing many points to it anyway (in comparison with the overall quantity of games), especially after you find a band-aid solution; unlike in the repertoire against 2. c4, you don't have to precise here to ensure a decent overall winrate.

Speaking of the Trompowsky, I've learned from databases that a good band-aid vs it is 2... d5 3. Bxf6 exf6 (allowing to double the pawns but getting a bishop pair) and then Be6 (to discourage c4), c6, Bd6, Nbd7, f5, getting control over the light squares. If White doesn't want to exchange the bishop, then 3... g6 is good, followed by Nbd7, Bg7 and a c5 or e5 break, or possibly Ne4 like in the Bg5 Gruenfeld - in any case, the bishop looks rather useless on g5 then.

Actually, that's how wlrs has met the Tromp in an ongoing game in a chess.com team match; he says his 13th move (Qa5) was questionable, but I approve of the first 12.

Svidler but also Caruana come to mind.

I also like d5. I'm not sure about playing g6, as the structures after Bxf6 exf6 seem more favorable to White in those cases. (I wouldn't enter 1.d4 Nf6 2.Bg5 g6 3.Bxf6 as Black either.)

Wei Yi, probably a future World Champion, has been crushing players with the Black side of the Gruenfeld at the Tata Steel tournament....

Though I agree that g6 is a bit premature at move 2 (because White can then gain an edge by playing 3. Bxf6 exf6 4. c4/e4, so claiming the central light squares with d5 first is a priority for Black), I find g6 a harmless move later on because Black's kingside dark squares are quite invincible anyway (if White does play 4. Bxf6) due to the absence of White's DSB, plus Black's intention is to play f5 in due course to get a grip on e4, and if White then opts to play g4 and gxf5, the f-pawns can get undoubled because Black can recapture on f5 with the bishop.

I'm not saying that 2... d5 3. Nf3 g6 totally neutralises the Tromp, but 4. Bxf6 seems to be bad for White because it gives up an important piece. White's another plan is a passive Stonewall type formation with c3, Nbd2, e3 and Bd3, but then Black plays Bg7, O-O, Nbd7 and Re8, threatening the e5 break, to which White can respond by the preemptive e4 break, but even then the position looks equal, either after exchanges on e4 like in Berg vs Cifuentes Parada, or a more entertaining e5 counterbreak like in Vokoun vs Studnicka. Needless to say, if Black does get e5 in, White is no longer dominant in the centre and has few winning chances.

That said, 3... c5 followed by Nc6 and Qb6 is also a notable response to 3. Nf3. If White refuses to exchange on f6, the position becomes like a Caro-Kann, but with Black controlling e4 (where the knight jumps), so Black is definitely not worse. But if White does exchange on f6, the fact that Black's c-pawn can't return to c6 is in White's favour. So, intuitively, I prefer to keep the c-pawn on c7 for a while to have both c6 and c5 available depending on how White reacts. (The same applies to 3. e3, 3. Nbd2 and 3. c3; only 3. Nc3 leads to a different plan.) I have to engine-check the lines, though - databases are helpless here.

Edit: Stockfish is showing smth ridic. It recommends 3... Ne4 4. Bf4 f6 (oh really?) 5. c4 g5, 5. h4 c5 6. e3 cxd4 7. exd4 e5 8. Be3 Nc6 9. c3 Be7 10. h5 O-O 11. h6 g6, giving Black an edge of over half a pawn. That's totally counterintuitive. In fact, White has scored 62.5% in the four 5. h4 games listed by Chesstempo because the Black players didn't find a good plan.

Hmm, actually, SF has changed its mind and now recommends not the 3... Ne4 4. Bf4 f6 5. c4 g5 ('met' by a paradoxical 6. Bc1), but 'normal' dull moves 4... c5 5. e3 Nc6 6. c3 Bf5 7. Nbd2 e6 8. Nxe4 Bxe4 9. Bd3 Bxd3 10. Qxd3 cxd4 11. exd4 Be7 =, transposing into a kind of Caro-Kann, zzzz. Here's an example 30|30 game of Stockfish (with default settings) with itself (that started from the position after 1. d4 Nf6 2. Bg5 d5), where Black lost only because White found a positional exchange sac (I don't think it's standard) and Black got into time trouble around move 50 (started playing on the increment).

I'm not sure which 3rd move is the best for Black. Tbh, I've recommended 3... g6 to specifically BJJ for a practical reason - it often transposes to the Petrosian System of the Gruenfeld, which is comfortable for someone who plays the Gruenfeld as the main part of the repertoire. Caro / French players will be well off choosing 3... Ne4 4. Bf4 c5 instead (4. Bh4 h5 - threatening to trap the bishop - 5. Nfd2 g5 6. Nxe4 gxh4 gives Black a tiny dynamic edge despite the doubled rook pawns).

Edit: now, speaking of 2. Nc3, the best response seems to be 2... d5 again, transposing to the Richter-Veresov Attack, which is usually followed up by 3. Bg5, thus coinciding with the 2... d5 3. Nc3 variation of the Tromp. Again, Black can go with either 3... c5 4. Bxf6 gxf6 or the g6, Bg7, (Be6,) Nbd7 setup, which can be now followed up by c5 (or c6 as well) because White's knight is blocking his c-pawn.

Third-move anti-Gruenfelds

After 1. d4 Nf6 2. c4 g6, the most popular non-Nc3 moves are Nf3, g3 and f3.

As you already know, 3. Nf3 renders the immediate 3... d5 really bad - after 4. cxd5 Nxd5 5. e4, there's no knight on c3 for the taking, so Black has to retreat the knight to f6 (which is a bit better than b6), and then White typically develops the LSB to c4 or e2 (or to b5 after Nb6), poised to convert the development advantage into an attack.

However, Black can do quite well by delaying d5 (moving Bg7 and O-O first) until one of the two things happens: 1) White develops the other knight to c3 (enabling Nf6xd5xc3), 2) White develops the LSB to g2 (in this case, after Nf6xd5-b6, Black is in no big trouble because White's LSB is not targeting the kingside).

That's of course applicable if Black isn't willing to play KID, which is quite good itself.

Vs 3. g3, the best option imo is the Catalan transposition - 3... c6 and 4... d5 (if cxd5, then cxd5, otherwise Black can play dxc4 and feel good because the g2 bishop is blunted by the c6 pawn).

3... d5 is playable, but note that the best response to 4. cxd5 is a gambit (a computer line that, amazingly, has never been played in games covered by free databases) - 4... c6! 5. dxc6?! Nxc6 6. Nf3 e5 7. dxe5?! Qxd1+ 8. Kxd1 Ng4 9. Ke1 Bc5 10. e3 Be6 11. b3 Ncxe5 12. Nxe5 Nxe5 13. Bb2 Bb4+ 14. Bc3 [forced, otherwise Black develops a big attack, e.g. Kd1 Bg4+] Bxc3+ 15. Nxc3 Rc8 and so on - Black is a pawn down but has the initiative.

The problem with 3... d5 4. cxd5 Nxd5 is that, after 5. e4 Nb6, White can develop the knight to e2, leaving the fianchettoed bishop more active, as opposed to the 3. Nf3 line, and then plays d5.

Vs 3. f3, as I've already said here, good responses are Bg7 followed by d6 (KID), c5 (the Modern Benoni) and e5 (the Leko Gambit), though I find 3. f3 the best anti-Gruenfeld out there. Which one to choose is again a matter of personal preference and familiarity with the Benoni and KID.

I hope you've understood the reasoning behind my rants, I'm sorry for their verbosity.

I should see if I still have a copy of Well's book in terms of what his recommendation was. It's been a long time since I played the Tromp.

But 1.d4 Nf6 2.Bg5 d5 3.e3 g6 4.Bxf6 seems pretty pleasant to me (for White) -- I see that originally neither of us really specified White's 3rd move :-) -- although transpositions to 3.Bxf6 lines are possible.

i always just played 2. ...Ne4 against the tromp.