[Update:] It's called "Tan gambit" after an Alekhine game. Analyzed below. White should dicline the pawn, since. It can be played as a drawing line for black (no queens).

  1. d4 d5 2. c4 c6 3. cxd5 Nf6!?

This seem to me a good gambit line. I think it's a good way of playing but no one does according to www.365chess.com Some people find exchange variation boring. What is the reason that no one is playing it?

Take a look at the position after 1. d4 d5 2. c4 c6 3. cxd5 Nf6 4. dxc6 Nxc6 5. Nf3 Bg4 6. e3 e5. Black is very active.

Of course white can avoid it by 3.Nf3. Meaning that slav players still have to know the exchange variation. A gambit does not need to be 100% correct. But a gambit no one is playing?

(I found it via 1.d4 d5 2. c4 Nf6. With no one I don't mean 0).

[FEN ""]

1. d4 d5 2. c4 c6 3. cxd5 Nf6!? (3. cxd5 cxd5) (3. cxd5 Nf6 4. dxc6 Nxc6 5. Nf3 Bg4 6. e3 e5)

[Update]: I used lichess (poor) analysis feature to analyze some games. [Between strong players].

  1. Nf3 Bg4. [What I would play].

7. Be2 [What I would be most afraid of]. ½-½, Reinhold (2139) vs. Stevermueer (2150)

7. Bb5 [Messy internet game] 0-1, Drozdovskij (2509) vs. Dobrov (2515)

  1. Nf3 e5. [Queens are exchanged].

6. dxe5 1-0, Sandipan (2637) vs. Elorta (2363)

6. Nxe5 ½-½, Vorobiov (2577) vs. Zvjaginsev (2646)

  1. Nf3 Bf5.

5. Nf3 Bf5 1-0, Alekhine vs Tan

  • My thought is why? The slav defense is sound and quiet with no real problems.
    – yobamamama
    Commented Dec 10, 2019 at 19:41

3 Answers 3


If you want to play this gambit, the best way to get there is by playing 2...Nf6!? 3.cxd5 c6!. White is almost certain to play 3.cxd5 which is both the natural move and the theoretical refutation of Nf6. Of course he can still ignore your pawn on c6 and transpose to the Exchange Slav, but at least with this move order it's not a sure thing that he wanted to play the exchange variation.

The gambit 1.d4 d5 2.c4 Nf6 3.cxd5 c6 was played in the game Udovcic–Canal, Venice 1953. (Thanks to @r4 for the link.)


In the chessok opening tree there are 28 games with this gambit. Going by the computer evaluation it seems to be playable, but the score is atrocious (>80% for white).

So here you have a gambit that is pretty unknown, scores horribly and can easily be ignored if white is not in a mood for adventures. Also it doesn't seem to lead to dangerous complications as far as I can see. There is no automatic attack on the white king, in the worst case white just returns the pawn.

All in all not very surprising that it is rarely played.

  • 1
    The 80% for white seems to be because 1. Blacks about 200pts lower rated 2. Black is playing crap moves.
    – r4.
    Commented Mar 31, 2016 at 5:25
  • I'm not comfortable by the many of the 28 games in wich. One player has a rating, the other don't.
    – r4.
    Commented Mar 31, 2016 at 5:26
  • 1
    Sure, I'm not saying it's not playable. I'm just arguing that a bad score dissuades people from playing it. And of course only weak players playing it also isn't exactly a recommendation. Commented Mar 31, 2016 at 7:48
  • 1
    @BlindKungFuMaster If players followed the sensible strategy of playing solidly against lower rated players and going for aggression and complications against higher rated players, then you'd expect the statistics to show that the gambiteers usually have a lower rating than their opponents.
    – bof
    Commented Mar 31, 2016 at 8:40

Black is active, but he's hardly "sac-a-pawn" active. Once white finishes his development and places his rooks and pieces on the open files and lines, I see no way for Black to stop white from exchanging pieces to a safe pawn up endgame. If I'm giving up a pawn like this, I want activity and something dynamic, like the ability to keep my opponent's King from castling. I just don't see enough from this pawn sac to see it as any good. Maybe in blitz once or twice. But I would hate to have to sit through a classical time control game with this position.

  • "Not any good". Wrong!. Engines says it is playable. (Same score almost as the, a6 slav defence).
    – r4.
    Commented Apr 6, 2016 at 18:53
  • Your answer would be better if you provided analysis. Right now it's just personal opinion on the position in the question. Commented Apr 15, 2016 at 10:56
  • 1
    Ok. Here's some moves. (Not everything, just some possibilities.) I played a few games with this at the chess club, but came up with nothing. 1 d4 d5 2. c4 c6 3 cxd5 Nf6?! 4 dxc6 Nxc6 5 Nf3 Bg4 6 e3 e5 (Tried different ideas here, but came up with nothing. 6...e6 7 Nc3 Be7 8 Be2 Rc8 9 0-0, and Black has nothing to show for his pawn. -If the person I was playing didn't blunder later on, then they usual just ground me down to a pawn down endgame with zero compensation to show for it) 7 dxe5 Qxd1 8 Kxd1 Nxe5 9 Bb5+ Nfd7 10 Nbd2 0-0-0 11Ke2 and White has very little to fear here.
    – JamieStarr
    Commented Apr 19, 2016 at 3:48
  • 1
    At lower levels, one can get away with almost anything, but every engine see's a healthy plus for White, and I bet very few GMs would prefer this for Black. A sacrifice entails gaining positional or tactical compensation. There is none here. Most Slav players relish the possibility of the pawn levers ...e5 and ...c5. Giving up the pawn on c6 loses that for nothing. Black has very little to show. Sure he has pieces that appear active, but, is that really worth a pawn that might have served a more useful purpose? This line sort of reminds me of this line of the Scandinavian: 1 e4 d5 2 exd5 c6?!
    – JamieStarr
    Commented Apr 19, 2016 at 3:56

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.