Some moves that are dismissed by experienced players as “obviously bad” moves that beginners would make, turn out to be very good, and sometimes crop up in the games of the world’s elite.

Very inexperienced players sometimes begin games like this: 1.e4 e5 2.Nf3 d6 3.Bb5+. They may play 3. Bb5 because they overvalue the significance of giving check or perhaps because the move struck them as active and aggressive. If those players continue to study and learn, they’ll learn that moves like 3. Bb5+ are typically a waste of time. Black will play the useful move 3…c6, driving the bishop away while increasing his control over the center. Over time, moves like 3. Bb5 can even become invisible to players; they won’t even be considered as possible candidates. That turns out to be throwing the baby out with the bath water. There are plenty of examples where moves of this sort are playable or even advantageous for the side that seems to be losing a tempo.

Take the position arising in the Ruy Lopez, one of the oldest openings, after 1. e4 e5 2. Nf3 Nc6 3. Bb5 a6:

White can and sometimes does play 4. Bxc6, but 4. Ba4 is far more popular. It is reasonable to ask why that is true, because Black can follow up with 4…b5, forcing 5. Bb3. Why not just play 3. Bc4? The bishop ends up on the a2-g8 diagonal anyway, so it appears as if White is giving Black two free moves on the queenside to gain space and to possibly fianchetto the queen’s bishop.

The reasoning is logical, but as is probably clear, given the popularity of the Ruy Lopez, White is not harmed by the inclusion of …a6 and …b5. The most stable way to meet the Italian game is with 3…Bc5, when a very old main line runs 4. c3 Nf6 5. d4 exd4 6. cxd4 Bb4+ 7. Bd2 Bxd2+ 8. Nbxd2 d5, with equality. White could also try 6. e5, but Black has the nice counter 6…d5!

Black has already achieved equality.

By contrast, in the Ruy Lopez, after 3…a6 4. Ba4 b5 5. Bb3 playing the same way won’t work for Black: 5…Bc5 6. c3 Nf6 7. d4 exd4 8. e5 (see diagram below) gives White a big advantage, as this time 8…d5?? doesn’t attack White’s bishop.

Other openings illustrate the same theme. In the Grand Prix Attack after 1. e4 c5 2. Nc3 d6 (2…Nc6 is better, but only if the player who is Black is ready to play an Open Sicilian variation other than the Najdorf, or is willing to meet 3. Nf3 with 3…e5) 3. f4 g6 4. Nf3 Bg7:

For a long time, White used to play 5. Bc4 just about automatically. White’s plan is to castle, play d3, Qe1, f5 and try to checkmate Black. Black has a number of good counters, one of which is to put the kingside knight on e7 and play …d5. For example 5…Nc6 6. 0-0 e6 7. d3 Nge7 8. Qe1 0-0 9. f5 d5.

Frustrated by this plan, White found an ingenious tweak: 5. Bb5+ Bd7 6. Bc4!

In some sense White has lost a tempo, but it’s not a sense that’s relevant to a proper evaluation of the position. Black’s extra move, 5. …Bd7, can create problems, as the Black queen will no longer support the d7-d5 advance in the center. This idea defanged one of Black’s main defensive ideas, and forced Black to find some new defensive methods.

A recent game featuring Vladimir Kramnik, the former World Champion, who is currently ranked No. 3 in the world, was quite interesting. Kramnik had White against Rainer Buhmann, a German grandmaster, in the Sparkassen Chess Meeting in Dortmund in July. The game began 1. e4 e6 2. d4 d5 3. Nc3 Nf6 4. e5 Nfd7 5. f4 c5 6. Nf3 Be7 7. Be3 b6 8. Qd2 0-0 9. h4 Nc6:

Kramnik now played the surprising 10. Bb5. After 10…Qc7 11. 0-0-0 a6 Kramnik retreated the bishop to d3.

A natural question would be to wonder why Kramnik didn’t play 10. Bd3 or 10. 0-0-0 followed by 11. Bd3. The latter is dealt with easily enough: Black would meet 10. 0-0-0 with 10…c4, and the bishop won’t get to d3. The problem with 10. Bd3 is subtler, but one possible answer is that Black could have played 10…f5 (which is analogous to what happened in the game), and after 11. g4 cxd4 12. Bxd4 fxg4 13. Ng5 g6 14. Nxe6 Nxd4 15. Nxd4 Bxh4 Black would have been nearly winning:

By contrast, after the game continuation of 10. Bb5 Qc7 11. 0-0-0 a6 12. Bd3 Kramnik answered 12…f5 with 13. g4. The game could then have continued 13…cxd4 14. Bxd4 fxg4? 15. Ng5, which would have given White a winning advantage:

Castling queenside made a huge difference for White, and Black’s “free” move of …a6 didn’t do much to help him. Here is the complete game, which actually ended in a draw:

Kramnik, Vladimir vs. Buhmann, Rainer
Dortmund 44th | Dortmund | Round 3 | 12 Jul 2016 | ECO: C11 | 1/2-1/2
1. e4 e6 2. d4 d5 3. Nc3 Nf6 4. e5 Nfd7 5. f4 c5 6. Nf3 Be7 7. Be3 b6 8. Qd2 O-O 9. h4 Nc6 10. Bb5
10. O-O-O?! c4  )
10. Bd3 f5 After this the plan in the game won't work:
11. g4? cxd4 12. Bxd4 fxg4 13. Ng5 g6 14. Nxe6 Nxd4! 15. Nxd4 Bxh4+  )
10... Qc7 11. O-O-O a6 12. Bd3 f5 13. g4! c4 Here
13... cxd4 14. Bxd4 fxg4? is bad for Black:
15. Ng5 g6 16. Nxe6 Black is already lost, and obviously
16... Nxd4? makes no sense here.
17. Nxc7  )
14. gxf5 cxd3 15. fxe6 Ndb8 16. Nxd5 Qd8 17. Nxe7+ Nxe7 18. Ng5 h6 19. Qxd3 hxg5 20. hxg5 Bxe6 21. Qh7+ Kf7 22. d5 Bf5 23. e6+ Ke8 24. Qxg7 Qc7 25. Rh2 Nxd5 26. Qxf8+ Kxf8 27. Rxd5 Bh7 28. b3 Ke8 29. g6 Bxg6 30. Rh8+ Ke7 31. f5 Bxf5 32. Rxf5 Qc3 33. Bg5+ Kxe6 34. Rf6+ Qxf6 35. Bxf6 Kxf6 36. Rh6+ Ke5 37. Rxb6 Kd5 38. Kb2 Nc6 39. a3 Kc5 40. Rb7 Rg8 41. Rh7 Rg2 42. Rh5+ Kd6 43. Kc3 Rg3+ 44. Kb2 Rg2 45. Kc3 Rg3+ 46. Kb2 Rg2

Another ultra-modern finesse arises in the London System, which has recently become more popular among elite players. After 1. d4 Nf6 2. Nf3 e6 3. Bf4 d5 4. e3 c5 5. c3 Nc6 6. Nbd2 Bd6 7. Bg3 0-0:

The move 8. Bd3 used to be automatic, save for people occasionally playing 8.Ne5. But lately 8. Bb5 has become all the rage, with the salient point of meeting 8…a6 not with an exchange on c6 but with the surprising 9. Bd3. It might seem that the idea is to make Black’s a-pawn a target, and there may be a line where that is the case. The main point is a bit subtler. After 9…b6 10. e4 Bb7 – the same way Black would play with the pawn on a7 – White has 11. exd5.

In 25 games with 8. Bd3 rather than 8. Bb5 a6 9. Bd3 no one has bothered to play this because Black is absolutely fine after …Qxd5. In the version where Black is forced to play …a6, however, 11…Qxd5 isn’t very good because of 12. Nc4!, threatening to take on b6. That is the point. Here is an example of what can happen after 8. Bb5:

Nisipeanu, Liviu Dieter vs. Cornette, Matthieu
Bundesliga 1516 | Germany | Round 12.3 | 09 Apr 2016 | ECO: D02 | 1-0
1. d4 Nf6 2. Nf3 e6 3. Bf4 c5 4. e3 Nc6 5. Nbd2 d5 6. c3 Bd6 7. Bg3 O-O 8. Bb5!?
8. Bd3 b6 9. e4 Be7 10. exd5 Qxd5  )
8... a6 9. Bd3!? b6 10. e4 Be7 11. exd5 exd5
11... Qxd5?! 12. Nc4 gives White a very pleasant advantage after the queen's forced retreat to d8.  )
12. Ne5 Bb7 13. O-O cxd4 14. Nxc6 Bxc6 15. cxd4 Bb5 16. Bxb5 axb5 17. Qb3 b4 18. Rfe1 Nh5 19. Be5 Qd7 20. Nf1 Rfc8 21. Ne3 Ra5 22. Rac1 Rc6 23. Nf5 Bf8 24. Qh3

These odd finesses are not just in variations featuring Bb5. Here’s a funny example that’s been known for some time that had a high-level test in a blitz match between Magnus Carlsen and Alexander Grischuk earlier this week: 1. e4 c5 2. Nf3 d6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 a6 6. Be3 Ng4 7. Bg5 h6 8. Bc1!? Nf6 9. Bc4.

Of course, White could have played 6. Bc4 and reached almost the same position. The one important difference is the placement of Black’s pawn on h6. One line where this difference is in White’s favor would be 9…e6 10. Bb3 b5 11. 0-0 Be7 12. Qf3 Qc7 13. Qg3.

When the nearly identical position arises through a normal Fischer-Sozin Variation move order (6. Bc4 e6 7. Bb3 b5 8. 0-0 Be7 9. Qf3 Qc7 10. Qg3) Black can and should castle. White can and does play 11.Bh6, and it’s still an open question whether White’s initiative can be sustained after 11…Ne8 12. Rad1 Bd7, etc. In the version with Black’s pawn on h6, however, this line is obviously unavailable to Black, as after 13…0-0?? 14. Bxh6 gives White the initiative plus a pawn. The game between Carlsen and Grischuk ended in a draw:

Carlsen, Magnus vs. Grischuk, Alexander SF Blitz 5m+2spm 2016 | INT | Round 6 | 23 Aug 2016 | ECO: B90 | 1/2-1/2
1. e4 c5 2. Nf3 d6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 a6 6. Be3
6. Bc4 e6 7. Bb3 b5 8. O-O Be7 9. Qf3 Qc7 10. Qg3 O-O 11. Bh6 Ne8 12. Rad1 Bd7 is one of the main lines in the 6.Bc4 system.  )
6... Ng4 7. Bg5 h6 8. Bc1 Nf6 9. Bc4 e6 10. Bb3 Be7
10... b5 11. O-O Be7 12. Qf3 Qc7 13. Qg3 , in contrast to the similar line given in the note to White's 6th move, favors White, because the otherwise desirable
13... O-O?? loses the h-pawn in this version for no compensation at all.
14. Bxh6  )
11. O-O O-O 12. f4 b5 13. e5 dxe5 14. fxe5 Nh7 15. Qg4 Kh8 16. Be3 Bg5 17. Bf2 Nd7 18. h4 Be7 19. Qe4 Rb8 20. Nc6 Qe8 21. Nxb8 Nxb8 22. Rad1 Nd7 23. Ne2 Nc5 24. Bxc5 Bxc5+ 25. Kh1 Qe7 26. c3 Bb7 27. Qg4 f5 28. exf6 Nxf6 29. Qxe6 Qxe6 30. Bxe6 Re8 31. Nf4 g5 32. Bd5 Nxd5 33. Nxd5 Re4 34. Kh2 Bd6+ 35. Kg1 Bc5+ 36. Kh2 Bd6+ 37. Kg1 Bc5+ 38. Kh2

There are other examples in other openings, but it is clear that sometimes a seemingly pointless move that loses a tempo is in reality a brilliant way to obtain an otherwise unavailable benefit.


Dennis Monokroussos is a FIDE master who has written about chess on his blog “The Chess Mind,” since 2005. He has been teaching chess for almost 20 years and for the last 10 years has been making instructional chess videos, which can be found at Between 1995 and 2006, he taught philosophy, including a four-year stint at the University of Notre Dame.