Double Reversed Halloween Gambit

Forgive me for returning to 1.a3 and the Mengarini (1.e4 e5 2.a3) but you were warned in this previous entry.

1.a3 e5 2.e4 Nf6 3.Nc3 Nc6 4.Nf3 g6
This is a reversed version of Glek's Four Knight's line 1.e4 e5 2.Nf3 Nc6 3.Nc3 Nf6 4.g3!?.
5.Nxe5?! (Dia)

As you may or may not know the original Halloween Gambit arises after 1.e4 e5 2.Nf3 Nc6 3.Nc3 Nf6 4.Nxe5?! which probably is incorrect for several reasons. What is most relevant in our context is the fact that after 4...Nxe5 5.d4, Black's most popular - and probably best - move is 5...Ng6.

Therefore clever players discovered the Reversed Halloween Gambit (1.e4 e5 2.Nf3 Nc6 3.Nc3 Nf6 4.g3 (Glek) 4...Nxe4?!) which almost certainly improves over the original. Not only has the knight's best retreat option been eliminated, there is also a couple of kingside weaknesses which at least outweigh White's extra tempo.
The diagram position shows the Double Reversed Halloween gambit which in reality is the original Halloween Gambit with the extra moves a3 and ...g6. Surprisingly it's almost as popular as the more normal looking 5.d4 exd4 6.Nxd4 Bg7 7.Nxc6 bxc6 with a relatively balanced position.
5...Nxe5 6.d4 Nc6
As already mentioned this is a less attractive square than g6.
7.d5 (Dia)

This is more promising than 7.e5 which was tried in one of the first games with the variation.
7...Nb8
Here we see another small point of 1.a3 - the knight cannot go to b4 (as in the parallel line 1.e4 e5 2.Nf3 Nc6 3.Nc3 Nf6 4.Nxe5 Nxe5 5.d4 Nc6 6.d5 Nb4). Black's other options are:
a) 7...Ne5 8.f4 Neg4 9.h3 looks promising for White (9.e5 Bc5!? is complicated).
b) 7...Bg7! 8.dxc6 bxc6 is the sensible man's move, transposing to the line 5.d4 exd4 6.Nxd4 Bg7 7.Nxc6 bxc6 mentioned above.
8.e5 Ng8 9.d6 (Dia)


This can be considered the critical position. White obviously has a certain amount of compensation for his piece.
9...cxd6
Black opens the centre. G.Jones-Bae, West Bromwich 2005 continued 9...c6 10.Bc4 Qh4 11.Qe2 Bh6 12.g3 Qh5 13.g4 Qh3 14.Bxh6 Nxh6 15.Ne4 Kd8 16.Nf6 b6 17.0–0–0 Qh4 18.Rhg1 Qg5+ 19.Kb1 Ng8 20.f4 Qxf4 21.Rgf1 Qg5 22.Bxf7 Ba6 23.c4 Nxf6 24.exf6 Kc8 25.Bg8 Rxg8 26.f7 Qd8 27.fxg8Q Qxg8 28.Qe7 Bxc4 29.Rf8+ Kb7 30.Qd8 1–0 .
10.exd6 Qf6 11.Nb5 Na6 12.Bc4 Bh6 13.Qe2+ Kf8 14.Be3 Bxe3 15.fxe3 Qh4+ 16.g3 Qh5 17.Qf2 Qf5 18.Qe2 Qh5 19.Qf2 Qf5 20.Qe2 Nh6 21.Rf1 Qh5 22.Qf2 Qf5 23.Qe2 Qe5 24.0–0–0 Kg7 25.Rd5 Qe8 26.Qd2 Rf8 27.Qd4+ f6 28.g4 b6 29.g5 Nf5 30.gxf6+ Rxf6 31.Rdxf5 gxf5 32.Rg1+ Qg6 33.Rxg6+ Kxg6 34.Bd5 Rb8 35.e4 Bb7 36.exf5+ Rxf5 37.Qg4+ Kf6 38.Qh4+ Kg6 39.Bc4 Nc5 40.b4 Ne4 41.Nc7 Nxd6 42.Qg3+ 1–0 Carlsen-Nyysti, Helsinki 2002.

I seem to remember an article about this variation in the Norwegian chess magazine 'Norsk Sjakkblad' by Stokke and Hjortås a few years ago. I will return with the year and issue number and possibly some more analysis whenever I find the relevant issue.

Happy Halloween!

Addendum November 27th
From Stefan Bücker, the editor of Kaissiber, I have received some additional information for those interested in this variation:
The game Carlsen - Nyysti, Helsinki 2002, was published in Kaissiber #20, page 34 (source: Suomen Shakki 2002), in the historical introduction to Maurits Wind's extensive analysis of the Halloween Gambit (pp. 22-51 of that issue). The article starts with Bücker's historical overview, showing that the 4.Nxe5 gambit was invented in 1873 (or earlier) by Dr. Carl Theodor Göring, who is better known as the inventor of the Goering Gambit. Later issues of Kaissiber contained refined analyses on the Halloween Gambit by Maurits Wind.

I can warmly recommend the magazine Kaissiber to anyone who can read German and who is interested in chess history or unorthodox opening theory.

Why Not the Slav?

This is the second part of an answer to why I didn't follow up 'Win with the London System' with a book on the Slav (1.d4 d5 2.c4 c6).

The Slav and the London system are quite a popular pair of openings - frequently complemented by the Caro Kann against 1.e4. And if my first project had been on the Slav, quite likely the London would have appeared a natural continuation. However, with the London coming first, I didn't really consider following up with the Slav. This isn't quite as silly as it may sound:

While I enjoy playing a reversed opening with White (see my posts on 1.a3) - looking for a good way to use that extra tempo, I find it quite uncomfortable to go the opposite direction - taking an opening which I play with White and play it a tempo down. There certainly will be related lines but a tempo down they almost always are slightly worse.

In the case of the Slav compared to the London, this most clearly is illustrated with this position (1.d4 d5 2.c4 c6 3.Nf3 Nf6 4.Nc3):

a) 4...Bf5?! followed by ...e6 and a typical London position would be Black's preferred move if it was playable. However, very few strong players use this move regularly with Black - probably with good reason as White scores around 80% with 5.cxd5! cxd5 6.Qb3 (a database search hints that there may be more to this than is generally known so I may return to this subject).

b) 4...dxc4, which gives up Black's central foothold but wins some time for development, is Black's traditional main continuation. After 5.a4 Bf5 the position has some resemblance of a reversed London. However, the pawn exchange is generally considered undesirable and for London players mainly interesting as an example of how to handle the position if something goes wrong.

c) The Semi-Slav 4...e6, which locks in the light-squared bishop, can lead to very interesting positions if White plays ambitiously but isn't very similar to a London set-up.

d) 4...a6!? - the Chebanenko Slav - is a more recent try for Black. Now when Qb3 can be met by ...b5 or even the exotic ...Ra7, ...Bf5 becomes an option for Black in certain lines. This is a very interesting idea but seen in a London perspective not very attractive. While a3 frequently can be useful in London, it has quite low priority. Consequently Black is not just his normal tempo down - in many respects he is closer to two tempi behind a normal London line.

e) 4...Qb6!? has been played by many strong players, most recently by Kamsky. Black prepares ...Bf5 by defusing the Qb3 option but it must be admitted that this system is more solid than exciting and it seems that White has several paths to a small edge.

f) Black has other options like 4...Ne4, 4...Bg4 and 4...Nbd7 but none of them are very London-like.

The Reversed Kan Sicilian

There have been some months now since I had anything on 1.a3. I cannot promise this will be the last entry but right now I am not aware anything more of any importance missing.

1.a3 Nf6 2.e3 e5 3.c4

This position obviously can also arise from the move-orders 1.c4 e5 2.e3 Nf6 3.a3 or 1.e3 e5 2.c4 Nf6 3.a3.

3...d5!?

This enters an Open Sicilian position a move down and looks a bit too optimistic. However, the normal result of a correctly played Open Sicilian is a small advantage to White. It could well be that losing a tempo only leads to equality - which is an excellent result for Black.

4.cxd5 Nxd5 (Dia)

Now we have a Reversed Kan Sicilian on the board. With reversed openings it can sometimes be hard to say whether White has an extra move or Black has one less. But in this case it’s quite clear: White has got a free move. How can he best use it?

5.Nf3!?

a) I am confident that if this variation should ever become popular, 5.b4 would be one of the first moves to receive serious attention. So far I could find no practical examples.

b) 5.Qc2 is the main line in the comparable reversed position but could be too quiet to achieve anything against modest development by Black:

b1) 5...Bd6 6.Bd3 Qg5 7.Ne2 c6 8.Ng3 Bc7 9.0–0 Nd7 10.Nc3 N7f6 11.Nxd5 Nxd5 12.b4 h5 13.Bb2 Bh3 14.gxh3 h4 15.Be4 (15.Ba6) 15...hxg3 16.Bxd5 and White was clearly better in Rodriguez Lopez-Castillo Martinez, Mislata 2000.

b2) 5...Nc6 6.Nf3 a6 7.Nc3 Be7 8.Be2 0–0 9.h4 h6 10.b4 Be6 11.Bb2 Nxc3 12.Bxc3 Bf6 =+ Sattari-Cheparinov, Dos Hermanas Internet Blitz 2004.

5...Bd6

a) 5...Nc6 6.Nc3 (6.Bb5) 6...Be6 7.Qc2 a6 8.Be2 Be7 9.0–0 Qd7 10.b4 0–0 11.Bb2 Bf6 12.Na4 Qe8 13.b5 axb5 14.Bxb5 Bd7 (14...e4) 15.Nc5 Ncb4 16.axb4 Bxb5 17.Rxa8 Qxa8 18.Ra1 += W.Paulsen-Flechsig, Leipzig 1879.

b) 5...Nd7 6.Qc2 c6 7.Nc3 N5f6 8.Bc4 Bd6 9.b4 0–0 10.Bb2 Qe7 11.Ng5 Nb6 12.Ba2 g6 13.h4 Bf5 14.Nce4 Nxe4 15.Nxe4 Rad8 16.h5 Nd7 17.g4 Bxg4 18.hxg6 hxg6 19.Nc5 with a clear advantage to White in Schmittdiel-Medunova, Liechtenstein 1994.

6.Qc2

This is the normal Kan move but doesn't appear very threatening and may allow Black to reach an acceptable position with modest play. However, if Black plays modestly White's long-term advantages - in particular his central majority - may become an important factor. Alternatives include:

a) By parallel from the reversed lines, 6.Bc4 Nb6 7.Ba2 should be an important line.

b) 6.d3 Nc6 7.b4 a6 8.Bb2 Be6 9.Nbd2 0–0 10.Nc4 f6 11.Be2 Re8 12.0–0 Bf8 13.Rc1 Qd7 14.Qc2 Rad8 15.Rfe1 = Kunte-Suvrajit, Atul 2006.

c) 6.Nc3 Nxc3 7.bxc3 (7.dxc3 0–0 8.e4 a5 9.Bc4 a4 10.0–0 Bg4 11.h3 Bd7 12.Bg5 += Eichler-Baumgartner, Austria 2000) 7...0–0 8.d4 exd4 9.cxd4 c5 10.d5 Nd7 11.Bb2 Nf6 12.Bc4 a6 13.a4 Bf5 14.0–0 Rb8 15.a5 += Steinitz-Rosenthal, Vienna 1873.

d) 6.e4 Nb6 7.d4 exd4 8.Qxd4 0–0 9.Bd3 Nc6 10.Qc3 Re8 11.Bg5 f6 12.Be3 Ne5 =+ Symeonidis-Panagiotopoulos, Nikea 2004.

6...0–0 7.b4

White has also tried:
a) 7.d3 Re8 8.Nbd2 Bf8 9.Be2 a5 10.b3 a4 11.b4 c5 12.bxc5 Na6 13.Bb2 was very good for White in Bouhallel-Weemaes, Belgium 2003.

b) 7.Bc4 Be6 (7...Nb6 8.Ba2 Qe7 9.h4 N8d7 10.Ng5 Nf6 11.Nc3 Kh8 12.Ne2 e4 13.f3 Nbd7 14.fxe4 c5 15.d4 gave White the advantage in Bosboom-Ellenbroek, Enschede 1993) 8.Qb3 c6 9.Nc3 Nd7 10.d4 Rb8 11.e4 Nxc3 12.bxc3 Bg4 13.Be2 c5 14.d5 b5 =+ Pantsulaia-Dzagnidze, Istanbul 2006.

7...Nd7

Or 7...Bg4 8.Bb2 Nd7 9.Be2 Kh8 10.d3 f5 11.Nbd2 Qe7 12.h3 Bxf3 13.Bxf3 N7f6 = Bouhallel-Marchadour, Avoine 2006.

8.Bb2 Re8 9.Bc4 N7f6 10.d3 c6 11.Nbd2 Bg4 12.Ng5 h6 13.Nge4 Bf8 14.h3 Bh5 (Dia)

So far the play has looked rather calm and positional. Now White decides to use his active pieces, his central majority and his delayed castling in a kingside attack.
15.g4 Bg6 16.Rg1 Nxe4 17.dxe4 Qh4 18.Rg3 Rad8 19.Nf3 Qf6 20.h4

It's becoming clear that White's attack is very dangerous.

20...Qd6?

This fails for tactical reasons - the pin in the b1-h7 diagonal can be neutralized. After 20...Nb6 White's attack still seems promising but nothing is really decided.

21.h5 Bh7 22.g5 b5

Now 22...Nb6 is met decisively by 23.g6.

23.Bb3 hxg5 24.Nxg5 Be7 25.Nxh7 Nb6 26.Rg6 Qd3 27.Bxe5 Bxb4+ 28.axb4 Rxe5 29.Qb2 Kxh7 30.Qxe5 fxg6 31.hxg6+ Kh6 32.Rd1 Qxb3 33.Rxd8 Qxb4+ 34.Kf1 1–0 Kjartansson-Baldursson, Reykjavik 2006.

1.a3 - the Final Hurdle

It would be very surprising if 1.a3 were sufficient for a genuine opening advantage. But if White could expect comfortable equality in a position more familiar to himself than his opponent it would not be a bad opening option at all. To some extent I think I have demonstrated that this can be achieved against 1...e5, 1...g6, 1...c5 and 1...d5. Consequently the main challenge for White seems to be the flexible 1...Nf6:

1.a3 Nf6! 2.e3!? (Dia)

This is my best shot. It doesn't promise White an advantage but it may give unbalanced equality in a position where White can hope to be the best prepared. White has also tried:

a) Prie himself has declared that 2.d4 g6! leads to a King's Indian where the pawn move to a3 is mostly a waste of time.

b) 2.g3, planning a King's Indian set-up with Bg2, d3, Nd2, Nf3, 0-0 and e4 probably is sufficient for equality as a3 normally is vaguely useful in these set-ups.

c) 2.b4 may at first look consistent with White's first move. But when you look more closely at the position you realize that it could just as well have occured from the move-order 1.b4 Nf6 2.a3?! which looks rather pointless. A recent practical example went 2...d5 3.Bb2 g6 4.e3 Bg7 5.c4 c6 6.Qc2 0–0 7.Nf3 Bf5 8.d3 dxc4 9.Qxc4 Nbd7 10.Nbd2 Nb6 11.Qc2 Re8 12.h3 a5 = Suba-Moya Hernandez, Almeria 2006.

2...e5

a) After 2...e6, I quite like 3.Bc4 d5 4.Ba2 with an original position which I suppose is roughly equal.

b) 2...c5 3.b4 may have a little less bite than 1.a3 c5 2.b4 but still gives White fair chances for a central superiority. 3...b6 may be best.

c) 2...g6 3.c4 (3.b4 Bg7 4.Bb2 d6 5.c4 0–0 6.g3 c6 7.Bg2 Nbd7 8.Nf3 e5 9.d3 Qe7 10.Nbd2 Nh5 11.Qc2 f5 unclear Emelianov-Ozgibcev, Novokuznetsk 1999) 3...d6 4.Nc3 Bg7 5.Be2 c6 6.d3 a6 7.Bd2 Nbd7 8.Rc1 Rb8 9.Nf3 e5 10.0–0 0–0 11.Qc2 Ne8 12.d4 f5 unclear Patuzzo-Giordano, Lugano 2003.

d) 2...d5 of course is very sound, but it seems White can unbalance the play:

d1) 3.b4 e6 4.Bb2 leads to a position that frequently has occurred from the Sokolsky move-order 1.b4 d5 2.Bb2 e6 3.e3 Nf6 4.a3.

d2) 3.Nf3 e6 4.c4 Be7 and now White has tried:

d21) 5.b3 0–0 6.Bb2 c5 7.Be2 b6 8.0–0 Bb7 9.d3 Nc6 10.Nbd2 Ne8 11.d4 cxd4 12.Nxd4 Nxd4 13.Bxd4 dxc4 14.Nxc4 Nd6 15.Be5 Nxc4 16.Bxc4 Bf6 17.Qxd8 Rfxd8 18.Bxf6 gxf6 19.Rfd1 += Gelashvili-Miladinovic, Kavala 1999.

d22) 5.g4!? is the trademark move for the new 'no rules' generation of chess players. 5...c5 6.b4 Nxg4 7.Rg1 Nh6 8.bxc5 Bf6 9.d4 was unclear in Bosboom-Sonntag, Germany 2006.

d23) 5.b4 seems like the logical move. 5...0–0 6.Bb2 b6 7.Qc2 c5 8.bxc5 bxc5 9.Bd3 Nc6 10.cxd5 exd5 11.0–0 Bb7 12.Bxf6 Bxf6 13.Nc3 g6 = Kozul-Sosonko, Bled 1997.

d3) I like 3.f4!? even if Black has somewhat reduced White's options compared to the 1.a3 d5 2.f4 lines I discussed in a previous entry (e3 is unnecessary or even damaging in a reversed Dutch Leningrad).

d21) 3...g6 4.b4 is an interesting Bird/Dutch set-up which is not easy to achieve as Black. Note that a3 is not wasted as 1.f4 d5 2.b4? is refuted by 2...Qd6! (1.f4 Nf6 2.b4? Nd5! is a variation of the theme).

d22) Also after 3...c5, 4.b4 is an interesting option, e.g., 4...cxb4 5.axb4 Qb6?! 6.Nc3 Qxb4? 7.Bb5+ Bd7 8.Ba3 Qa5 9.Bxe7 and White's advantage is sizable.

3.c4
White is hoping for a reversed Paulsen/Kan Sicilian. Other possibilities include:
a) 3.d4 exd4 4.exd4 Be7 5.Nf3 0–0 6.Bd3 d5 7.0–0 Bg4 8.c3 Bh5 9.Bf4 Bg6 10.Ne5 Bxd3 11.Qxd3 Nbd7 = Filzmeier-Marjanovic, Nova Gorica 1999.

b) 3.b4 d6 4.Nf3 g6 5.h3 Bg7 6.Bb2 0–0 7.c4 Re8 8.Nc3 Nbd7 9.Be2 e4 10.Nh2 Ne5 =+ Basman-Kinlay, Bristol 1980.

c) I have a weak spot for 3.Bc4 d5 4.Ba2 which leads to a quite unique position. However, objectively I must admit that White is unlikely to achieve full equality.

3...c6 (Dia)

Probably this is best. It's known from the O'Kelly Sicilian (1.e4 c5 2.Nf3 a6) that ...a6 (here a3) is not particularly useful in set-ups with c3 (here ...c6).

a) 3...b6 4.Nc3 Bb7 5.Nf3 e4 6.Nd4 c5 7.Nf5 g6 8.Ng3 Bg7 9.Be2 0–0 10.0–0 ½–½ Galkin-Malaniuk, Sochi 1997.

b) 3...g6 4.Nc3 (4.b4 Bg7 5.Bb2 d6 6.d3 0–0 7.Nf3 Ng4 8.h3 Nh6 9.Qb3 a5 10.Nbd2 axb4 11.axb4 Rxa1+ 12.Bxa1 Be6 13.d4 exd4 14.Bxd4 Nc6 15.Bxg7 Kxg7 16.Be2 += Bosboom-Gulko, Wijk aan Zee 2001) 4...Bg7 5.Nf3 Nc6 6.Qc2 a5 7.Rb1 0–0 8.Be2 Re8 9.d3 d6 10.0–0 Bf5 11.Nd2 Rb8 ½–½ Hulak-Tkachiev, Istanbul 2003.

c) I was surprised to discover that there are quite a few games continuing 3...d5 4.cxd5 Nxd5. This seems quite accommodating as White is playing a Kan/Paulsen Sicilian a move up. It may tell something about the importance of one tempo in an unbalanced position that White in this line scores 61% (in 41 games) while Black in the corresponding position after 1.e4 c5 2.Nf3 e6 3.d4 cxd4 4.Nxd4 a6 scores 51% (in 39.435 games) - which is still an amazingly good result for Black. Maybe I will return to this in a later entry.

4.Nf3

Also 4.d4 seems reasonable: 4...exd4 (4...e4!?) and now:

a) 5.Qxd4 seems relatively safe (isn’t this somewhat reminiscent of the French line 1.e4 e6 2.d4 d5 3.Nd2 c5 4.exd5 Qxd5!?). A recent game went 5...d5 6.Nf3 Bd6 7.Nc3 dxc4 8.Bxc4 0–0 9.e4 Bc7 10.Qxd8 Rxd8 11.Ng5 Rf8 12.f4 h6 13.Nxf7!? Rxf7 14.e5 Nfd7 15.Be3 Nf8 16.Bxf7+ Kxf7 17.0–0–0 and it seemed that Black had slightly the better chances in a difficult position in Bosboom-Tiviakov, Hilversum 2007.

b) 5.exd4 would follow the parallel main line in the reversed 2.c3-Sicilian. After 5...d5, 6.Nf3 most likely would lead to an IQP position where a3 probably would be marginally useful. Also 6.c5, planning to meet 6...b6 with 7.b4 and a queenside space advantage makes sense.

4...e4 5.Nd4 d5
5...g6 6.Nc3 Bg7 7.Qc2 Qe7 8.b4 0–0 9.Bb2 d5 10.cxd5 cxd5 11.Be2 Nbd7 12.0–0 Ne5 unclear Kuligowski-Sosonko, Amsterdam 1982.

6.cxd5 cxd5 (Dia)

7.d3

a) I like 7.Qc2 here but it hardly changes the evaluation that White must thread carefully in order to keep the chances equal.

b) 7.b4 seems consistent but is hardly sufficient for equality:

b1) 7...Nc6 8.Bb2 (8.Nxc6 bxc6 9.Qc2 Bd7 10.d3 exd3 11.Bxd3 Bd6 =+ Haapasalo-Rantanen, Salo 1998) 8...Bd6 9.Be2 0–0 10.f4 Bd7 11.0–0 Rc8 12.Nb3= Talon-Dal Borgo, Belgium 2002.

b2) 7...a5 8.b5 Bd6 9.Be2 0–0 10.Bb2 Nbd7 11.f4?! (after 11.Nc3 White is close to equality in an unbalanced position) 11...Nb6 =+ Talon-Van den Brande, Westerlo 2004.

7...Bc5

Or 7...a6 8.Nc3 Bd6 9.dxe4 dxe4 (Bosboom-Van Wely, Wijk aan Zee 1999) 10.Qc2 +=.

8.Nc3 0–0 9.Be2 Qe7 10.0–0 Rd8 11.b4 Bxd4 12.exd4 Nc6 13.Be3 Bf5
= Milov-Godena, Cannes 2006.

Ask Dr. Johnsen

Today 'anonymous' left this message under my entry 'The Smallest Repertoire':

Q: I am due to play against a much stronger player in an on-line tournament soon, and wondered if you had a recommendation on how to approach it? I can see that he uses the French (I'm playing White) which I could prepare against (I'm an e4 player), but should I do something more psychological? Assume that he feels confident against me and adopt an opening as White that he has to attack me to win? Is there such a thing as a reversed Pirc, as I really enjoy that opening?
Any ideas welcome....

I don't really pretend to be a chess guru. But with too little time (and energy) to play tournaments myself, it's tempting to give advise to those who actually do play. So here are my 50 cents:

First of all: It's fully possible to play a reversed Pirc. 1.g3 d5 2.Bg2 e5 3.d3 might be a good attempt but it really depends on which lines you favor. However, I wouldn't expect you to gain much by heading for a reversed Pirc. Quite likely your opponent will choose a line which is not considered to give (White) a lasting advantage but which gives (Black) easy equality when reversed.

If you are willing to invest some time and work in this game, my main recommendation is to head for the sharpest mainline. Most strong players hates to risk losing to a weaker player who happens to be better prepared. Consequently they are quite likely to chicken out with an inferior move in order to avoid prepared surprises. With some luck this will give you a clear opening advantage. If the difference in playing strength isn't too great this could offer a chance for a win (or a draw offer from a position of strength).
More concretely:
1) Find out as much as possible about your opponents preferences in the French. Does he play the Winawer? The Classical? Which subline(s) does he prefer? Does he tend to follow the mainlines or does he play some home-brewed mixture?
2) Find a good book about the line you are most likely to encounter. Concentrate on the sharpest mainlines (that's the most ambitious lines for White). You will not have time to prepare for all the ways he may deviate but at least you should know when he leaves your prepared line (and you have to start thinking rather than remembering).
3) During the game, try to give an impression that you are well prepared and looking forward to a theoretical duel.

If this approach should fail (he is after all a much stronger player), you at least are likely to learn something from your preparations. If theoretical mainlines are not to your taste, you have got two main options:

1. Try to complicate from move two, and hope he gets more confused that you. You most likely will lose, but there always is a chance that he blunders first. 1.e4 e6 2.b3 d5 3.Bb2 dxe4 4.Nc3 Nf6 5.g4 could be a try.
2. Play unashamedly for a draw. If he isn't too much stronger than you, he may worry if he really will be able to win a dead level ending and try to introduce some complications himself. That could backfire and give you an undeserved chance. 1.e4 e6 2.d4 d5 3.exd5 is the obvious (and banal) solution.

Option 1 has the added bonus that it will give you a reputation as a dangerous chess pirate. Yet I suspect that option 2 is most likely to bring success.

A Double Mengarini, Please

The Mengarini Opening to some extent is an attempt to turn the tables in the opening struggle. It comes in at least three different versions:

• 1.e4 e5 2.a3
• 1.e4 e5 2.Nc3 Nf6 3.a3
• 1.e4 e5 2.Nf3 Nc6 3.Nc3 Nf6 4.a3
  

Against each of these move-orders Black can resolutely turn the table back again with the reply ...a6. Instinctively one would think that these extra moves - which mainly appears to have defensive qualities - would favor Black who by nature is the defending part. However it's not that clear in practice. Actually it turns out to be quite hard to figure out how the extra moves influence the resulting double king-pawn positions:

1.e4 e5 2.a3 a6 (Dia)

3.Nf3

a) Sadly 3.f4!? with a Pseudo King's Gambit seems untested.

b) 3.Nc3 Nf6 4.Nf3 would transpose but the Vienna style moves 4.Bc4 or 4.f4!? may be worth a try.

c) 3.d4 exd4 4.Qxd4 Nc6 5.Qe3 d6 looks like a slight improvement over the well-known 2.d4 exd4 3.Qxd4 Nc6 4.Qe3 line of the Centre Game. However, in Hofstetter-S.Mueller, Kahl 1996 chances were equal after 6.Nc3 Nf6 7.Bd2 Be7 8.Bc4 Be6 9.Nd5 0–0 10.Ne2 Ne5 11.Nxf6+ Bxf6 12.Bxe6 fxe6.

3...Nc6 4.Nc3

a) 4.Bc4 b5 5.Ba2 Nf6 6.Ng5 d5 7.exd5 Nxd5 8.Nxf7 Kxf7 9.Qf3+ Ke6 10.Nc3 Nce7 11.d4 Qd6? (11...Bb7) and in Vasic-Ristovski, Portoroz 2003 White could have obtained a winning position with 12.Ne4.

b) 4.d4 exd4 and now:

b1) I really don't understand why nobody has tried 5.Nxd4 Nf6 6.Nxc6 bxc6 7.e5. In this line it's quite likely that Black's ...a6 will be to his disadvantage, as he after 7...Qe7 8.Qe2 doesn't have the sharp option ...Ba6.

b2) 5.c3 dxc3 6.Bc4 d6 7.0–0 Bg4 8.Qb3 Qd7 9.Ng5 (after 9.Bxf7+ Qxf7 10.Qxb7 Kd7 11.Qxa8 Bxf3 12.gxf3 Qxf3 Black at least has a perpetual check) 9...Bh5 10.Bxf7+ Bxf7 11.Nxf7 Na5 12.Qxc3 Qxf7 13.Qxa5 and White had a clear advantage in Zagrapan-Ilkovics, Slovakia 1998.

4...Nf6 (Dia)

One of the strengths as well as the limitations of the Three Knights Opening is the fact that it reduces the active options for both sides. With a3 and a6 eliminating all lines with Bb5 and ...Bb4 this becomes even more pronounced. 4...Bc5?! is met strongly by 5.Nxe5!

5.d4

5.g3 Nxe4?! is yet another version of the Halloween Gambit. I cannot even guess if the extra a-pawn moves favor any of the sides but a sensible solution is as usual to return the piece with 6.Nxe4 d5 7.Nc3 d4 8.Bg2 dxc3 9.bxc3 which would be a well-known theoretical position if it weren't for these extra a-pawn moves. But maybe the entire gambit is silly as Black after 8.Ne2 the doesn't have option of ...d3 followed by ...Nb4.

5...exd4 6.Nxd4 (Dia)

6...Bc5

a) 6...Nxd4 7.Qxd4 simply gives White the freer game and slightly the better chances, e.g: 7...d6 8.Be2 Be7 9.0–0 0–0 10.Bf4 Be6 11.Rad1 += Bucher-Knaus, Switzerland 2006.

b) 6...g6 7.Bg5 (7.Nxc6 bxc6 8.e5 Qe7 9.Qe2 Nd5 10.Ne4 Bg7 11.c4 +=) 7...h6 8.Bh4 Bg7?? (8...d6 9.Bxf6!? Qxf6 10.Nd5 Qxd4 11.Nxc7+ Kd8 12.Nxa8 Qxe4+ 13.Qe2) 9.Nxc6 bxc6 10.e5+- Gunsberg-Zukertort, London 1887.

c) 6...d6 7.f3 (7.Nxc6 bxc6 8.Bc4 Be7 9.0–0 +=) 7...g6 8.Be3 Bg7 9.Qd2 Qe7 10.0–0–0 Be6 11.g4 += Nanu-Chirpii, Eforie Nord 1999.

d) 6...d5 7.exd5 Nxd5 8.Nxc6 bxc6 9.Bc4 Qe7+ 10.Ne2 Qe4 11.Bxd5 Qxd5 (11...cxd5 12.0–0 +=) 12.0–0 Bd6 13.Bf4 += Bhend-Mottas, Pizol 1997.

7.Nxc6

a) After 7.Nb3 Ba7 only Black has any use of the extra a-pawn moves.

b) 7.Be3 may well be best.

b1) Instinctively 7...Bb6 looks less compact with the pawn on a6.

b2) After 7...Ba7 8.Nxc6 bxc6 9.Bxa7 Rxa7 Black's rook looks a little silly.

b3) 7...Nxd4 8.Bxd4 looks more comfortable for White: 8...Qe7 (after 8...Bxd4 9.Qxd4 d6 10.0–0–0 0–0 11.e5 Ng4 12.f4 Qh4 13.exd6 cxd6 14.g3 Qh5 15.h3 White was clearly better in Eichner-Feldmann, Germany 2006) 9.Be2 d6 10.Bxc5 dxc5 11.0–0 0–0 12.Qd3 += Pirttimaki-Nippula, Finland 1993.

7...bxc6 8.e5

This must be better than 8.Bc4 d6 9.h3 h6 10.0–0 0–0 = of E.Reppen-Garcia Serrano, Copenhagen 2004.

8...Qe7 9.Qe2 Nd5 10.Ne4 (Dia)

10...0–0?!

10...Bd4!? 11.f4 f5 12.c3 Ba7 is better and at first glance unclear.

11.c4 Nb6 12.Bg5

White's opening has been a success - White is at least somewhat better, Braeuning-X.Garcia, Barcelona 1997.

A 1.a3 Mystery

BigBase 2007 offers this mysterious game:

R.Piepjohn (1920) - G.Sax (2600), Naestved 1988

1.a3!?

Grandmasters are usually quite aware that they have to fight for every point - even against modest opposition. However, if Black in any way was inclined to underestimate his lowly rated opponent this move may have been a clever choice.
1...c5!? (Dia)
This seems a very sensible reaction to White's first move. Now the opening can be considered a reversed English.
2.g3

We will have a look at White's alternatives at the end of this entry.

2...g6 3.Bg2 Bg7 4.d3 d6

Black has also tried 4...Nc6 5.Nc3 d6 6.Nf3 Nf6 7.0–0 0–0 which looks fairly equal. Here White may have been a tad too optimistic when he initiated a kingside attack with 8.Nh4 e5 9.f4 exf4 10.gxf4 d5 11.f5. After 11...d4 12.Na4 Ng4 13.Qe1 Bf6 14.Nf3 Qe7 15.fxg6 hxg6 16.c4 Bf5 17.Rb1 Rae8 it was obvious that Black was better in Huettig-Klundt, Ditzingen 2000.

5.Nd2 Nc6 6.e3 e5 7.Rb1 a5 8.b3

Obviously White's success in this game was not directly related to his opening play.

8...Nge7 9.Ne2 0–0 10.0–0 Be6 (Dia)

The opening phase is more or less over, and despite Black's slight space advantage, the position looks roughly equal.

11.h3 Qd7 12.Kh2 f5 13.c4 g5 14.f4 g4 15.h4 Rad8 16.Qc2 b6 17.Bb2 Bf7 18.Rbd1 d5!?

This opens up the position somewhat but not really to Black's advantage.

19.fxe5 Nxe5 20.d4 cxd4 21.Bxd4 Qc7 22.c5 b5 23.Nc3 Be8 24.a4 b4 25.Ne2 N7g6 (Dia)

The position has become quite unbalanced. Now the armies engage in close combat and surprisingly it's the GM who fails in the calculation test.

26.Rxf5 Nxh4?!

Probably 26...Rxf5 27.Qxf5 Nxh4 is better. After 28.Bxe5 Qxe5 29.Qxe5 Bxe5 there are some long forcing lines and one of them goes 30.Nc4 Nxg2 31.Nxe5 Nxe3 32.Rd3 Nc2 33.Nxg4 d4 34.Nf6+ Kh8 35.Nxe8 Rxe8 36.Nxd4 Rd8 37.c6 Nxd4 38.c7 Rc8 39.Rxd4 Rxc7 =.

27.Rxf8+ Bxf8 28.Nf4 Nxg2 29.Kxg2 Qe7

Possibly 29...Bf7 planning to meet 30.Qf5 with 30...Re8 is better.

30.Qf5

Now White's advantage is obvious.

30...Bg7 31.Qe6+ Qf7?

After 31...Kf8 32.h4 White is objectively winning but there is still a fight and the 680 rating points might still have influenced the result.

32.Bxe5 1–0

It would indeed be a nice story if "super GM" Sax lost to an unknown amateur who opened the game 1.a3. At this time Sax was close to his peak as a player. So who was Reinhardt Piepjohn who so easily matched him? BigBase 2007 has only this single game by him and claims that it was played in the ninth and last round on October 4. But something isn't quite right as the tournament seems to have been played July 30th to August 7th as this Danish site shows. The tournament rating favorite Sax didn't do too badly and ended well ahead of Piepjohn. Can it be that the round number as well as the result is wrong? Or was Black a different player? Any information is appreciated!

And now let's return to some alternatives to 2.g3:

1.a3 c5 2.b4!? (Dia)

This looks logical and strong as White will now get a numerical superiority in the centre. However, I remember seeing 1...c5 recommended as a good antidote to 1.b4 and I don't believe 2.a3 is White's best try in that position. Other options are:

a) 2.Nf3 is a flexible and good move but I have not really found any lines that make sense of White's first move.

b) 2.e4 leads to an Anti-Sicilian line which has recently gotten quite a lot of attention. The main source of information is no doubt 'Challenging the Sicilian with 2.a3!?' - a 206 pages work by Alexei Bezgodov. However, for historical information you should also read Hans Ree's article for 'The Chess Cafe'. I assume that if you want to reach this position, 1.e4 is your best bet.

c) 2.c4 too makes sense as queenside expansion with a3, Rb1 and b4 is a common plan for White in the symmetrical English. Here are two game fragments that actually started with our move-order.

c1) 2...Nf6 3.Nc3 d5 4.cxd5 Nxd5 5.g3 Nc6 6.Bg2 e6 7.Nf3 Be7 8.0–0 0–0 9.Nxd5 exd5 10.d4 c4 unclear Andrejic-Andrijevic, Obrenovac 2004.

c2) 2...g6 3.b4 Bg7 4.Ra2 d6 5.Nf3 Nf6 6.e3 0–0 7.Be2 Nc6 8.b5 Na5 unclear Ambrus-Balashov, St Petersburg 2001.

d) 2.d3 doesn't have much independent significance compared to 2.Nf3 or 2.g3. After 2...d5 3.Nf3 Nc6 4.Nbd2 Nf6 5.e4 Bg4 6.Be2 Qc7 7.c3 Rd8 8.Qc2 e6 9.0–0 Be7 chances were balanced in Markus-Buljovcic, Subotica 2001.

e) 2.e3 too is relatively non-descript. One example is 2...b6 3.Nf3 Bb7 4.d4 e6 5.Be2 Nf6 6.0–0 Be7 7.b3 0–0 8.Bb2 cxd4 9.exd4 Nc6 with equal chances in Z.Markovic-Perunovic, Novi Sad 2000.

2...cxb4!?

This strengthens White's central influence and opens the a-file for his rook. Still it is probably critical as it exposes White's b-pawn to attack.

a) 2...e6 3.Bb2 Nf6 4.bxc5 Bxc5 5.e3 Nc6 6.Nf3 0–0 7.c4 d5 8.d4 Bd6 9.Nbd2 Qe7 10.Be2 e5 11.cxd5 Nxd5 12.Ne4 exd4 13.Nxd6 Qxd6 14.Nxd4 += Kulicov-Stiri, Athens 2006.

b) 2...e5 3.Bb2 (3.bxc5 Bxc5 4.Bb2?? Qb6 –+ Monastyrev-V.Karpov, Tomsk 1999) 3...e4 4.e3 Nf6 5.bxc5 Bxc5 6.Bxf6 Qxf6 7.Nc3 Qe5 8.Bc4 Nc6 9.Nge2 b6 10.Bd5 f5 11.0–0 Bd6 12.Ng3 Ba6 13.f4 Qf6 unclear Ermenkov-Adorjan, Riga 1981.

3.axb4 Qb6 (Dia)

4.Nc3!?

White cannot expect any advantage after 4.c3 or the strange looking 4.Ra4 d5 5.Nc3 e6 6.e3 Nf6 (6...Bxb4? 7.Bb5+ +-) 7.Nf3 Bd6 8.Nb5 Be7 9.Bb2 Nc6 10.Ne5 0–0 = Bettman-Wirschell, Soest 2000.

4...Qxb4!?

This is untested but must be critical. The sensible 4...e6 5.b5 d5 6.e3 Nf6 7.Nf3 Bd6 8.Bb2 Nbd7 9.Be2 0–0 10.0–0 e5 11.Ba3 Bxa3 12.Rxa3 Re8 lead to roughly equal chances in Frank-Bokelbrink, Pinneberg 2002.

It's not at all easy to give a meaningful continuation from here but it seems obvious that White's superior development gives him some play for the pawn. One almost absurd line goes:

5.e4 Qb6 (Dia)

6.Bc4 Nf6!?

7...e6 looks safer.

7.e5 Qc6

Could 7...d5 8.Bb5+ Nfd7 9.Nxd5 Qd4 really be better? 10.Nc7+ Kd8 11.Nxa8 Qxe5+ 12.Ne2 Qxa1 13.0–0 isn't too convincing.

8.d3 Ng4!

8...a6 9.exf6 b5 10.fxe7 Bxe7 11.Nd5 bxc4 12.Nxe7 Qe6+ 13.Ne2 Qxe7 14.Ba3 looks winning for White.

9.Qxg4 d5

...and it seems like Black survives.

Reversed Dutch

A reversed Dutch? Isn't that the Bird's Opening (1.f4)?

Actually not. 1.f4 d5 is indeed a reversed Dutch but I have always been slightly skeptical about 1.f4 because of 1...Nf6! which avoids the weakening of the e5-square and at least temporarily stops e4. Correspondingly I find 1...d5 a slightly illogical reply which to some extent justifies White's first move. That's why I consider 1.a3!? d5 2.f4!? an interesting sequence of moves while 1.f4 d5 2.a3 appears rather meaningless.

But is there actually any advantage in having the extra a-pawn move in a reversed Dutch? There obviously are certain Dutch lines with an early ...Na6 that simply become impossible to play with a pawn already occupying that square. However, I assume that with a deep understanding of Dutch strategy and a little imagination it should be possible to steer for positions where a3 is useful. Here is an example where White succeeded wonderfully:

N.McDonald - B.Thipsay, Banwell mem London 2001

1.a3 d5 2.f4 Nf6

Quite possibly Black should immediately ensure control of the long diagonal with 2...g6.

3.Nf3 g6 (Dia)

Black has chosen a sensible set-up that can be completed without taking on too many obligations in the centre.

4.g3

With this move White heads for a reversed Leningrad system. Actually 4.b4!? - to some extent preventing ...c5 - seems sensible too. One of Black's main problems in the Dutch is to complete the development of his queen-side. With that problem solved, this reversed version cannot be too bad.

4...Bg7 5.Bg2 0–0 6.0–0 c5

Finally Black accepts his role as White! Probably 6...c6 would have given comfortable equality.

7.d3 Nc6 8.Nc3!

Probably this is White's best chance to make something out of his pawn on a3. From a comparison with the parallel position with reversed colours it seems that 8.Qe1 and 8.c3 are sensible alternatives.

8...d4 9.Na4! (Dia)

After 1.d4 f5 2.Nf3 Nf6 3.g3 g6 4.Bg2 Bg7 5.0–0 0–0 6.c4 d6 7.Nc3 Nc6 8.d5, 8...Ne5 is a little more popular than 8...Na5 but odds are much better that White can make use of a3 in this line.

9...Qd6 10.c4 Ng4

In the reversed position the main continuation is 10.b3 a6. However, 10...b6 could nevertheless have been considered as White's score in the reversed line is 59%. Only experience can tell how much difference one tempo can make but in a sharp position where both players pursue different plans, it may well make the difference between a win and a loss.

11.Rb1

Philosophically White can claim some success as the two main continuations in the parallel reversed position are 11...Rb8 and 11...a6. What's more: in the 28 games with these two moves, the score is 9 wins for White, 7 draws and 12 wins for Black!

11...Rb8

11...e5 seems a more consistent follow-up of Black's 10th move.

12.b4

A relevant game for comparison is Benko-Tal, Candidates Tournament Bled/Zagreb/Belgrade 1959: 1.Nf3 f5 2.g3 Nf6 3.Bg2 g6 4.c4 Bg7 5.Nc3 0–0 6.0–0 d6 7.d4 Nc6 8.d5 Na5 9.Qd3 c5 10.Ng5 a6 11.Rb1 Rb8 12.Bd2 Qe8 13.b3 b5 14.a3 Ng4 15.Nf3 bxc4 16.bxc4 Rb3 17.Rxb3 Nxb3 18.Rb1 Nd4 19.e3 Nxf3+ 20.Bxf3 Ne5 21.Qe2 Nxf3+ 22.Qxf3 e5 23.Qd1 e4 24.Qa4 Qe7 25.Qc6 f4!? 26.Rb8 Bh3 27.Rxf8+ Qxf8 28.exf4 Qb8! 29.Ne2 Qb1+ 0–1.

12...b6 13.bxc5 bxc5 14.Rb5 Rxb5 15.cxb5 Nd8 16.Qc2 Ne6 17.f5 gxf5 18.Nh4 Bf6

Fritz suggests 18...Nc7 with roughly equal chances.

19.Nxf5 (Dia)

The pawn structure is quite murky but White's advantage is not in doubt as his piece activity is superior.

19...Qe5 20.b6 axb6 21.Nxb6 Ng7 22.Nc4 Qc7 23.Bf4 Qd8 24.Nh6+ Nxh6 25.Bxh6 e5 26.a4 Re8 27.a5 Ne6?

27...Ba6 would have been a better try.

28.Bc6!

This wins an exchange thanks to the hanging bishop on f6.

28...Bg7 29.Bxe8 Bxh6 30.Bxf7+ Kg7 31.Bxe6 Bxe6 32.a6 Qa8 33.Qb1 Qxa6 34.Qb8 Bxc4 35.Qf8+ Kg6 36.Qf5+ Kg7 37.Qxe5+ Kg8 38.Qe8+ Kg7 39.Qf8+ Kg6 40.Qf5+ Kg7 41.dxc4 1–0

The Prie System

To me the main problem with starting a game 1.a3 always has seemed to be the closed defences, 1...d5 and 1...Nf6. That is still so, but lately a new system has appeared that is very relevant for 1.a3 players - the Prie System which goes 1.d4 d5 2.a3 (Dia) and may at first seem mysterious. The move is not new - it was first used (repeatedly) by Somacarana in the eighteen-fifties - but it seems Prie deserves the naming rights as he is the first top player not only playing it successfully but also writing about it.


I believe Prie's reasoning goes roughly like this:
1) 2.c4 is White's best move but some lines demand quite a lot of study.
2) Systems with an early bishop development to f4 or g5 fail to gain an advantage and possibly even to achieve equality because of lines with an early ...c5 and ...Qb6.
3) 2.a3 to some extent discourages 2...c5 and a3 is a useful 'extra' move against most reversed Queen's Pawn openings.
4) a3 is useful in a few of White's normal Queen's Gambit lines, so White may well consider a delayed c4 should Black play too passively.
5) The possible switching between black and white strategies and speculating in which lines a3 is useful, irrelevant or even damaging can be quite confusing.

In short the Prie System may not lead to any objective advantage to White but it leads to 'normal looking' positions where the strongest player is likely to win. Here is a game where Prie definitely was the stronger:

Prie - D.Adams, British Cht 2005
1.d4 d5 2.a3

This is Prie's preferred move-order. He also considers 1...e6 2.a3 a valid try but against 1...Nf6 he prefers 2.Nf3 because of 2.a3?! g6.

2...e6

2...c5 definitely is risky in view of 3.dxc5 when White may actually try to hang on to the pawn.

3.Nf3 Nf6 4.Bg5 Be7 5.e3 h6 6.Bf4 0–0 7.Nbd2 b6 8.Ne5 Bb7 9.Qf3 c5 10.c3 (Dia)

Now it looks like a London system where it's not easy for Black to make use of his extra tempo.

10...a5 11.h4 Nfd7 12.Bd3 Nc6 13.Rh3! Ndxe5 14.dxe5 Qc7 15.Qh5

Most likely White is winning already.

15...Ba6 16.Bc2 Rfd8 17.Rg3

Now it's definitely decided.

17...Bf8 18.Bxh6 Qxe5 19.Bg5 f5 20.Nf3 Qd6 21.Bxd8 Nxd8 22.Ng5 Be7 23.Qh7+ 1–0

The London is such an easy game to play!

Mengarini's Opening

My apologies to readers waiting for some serious analysis - I still have some things to say about 1.a3:

1.a3 e5 2.e4 (Dia)
I suspect this move will come as a surprise to many players. We now have a position that could as well arise from 1.e4 e5 2.a3, which at first seems rather meaningless in a set of openings where we have learned that rapid development is essential.
2...Nf6
This is natural and probably best. It is debatable whether the extra a-pawn move has any significance in the reversed King’s gambit arising after 2...f5!?. What is certain is that this is an unlikely move to encounter unless your opponent is a regular King’s Gambit player. One of the relatively few practical examples is 3.exf5 Nf6 4.Be2 (4.g4!?) 4...Bc5 5.Nf3 d6 6.d4 exd4 7.Nxd4 0–0 8.0–0 Bxd4 9.Qxd4 Bxf5 with fairly equal chances in Anbuhl-T.Kristiansen, Gausdal 1981.
3.Nc3
Now we enter the so-called Mengarini Opening (which normally would arise after 1.e4 e5 2.Nc3 Nf6 3.a3). It can be considered a reversed version of the Open Games where the Ruy Lopez obviously is ruled out. It seems unlikely that the extra move will make a reversed Latvian with 3.f4?! a tempting option.
3...Nc6
With 3...Bb4?? (a reversed Spanish) out of the question, one might expect 3...d5 (a reversed Scotch) to be Black’s best try, as Kasparov has made a case for this being White’s only serious alternative to the Spanish. However, it turns out that after 4.exd5 Nxd5 5.Qh5!, may be a major obstacle. In the parallel line 1.e4 e5 2.Nf3 Nc6 3.d4 exd4 4.Nxd4 Qh4, White has the promising pawn sacrifice 5.Nb5 (or possibly a delayed version with a later Nb5). As this obviously is not an option here, 5...Nc6 6.Bb5 may be quite unpleasant for Black, e.g. 6...Qd6 7.Ne4 Qe6 8.Nf3 Bd6 9.Nfg5 Qg6 10.Qf3 f5 11.Nxd6+ Qxd6 12.d4 0–0 13.Bc4 Nce7 14.Bd2 exd4 15.0–0–0 h6 16.Rhe1 with a clear advantage to White in Czarnota-Korosciel, Poraj 2003.
3...Bc5 4.Nf3 d6 obviously must be sound. One of the more high-powered games continued 5.h3 Nc6 6.d3 a6 7.g3 0–0 8.Bg2 Be6 9.0–0 h6 10.Kh2 d5 with equal chances, Wahls-Brueckner, German Cht 1990.
4.Nf3 d5 (Dia)
This probably is critical.
a) 4...Bc5 5.Nxe5! is good for White, and one of the basic ideas behind the Mengarini, e.g. 5...Nxe5 6.d4 Bd6 7.dxe5 Bxe5 and now:
a1) 8.Bd3 0–0 9.0–0 Re8 10.Ne2 d5 =+ Levitsky-Steinitz, Moscow 1896.
a2) 8.Nb5 a6 9.f4 axb5 (9...Bxb2 10.Bxb2 axb5 11.e5 Qe7 12.Bxb5) 10.fxe5 Nxe4 11.Qg4 d5 12.Qxg7 +/- Molander-Van Hoolandt, Gausdal 2002.
b) 4...a6 makes some sense; Black reclaims his right to be Black. The game Gunsberg-Zukertort, London 1887 casts some doubt about the playing level at the time: 5.d4 exd4 6.Nxd4 g6 7.Bg5 h6 8.Bh4 Bg7?? 9.Nxc6 bxc6 10.e5 g5 11.exf6 Qxf6 12.Bg3 when White was winning.
c) 4...d6 allows White to take the initiative in the centre with 5.d4 exd4 6.Nxd4 when Black seems unable to take advantage of White’s tempo-loss:
c1) 6...Nxd4 7.Qxd4 Be7 8.Bc4 0–0 9.0–0 Kh8 10.Bg5 Ng4 11.Bxe7 Qxe7 12.f4 (12.Nd5) 12...Be6 13.Be2 += Carlsen-Potapov, Peniscola 2002.
c2) 6...Be7 7.Be2 (7.Bc4 0–0 8.0–0 Re8 9.h3 Nxd4 10.Qxd4 Nd7 11.Nd5 Nb6 12.Nxe7+ Qxe7 = Gullaksen-Simonsen, Torshavn 2003) 7...0–0 8.Be3 Re8 9.Qd2 Nd7 10.0–0–0 Bf6 11.f4 Nb6 12.g4 Bd7 13.g5 Bxd4 14.Bxd4 Nxd4 15.Qxd4 += Motwani-Winants, Belgium 2001
d) 4...g6 is a reversed version of Glek’s Four Knight’s line. White’s most entertaining move is 5.Nxe5!?, to which I may return in a later entry. It has however limited theoretical significance as Black after 5...Nxe5 6.d4 Nc6 7.d5, can return the piece with 7...Bg7! 8.dxc6 bxc6 and reach exactly the same position as after 5.d4 exd4 6.Nxd4 Bg7 7.Nxc6 bxc6.

5.Bb5
Also 5.exd5 Nxd5 6.Bb5 Nxc3 7.bxc3 has been tried, but White has been unable to prove any advantage: 7...Bd6 8.d4 exd4 9.cxd4 0–0 10.0–0 Bg4
a) 11.c3 Qf6 12.h3 Bxf3 13.Qxf3 Qxf3 14.gxf3 Ne7 15.c4 Nf5 16.Be3 Nh4 17.f4 c6 =+ Bae-Porat, Port Erin 2003
b) 11.Be3 Ne7 12.h3 Bh5 13.Bd3 Nd5 14.c4 Nxe3 15.fxe3 c5 = Golubovic-Z.Szabo, Budapest 1995
5...d4
5...Nxe4 may be better:
a) 6.Qe2 Nxc3 7.Qxe5+ Qe7 8.dxc3 Bd7 1/2–1/2 Gullaksen-A.Moen, Stockholm 2004.
b) 6.Nxe5 Qf6 7.Nf3 Be6 8.Qe2 Nxc3 9.dxc3 Bd6 10.Bg5 Qg6 11.Bd3 Qh5 12.Bf5 Ne5 (12...0–0!) 13.Bxe6 fxe6 14.Nxe5 Qxe2+ 15.Kxe2 Bxe5 = Djurhuus-A.Moen, Gausdal 2002
6.Ne2
The position has become semi-closed. I doubt White has any advantage, but there should be plenty of opportunities to outplay a weaker opponent. One typical game is G.Welling-Reimer, Dinard 1986:
6...Qe7
6...Nxe4 7.d3 Nf6 8.Nxe5 Qd5 9.Bxc6+ bxc6 10.Nf3 c5 11.0–0 Bd6 12.c4 Qb7 13.b4 += Glek-Zaja, Austria 2005.
7.d3 Bd7 8.0–0 g6 9.c3 dxc3 10.bxc3 (Dia)

White is better developed, better co-ordinated and has more pawns in the centre. But with accurate play it is still possible that Black could hang on.
10...Nd8?! 11.a4 c5 12.d4 Bxb5 13.axb5 Bg7 14.dxe5 Ng4
14...Nxe4 15.Re1 0–0 16.Nf4 Nxc3 17.Qb3 is no better.
15.Bg5 Qc7 16.Nf4 Nxe5 17.Nd5 Nxf3+ 18.Qxf3 (Dia)



18...Qc8
Black's position is just too bad. 18...Qd7 19.Rfd1 and 18...Qd6 19.Rfd1 are just as hopeless.
19.Bf6 1–0

More Reversed Openings

It's time to expand on 1.a3 and the theme of reversed openings. Unless Black has prepared the frequently recommended 1...g6, it’s quite likely your opponent will have a short think after 1.a3. Most likely he will be considering how his various options will fit into his normal opening repertoire (reversed or not).

In some ways 1...e5 is the obvious attempt to swiftly punish White for his non-developing first move. After all it’s possible that White is a pure beginner who has no ideas about development or the centre. 2.b4 can hardly worry Black as it should not be considered a reversed St. George opening (1.e4 a6 2.d4 b5) but rather a toothless Sokolsky line (1.b4 e5 2.a3?!). Nevertheless there are a couple of reasons for Black to reject 1...e5:

1) If Black doesn’t usually open his white games with 1.e4, it may seem illogical to play that move with Black, when White even has a small extra move. We will see one of these reversed lines today. Another is the Mengarini opening 1.a3 e5 2.e4 (or 1.e4 e5 2.a3) to which I will return later.

2) Even if Black normally opens 1.e4 and believes he may steer the game into lines where White’s extra move is irrelevant or even damaging, he may be worried because he doesn’t usually meet 1.c4 with 1...e5 and now fears that he could be lured onto unknown territory with 2.c4 (even though few English players plays 2.a3 after 1.c4 e5, it is not at all uncommon for White to play a somewhat delayed a3).

Z.Plenkovic (2240) - M.Zufic (2370)
Goodbye Summer open Omis 2004
1.a3 e5 2.d4!?
This seems to be a speciality of Plenkovic, who in MegaBase 2006 scores an impressive 4/4 with it. It is one of several reversed approaches that makes sense as it may lead to quite sharp play if White (normally Black) so desires.
2...exd4 3.Qxd4 Nc6
One of Plenkovic wins went 3...Qf6 4.Qd3 b6?? 5.Qe4+ 1–0 Plenkovic-Bazant, Rabac 2004!
4.Qd3
Also 4.Qa4 may lead to sharp Scandinavian positions where the extra a-pawn move is a very useful bonus. Compare with the line 1.e4 d5 2.exd5 Qxd5 3.Nc3 Qa5 4.d4 Nf6 5.Nf3 Nc6?! which in practice has been scoring nicely for Black. But after 6.Bd2! Bg4?! 7.d5! Black's position is so difficult that he has started playing 6...a6 in order to continue developing quickly with ...Nc6, ...Bg4, ...Nf6 and ...0-0-0. With ...a6 already in place, the line could be quite attractive.
4...Nf6 5.Nf3 Be7
Here Black has also tried:
a) 5...h6 6.b4 a6 7.Bb2 d6 8.g3 Be7 9.Bg2 Be6 10.Nbd2 0–0 11.0–0 Qd7 12.Rfe1 Bf5 (12...Bh3 13.Bh1) 13.e4 += Plenkovic-Brkic, Split 2005
b) 5...d5 6.Nbd2 6...Be6 7.b4 Nd7 8.Bb2 f6 9.g3 Bd6 10.Bg2 Qe7 11.0–0 Nde5 12.Qb3 Qd7 13.Rad1 Ne7 14.Nxe5 fxe5 15.c4 += Plenkovic-Zvan, Pula 2005.
As can be seen from these examples, Plenkovic likes to play in positional style, with a kingside fianchetto in connection with a pawn expansion on the queenside, but also more primitive play with.Nc3 Bg5 and 8.0–0–0 should be possible.
6.g3 0–0 7.Bg2 d6 8.0–0 Bg4 9.Nbd2 Qd7 10.Re1 Rfe8 11.b4
White again has gone for his favourite set-up. Probably the position is roughly equal, but it is slightly imbalanced and probably feels more familiar to White than Black.
11...Bf5 12.e4 Bg6 13.Bb2 a6 14.Nh4 Ne5 15.Qb3 Qc6 16.Bd4 Qd7 17.Nxg6 Nxg6 18.Rad1
Probably White now has an edge thanks to his bishop-pair.
18...Rad8 19.Nc4 Qc8 20.Ne3 c5 21.Bc3 b5 22.Nf5 Qe6 23.Qb2 Ne5 24.f4 Nc4 25.Qa1 Bf8 26.Bh3!
This should be winning. If Black does nothing, White will play Kg2 and pick up material by a discovered attack on Black's king. But the tactical lines are complicated and White soon spoils his position.
26...Nxe4 27.Bxg7!? Qxf5!
A very good way to mix up things!
28.Bxf5?
Probably the right course would have been 28.Bxf8 Qf6 29.Qxf6 Nxf6 30.Be7 with a decisive advantage to White.
28...Bxg7 29.c3 Nxc3 30.Qc1 Ne2+?
After this check White again gains the upper hand. 30...Nxd1 31.Qxd1 Bd4+ 32.Kh1 Nxa3 33.Bxh7+ Kxh7 34.Qd3+ Kg7 35.Rxe8 Rxe8 36.Qxa3 Re1+ would have given Black a relatively clear advantage.
31.Rxe2 Rxe2 32.bxc5 Bb2 33.Qb1
The position still is very complicated, but I trust Fritz when he claims that White should come out on top.
33...Bxa3 34.c6 Bc5+ 35.Kh1 Bb6 36.Bd3 Rb2 37.Qc1 Be3
This is hopeless but also 37...Rb4 38.Qc3 Ne3 39.Re1 Nd5 40.Bxh7+ Kxh7 41.Qd3+ Kg7 42.Qxd5 is winning for White.
38.Qa1 1–0
It's possible that Black lost on time, but his position is lost. He must save his rook, but then the a-pawn falls and the white c-pawn decides.

Reversed or Not?

I always have had a special interest in what you (a little pretentiously) may call ‘existential opening questions’. That is opening questions that may influence your entire understanding of the game of chess.

One such theme is the idea of playing reversed opening systems. At a theoretical level, this may tell something about the nature of White's initial advantage. Several sharp opening lines for Black would be virtually winning if he only had one extra tempo. The practical problem, of course, is that if you try to achieve this by obvious means, Black will play a quieter set-up and assume that he will still be at least equal.

The game below to some extent illustrates these ideas:

R.Gerbery (2307) - J.Tajti (2147) Zemplen Cup-A, Sarospatak 2000
1. a3
This is sometimes called Anderssen's Opening and is one of White’s most obvious ways of heading for a reversed opening with an extra move. The move isn’t really weakening (although White may not want to combine it with long castling) and quite useful in several of Black’s most popular defences (in particular the Sicilian comes to mind, where ...a6 is a regular ingredient in most lines).
1...g6!?
I will return to Black’s other replies in a later entry. However, I have several places seen this move recommended as the best antidote to 1. a3, and indeed it appears logical. Black discourages 2.b4 while refusing to take on ‘white clothes’ as he would with 1...e5 or 1...d5. In most Pirc/King’s Indian systems a3 is a wasted (or at least harmless) move that contributes little to White’s queenside play.
2. e4!
In theory this position could also arise from the move-order 1.e4 g6 2.a3. I practice that rarely happens as White has many other tempting moves. The Modern Defence (1.e4 g6) is one of Black’s most challenging opening systems, but I am no great believer in its practical value. Black’s position tends to become somewhat cramped and with a few weaknesses. That may not be such a bad deal if you are genuinely stronger than your opponent and give priority to unbalancing the play. But against an opponent of equal strength I believe Black is handicapping himself. As a matter of fact I would be willing to meet 1.e4 g6 with 2.a3 on a regular basis if I knew that this would significantly increase my chances of meeting 1...g6!
2...Bg7
The obvious move. 2... c5 is logical too, and leads to positions that have lately arisen frequently after 1. e4 c5 2.a3 g6. White has reasonable chances to gain an advantage after 3. b4 Bg7 4. Nc3.
3. Bc4

3.Nc3 may be a slightly more flexible way to play for the same positions that arise in the game (knights before bishops). But if White is to have any use of his a-pawn move, it seems that the bishop should take the a2-g8-diagonal.
3...d6
3... c6 4. Nc3 probably would lead to a similar position.
4. Nc3 Nf6 5. d3 c6
It may appear that White after 5... c5 6.f4 would have an inferior version of a Sicilian Grand Prix attack, as White has spent a move on the quiet a3 in a position where he normally is in a hurry to attack on the kingside. But a3 is quite useful as it helps White keep his important light-square bishop and in addition there is a trade-off: a3 is slow, but so is ...d6, as ...e6 and ...d5 is an important counter-attacking idea against GP-Sicilians with Bc4.
6. f4
Now the position takes on distinct features from the Sicilian GP Attack, even with Black’s c-pawn still on c6.
6...b5 7. Ba2 Nbd7 8. Nf3 O-O

9. Qe2
In the similar Sicilian GP-attack positions, White usually plays 9. O-O followed by Qe1-h4 and f5. That might well have been a good plan here too.
9... Nb6 10. h3 a5 11. Be3 Ba6 12. Qf2 Nfd7 13. Rb1!?
The idea of this move would probably never have crossed my mind. But human minds work in different ways and that’s one thing that makes chess such a charming game.
13...b4 14. Ne2 Rb8 15. f5

Now the full force of White’s attack is obvious. The soon-to-be open f-file in combination with the diagonals a2-g8 and c1-h6 ensures White free access to Black’s kingside.
15...d5 16. fxg6 fxg6?
This recapture away from the centre weakens the e6-square. After 16... hxg6 White still has a promising kingside attack, but there is nothing concrete.
17. Ned4 Qc8 18. Ne6 Rf6 19. Nxg7 Kxg7 20. Qh4 e6?
This loses material. With 20... e5 21. O-O Qe8 Black could still have fought on from a difficult position. But without his dark-squared bishop it would probably be a hopeless task to defend his weakened king’s position in the long run.
21. Bd4 c5 22. Bxf6+ Nxf6 23. e5 Ng8 24. O-O c4 25. Ng5 Nh6 26. Rf7+!
This short combination decides immediately.
1-0