THE “PORTSMOUTH” GAMBIT (Part One)
An exploration of a relatively unknown sideline of the Sicilian Wing-Gambit
By FRED CLOUGH
The Sicilian Wing Gambit is as old as the Sicilian Defence itself, being briefly analysed in
Staunton’s Chess Players Handbook (1847) which in turn quotes earlier sources. Since then it
has been employed by a number of great attacking players including Spielman, Marshall, Keres
and Bronstein. The basic idea is to deflect the black c-pawn from the centre allowing White to
play d4, often in conjunction with rapid development of the pieces.
The Gambit is most usually played when Black plays 2…d6 but what about 2…Nc6. This
article represents my analysis of a little-known sideline – the so called “Portsmouth “Gambit
1.e4 c5 2.Nf3 Nc6 3.b4!? My attention was first drawn to this in November 1978 by an excellent
Guardian newspaper article by Leonard Barden. He had been so impressed by the successes
of a group of Portsmouth players that he named it the “Portsmouth” Gambit. About the same
time some German theorists (particularly Volker Hergert in Gambit Revue No.5) were also
recommending the same line. So what is different about this specific version of the Sicilian
Firstly, declining the gambit is unattractive due to the immediate threat of b5. But the main
advantage is that it sidesteps Black’s usual recommended defensive moves most normally
essayed in the 2…d6 system (that of 4…Nf6 followed by 5…d5!) In the Portsmouth variation,
4…Nf6 is met by 5.d5! hitting Black’s knight with advantage. Therefore, Black must find other
ways of dealing with this aggressive gambit. Incidentally, Black can now forget any hopes of
playing his favourite “Sveshnikov” or other related …e5 systems!
Of course, the Portsmouth Gambit also allows Black the alternative 3…Nxb4 somewhat in the
style of an Evans Gambit. Therefore we have two major variations (3…cxb4 and 3…Nxb4) ripe
for exploration. Whilst at least 200 games can be found from a computer database search (only
a handful being British) the Gambit is entirely missing from NCO and Nigel Davies’ recent
otherwise excellent book “Gambiteer 1” focuses entirely on the classical wing-gambit (2.b4).
Given this neglect, I present below my exploration of this fascinating gambit based largely
on my own OTB and CC games and incorporating my own original analysis (although using
Barden’s article as a starting point). This has been an exercise primarily for my own benefit, to
advance my own opening preparation and to satisfy my curiosity. But perhaps it has also opened
up a veritable “can of worms”!
1.e4 c5 2.Nf3 Nc6 3.b4!? This can be accepted in two ways by either 3…cxb4 or 3…Nxb4.
Firstly, I will consider the capture by the pawn
3…cxb4 4.d4 Black now has two main alternatives in 4…d5 and 4…e6. I consider 4…e6 to be the
critical test of the Portsmouth Gambit and this will be examined in Part Two of this article.
If 4...g6 Barden considered this weak on account of 5.d5 Bg7 6.c3 bxc3? 7.dxc6 c2 8.Qxc2 Bxa1 9.c7! winning. But Yilmaz - Kurajica (Kavala Zonal 1985) improved on Black’s play with 6…Ne5 7.Nd4 d6 8.Qa4+ (8.f4 is interesting 8…Nd7 9.cxb4 Ngf6 10.Bb2!? Nxe4 11.Ne6!!) 8…Nd7 9.Qxb4 Ngf6 10.f3 0-0=.
Whilst further research is needed, this line has much to recommend it to avoid some of the more risky variations to follow.
So, our first main line is…
1.e4 c5 2.Nf3 Nc6 3.b4 cxb4 4.d4 d5
The most common & natural response.
5.exd5 Qxd5 6.c4
(See Diagram 1)
Now we have a further split 6…bxc3ep
or…. 6...Qa5 7.d5 b3+ 8.Bd2 (8.Nbd2 can also be played here 8…Qc3 9.Ba3 b2 10.Bxb2 Qxb2 11.dxc6 bxc6 12.Bd3 Nf6 13.0-0 Here White has full compensation for the pawn with much better development and moves like Qa4 and Ne4 in the offing) 8…b2 9.Bxa5 bxa1Q 10.dxc6 b6 11.Bc3 Qxa2 12.Be2 Nf6 13.0-0 e6 (13...Bf5 is perhaps better) So far this is Thal – Granitzky East German Championship 1958 and “one of the few examples of this gambit in chess literature before the Portsmouth players took it up.” Barden.
14.Qd3! Be7 15.Ng5! a5 Black could have tried 15…Ba6 with the idea of giving back the exchange and castling
to safety. But after 16.Nd2 Rd8 17.Qe3 Rxd2 18.Bxd2 0-0 but Black is not safe at all19.Nxe6! fxe6 20.Qxe6+ Rf7 21.Bg5 Black is in all sorts of trouble with Rd1 to follow. If 15…h6 16.Bxf6 Bxf6 17.Nxf7!! Kxf7 18.Bh5+ Ke7 19.Rd1 Be5 20.c7! winning easily (Hergert-Geibig 1987).
16.Nd2 Qa3 17.Nge4 Nxe4 18.Nxe4 Bf8 19.Rd1 Qe7 20.Be5 Bd7 21.Nd6+ Qxd6 22.cxd7+ and Black resigned.
As Barden says, since the tactical refutation of White’s opening (6…Qa5) works out badly, Black should try 6…bxc3ep.
So, back to the main line.
1.e4 c5 2.Nf3 Nc6 3.b4 cxb4 4.d4 d5 5.exd5 Qxd5 6.c4 bxc3 7.Nxc3 Qa5
This is the most common variation of the Portsmouth Gambit that I have encountered in practice.
(See Diagram 2)
8.d5?! A popular option for White at all levels and very effective against the unprepared Black player, especially OTB. Now 8…Qxc3+ 9.Bd2 plays into White’s hands allowing dangerous attacking opportunities against Black’s exposed queen and king. A recent example is Clough – Grobler (CC171 pg 14). Instead 8…e6! 9.dxc6 (9.Bd2 transposes to the main line variant where 8.Bd2 e6 9.d5) 9… Bb4! 10.Bd2 Bxc3 regains the piece and the initiative. Attempts by White to defend with 11.Qc1, 11.Rc1 and 11.Be2 have fared badly in practice (Black wins seven out of eight gamesin my database) as White cannot avoid the exchange of pieces on d2 and c3 bringing his attacking prospects to an end. To his merit, Barden appreciated the essential problem with his (untried) suggestionof 10.Nd2 (to preserve the Bc1). Unfortunately, this can be refuted theoretically. E.g. 8...e6! 9.dxc6 Bb4 10.Nd2 Bxc3 11.cxb7 (11.Qf3 Bxd2+ 12.Bxd2 Qe5+ 13.Be2 Qxa1+ 14.Bd1 Qe5+ 15.Be2 Qd5 wins for Black.) 11...for Black.) 11...Bxb7 12.Qb3 Bxd2+ 13.Bxd2 Qe5+!! 14.Be2 Qxa1+ 15.Bd1 Qe5+ 16.Be2 Ne7 17.Qxb7 Rb8!! And Black wins.
So White must play 8.Bd2 rather than 8.d5?!
8...Bg4?! 9.d5! Bxf3 10.Qxf3 Ne5 11.Qe4 Nd7 12.Nb5 Qb6 13.Rc1 Ngf6 14.Qa4 Nxd5 15.Be3! with a winning attack (Cullinane– Thompson 1974). 8…e6 9.Nb5 Bb4 10.Nd6+ Ke7 11.Nc4 with advantage to White says Barden. However 11…Bxd2+ 12.Nfxd2 Qd5! Looks good for Black to me. So, after 8…e6 White could try 9.Qb3 Be7 (If 9...Bb4 10.Bb5 Nge7 11.0-0 0-0 12.Rfc1 unclear) 10.d5 exd5 (10…Nd8 11.Nb5 with an easy win. Felbecker – Burckert Corr 1961) 11.Nxd5 Qd8 12.Rd1 Nf6 13.Bc4 0-0 14.0-0 Nxd5 15.Bxd5 Bg4 16.Rfe1 Qc7 17.Re4 with a kingside attack.Or, after 8…e6 9.d5!? exd5 10.Bb5 Bb4 11.Nxd5 (If 11.0-0 Nge7 is fine for Black.) 11... Bxd2+ 12.Qxd2 Qxd2+ 13.Nxd2 Kf8 14.0-0 (14.Bxc6?! resulted in a win for White against his GM opponent (Bereza – Schleber 2004).But a safer continuation might be 14.0-0 Be6 15.Nc7 Rd8 16.Nxe6+ fxe6 17.Nf3) 14…Be6 15.Nc7 Rd8 16.Nxe6+ fxe6 17.Nf3 better for White. 9.Bb5 e6 10.d5 exd5 11.Nxd5 11.Qe2+ Be7 12.Nxd5 Qd8 13.Nxf6+ gxf14.0-0 Be6 15.Bc3 good for White (Clough– Finnie BCCA 1989). 11...Qd8 12.Nxf6+ Qxf6 13.0-0 “followed by Bg5 with active play for the pawn” Barden. My own game, Clough v Hammond Corr 1993, continued 13…Bg4 14.Qc1! Bxf3 15.Re1+ Kd7 16.Re3 Bd5 17.Rd3 with a winning attack. I therefore conclude that White has a viable gambit after Black’s 4…d5, although the sideline 8…e6 may be worth further analysisby players seeking better play for Black. It is now time to examine Black’s alternative capture 3…Nxb4. Leonard Barden says this is too timewastingfor Black. The natural moves are
4.c3 Nc6 5.d4
Should Black now exchange pawns? Or is it better to play 5...d5 denying White’s knight the typical & dangerous Nc3 attack on the Qd5. In the absence of theory, I have investigated these problems with my own analysis. Black has four likely continuations:- A/ 5...e6? Not to be recommended
6.d5! exd5 7.exd5 When Black probably has nothing better than 7...Nb8 leaving White with a significant
advantage. If 7...Nce7? 8.d6 Nf5 9.Qe2+ wins. B/ 5…d6 is an important option keeping the position closed. White can build up behind his pawn centre with 6.Nbd2, 6.Bc4, Bd3 or Bb5 with compensation of space and time. However, White’s c3 pawn can also obstruct the useful Bb2 development available in other lines. White could be better by simply responding 6.d5 Ne5 7.Nxe5 dxe5 8.Qh5! when Black has no satisfactory way of defending the e-pawn.
C/ 5...cxd4 6.cxd4 Now Black can transpose into either of the two main lines discussed under 3…cxb4.
C1/6...e6 7.Bd3 7.d5 allows Black to enter into a favourable version of the Exchange Sacrifice variation to be discussed in Part Two: 7...Qf6 8.dxc6 Qxa1.
7...d5 8.exd5 Qxd5 9.0-0 Followed by Nc3 gives White a good game.
Or… C2/ 6…d5 7.exd5 Qxd5 8.Nc3 Qa5 9.Bd2
We arrive at exactly the same position already discussed under 3...cxb4 (see above) where I believe White has a viable gambit. Thus my main focus here is on Black’s third
D/ 5…d5 The mainline continuation must undoubtedly be…
6.exd5 Qxd5 Where White has three further options.
D1/ 7.Be2 Boleslavsky’s suggestion, to be followed by 0-0 and c4. 7…cxd4! 8.cxd4 e6! 9.a3 Preventing ...Bb4
9...Nf6 10.0-0 Be7 11.Nc3 Qa5 12.Bd2 0-0 White has the initiative but probably not quite enough for the pawn.
D2/ 7.Nbd2 Once again Black has the same resource used previously against 7.Be2 viz 7…cxd4 and 8…e6! blockading d5. 7…cxd4! 8.cxd4 If 8.Bc4 Qa5 9.cxd4 Nf6 10.0-0 e6 and White has gained nothing for his gambit pawn except a lost position! (Clough – Burkett, West Wales League OTB 2003) This was a criticalgame for me and served as a “wake-up call” to save my favourite opening! 8...e6! Black is better.
If 8...Nxd4?! 9.Qa4+ Nc6 10.Bc4 with a strong attack for White. After deciding that neither 7.Be2 nor 7.Nbd2
provided White with sufficient compensation for the pawn, it was back to the drawing board to salvage the Portsmouth Gambit. And I found…
This is my saving idea. Since Black has prevented White’s thematic Nc3 attack on the Qd5, then 7.Na3 provides an alternative threat via b5. This also keeps open the diagonal of White’s dark-squared bishop (as opposed to the 7.Nbd2 variation) and gives possible additional support to the square d4 via Nb5 or Nc2. Transferring the knight to c4 is another possibility. Furthermore, it seizes some initiative with its immediate threatsof Nb5 and Bc4. I envisage two important continuations:-
Poor in view of …
8.Bc4 is also playable. After 8…Qe4+ 9.Be3!? Black has 9…dxe3, 9…Bg4 or 9…e5
with perhaps unnecessary complications from White’s point of view. The text is “simple” in
comparison. 8…Qd8 8...Kd8 9.Nbxd4± 8…Qe4+ 9.Be2 d3 10.Nc7+ wins the
9.Bf4 e5 10.Nxe5 Qe7 11.Qe2 d3 12.Nxd3! Qxe2+ 13.Bxe2 Kd8 14.Nc7 Rb8 15.Nd5 Ra8 16.0-0 Nf6 17.Nc7 Rb8 18.Rfd1 With a winning position for White.
Therefore the critical test of this line.
If 8.Bc4 Qe4+ (8...Bxf3 9.gxf3 Qh5 10.Rb1 b6?! 11.Bb5! Qd5 12.c4±) 9.Be3 0-0-0 (9…e6 10.0-0 cxd4 11.Nb5 0-0-0 with advantage to White Bosboom –Van den Bosch, Antwerp 1991) 10.0-0 cxd4 11.cxd4 e5 12.Bd3 Bxf3
13.gxf3 gives White open lines & attacking possibilities. But structural defects make this position ripe for further in-depth analysis.
Perhaps castling into insecurity but if 8...Nf6 9.Nb5 Rc8 10.c4 Qe4 11.d5 Ne5 12.Nc3 Qf5 13.Nxe5 Qxe5 14.Bb2 with a strong initiative for White.
10.Nxe5 Bxe2 11.Qxe2 Nxe5 12.dxe5
White has recovered the pawn with a strong position and can look forward to the rest of the game with confidence.
Some German commentators have concluded that accepting the variation via 3… Nxb4 is safer than 3…cxb4. However, I have tried to inject new life into White’s play with the idea of 7.Na3 and fully believe that White has a viable gambit here. It remains fertile ground for exploration by correspondence players.
To summarise: after both:-
1.e4 c5 2.Nf3 Nc6 3.b4 cxb4 4.d4 d5
1.e4 c5 2.Nf3 Nc6 3.b4 Nxb4 4.c3 Nc6 5.d4
I fully believe White has a good game and can look forward with confidence. As alluded to in the introduction, the critical test of the Portsmouth Gambit is
1.e4 c5 2.Nf3 Nc6 3.b4 cxb4 4.d4 e6!
which will be explored in Part Two of this article (next issue).
THE “PORTSMOUTH” GAMBIT (Part Two)
An exploration of a relatively unknown sideline of the Sicilian Wing-Gambit
By FRED CLOUGH
1.e4 c5 2.Nf3 Nc6 3.b4
In Part One we looked at both 3…cxb4 4.d4 d5 and 3…Nxb4. Now we look at Black’s critical
5th move response of 4…e6 where I feel White is compelled to pursue the “Exchange Sacrifice”
line of 5.d5!? The whole viability of the Portsmouth Gambit may stand or fall on this line so I
have devoted many hours of analysis to the extensive complications involved. However, there is
still plenty of scope for correspondence players to delve deeper and find new lines of play.
1.e4 c5 2.Nf3 Nc6 3.b4 cxb4 4.d4 e6 5.d5!?
5.Bd3 is the quieter alternative for White. It is by no means bad but Black retains the extra pawn leaving White to prove the compensation 5…d5 6.Nbd2 (6.exd5 Qxd5 7.0-0 Nf6 8.a3! This idea will be familiar to regular Winggambit
players Be7 9.axb4 Nxb4 10.c4 Qh5 11.Nc3 0-0 White has a strong initiative for the pawn and a very playable position) 6...dxe4 7.Nxe4 Be7 8.0-0 Nf6 9.Re1 (9.a3 is again possible here 0-0 10.axb4 Nxd4 11.Nxd4 Qxd4 12.Bg5 Qxb4?! 13.c3! and Black has to be careful) 9...0-0 10.Bb2 Nd5 when I personally prefer Black’s extra pawn but again 11.a3 is possible.
Now the rook on a1 is gone.
6...Ne5 Saverymuttu - Lees, Rhyl 1969, continued 7.Nd4 exd5 8.Nb5 Bc5 9.f4 Kd8 10.Qxd5 d6 11.Nd4 when “Black is in difficulties in the central files” (Barden). But neither is White’s king entirely comfortable after 11...Ng4 12.e5!? which looks very double-edged to me. As White I twice encountered an unusual twist in this variation. In 1996-7 we had the Portsmouth Gambit in our BCCA Gambit tourney thanks to Jonathan Tait, and two opponents each played 7...Qh4!? White must now lose his e-pawn and both games followed identical paths for the next six moves. 8.Be2 Qxe4 9.dxe6!? (9.0-0 is an alternative where White also stands well Qxd5 10.Nb5 Qc6 11.cxb4 Nf6 12.Bf4 d6 13.N1a3 Nd5 14.Bxe5 dxe5 15.Rc1 Qb6 16.Nc4 Qd8 17.Ne3 Bd6 18.Bg4 f5 19.Bxf5 exf5 20.Qxd5 Bb8 21.Qb3
and White is better) 9...Qxg2 10.exf7+ Kxf7 11.Rf1 Nf6 12.Bf4 Qe4 13.Bg3 bxc3 14.Nxc3 led to a win v McGowan 1997 and 14.f3 led to a draw v Eastwood 1996. But I dare not venture further analysis in these chaotic positions.
7.dxc6 c2! 8.Qxc2 Qxa1
This is the critical position where White must justify the loss of the exchange and a pawn. White has three possible plans:-
PLAN A: Contain and trap the black queen. PLAN B: Aim to drive the black queen (or allow its escape!) back along the diagonal to f6, gaining tempo in the process, and then keep it there out of play on the kingside whilst White
attacks in the centre and on the queenside. PLAN C: Forcing the black queen to e7 or d8 where it becomes vulnerable.
This is my recommendation. Other options are:- 9.e5 This is the most direct attempt to trap the black queen. Indeed, Barden claimed that “9.e5 followed by Bc4 and Bb2 wins”, but I think he may have underestimated Black tactical resources. 9.e5 Bb4+ 10.Kd1 Ne7 11.a3 (11.Bc4? Nd5 12.Bb2 Nc3+) 11...Nd5 12.axb4 (12.Bb2 Nc3+) 12...Nxb4 13.Qb3 Nxc6 14.Nc3 Qa5 15.Nb5 0-0 and Black has rook and three pawns for bishop and knight. Even better is where the black knight sets out to rescue the queen with 9...Ne7! 10.Bc4 Nxc6 11.Bb2 (11.a3 Nd4!) 11…Nb4! 12.Qc1 Nxa2! Black will have a large material advantage whether queens are swapped or not. 9.e5 Ne7 10.a3 Nxc6 11.Bb2 (11.Bc4 Nd4!) 11...Nb4 12.axb4 (12.Qc1 Na2 [Here a GM draw was agreed Yilmaz v Kouatly, Thessalonika 1984]
13.Bxa1Nxc1 14.Bc4 b5! Black stays the exchange plus one pawn ahead) 12...Bxb4+ 13.Kd1 Qa5 14.Bd3 b6 and Black has rook and three pawns plus dangerous counterplay for his two knights. In conclusion I feel Barden was wrong and that White will not relish any of these 9.e5 outcomes which fail to trap the black queen.
9.Qb3 This move also prepares to trap the black queen after the moves Bd3 & Bb2. It prevents Black’s disruptive ...Bb4+ and combines this plan with an immediate attack on b7. On the other hand the queen move from c2 to b3 requires an extra tempo and the queen may be exposed to attack from ...a5...a4. 9.Qb3 incidentally has a special significance since it can transpose into an identical position arising the ...Nxb4 variant. So 9.Qb3 bxc6 10.e5 a5 11.Bc4 a4 12.Qb6 Ne7 13.0-0 Nd5 14.Bxd5 cxd5 15.Bb2 Qxa2 16.Nc3 “followed by Nb5 and Nc7+ winning
(Barden)” But 16...Qc4! 17.Nb5 Bc5 18.Nc7+ Ke7 Should not the opposite evaluation apply? 19.Ba3 (19.Qb5 Qxb5 20.Nxb5) 19...Bxa3 20.Nxa8 and Black is winning readily. 9.Qb3 a5 was tried by Black (Stanton- Gallagher 1980’s) 10.cxb7? Bxb7 11.Bd3 Bb4+ 12.Ke2 Qf6! (if 11.Qxb7 Bb4+ 12.Kd1 Rd8 and Black creates active play for his queen)
An improvement for White here could be 9.Qb3 a5 10.Bd3 perhaps this makes more sense having the immediate deadly threat of Bb2. 10...Bb4+ (10...dxc6 11.Bb2) 11.Ke2 Qf6! 12.Bg5 Qg6 13.a3 (13.h4!? f6!) 13...Be7 14.Bxe7 Kxe7 15.cxb7 Rb8 16.bxc8N+ Rxc8 17.g3=
Here is a rare example of a Plan B type attack from master play (Bolding v Lerch, French Chp 2001) which shows the sharp tactics typical of this whole variation. White tried 9.cxd7+ Bxd7 10.Qb3 Rc8 11.Bd2 Qf6 12.Bg5 Qg6 13.Qxb7 f6!? 14.Bb5 Rd8 15.0-0 Be7 16.Nc3 Bxb5 17.Nxb5 fxg5 18.Nc7+ Kf8 19.Qb3 Rd6 20.Ne5 Qf6 21.Qb5 g6
22.f4 gxf4 23.Ne8 Rb6 24.Nd7+ Kf7 25.Ne5+ Kf8 26.Nd7+ Kf7 27.Ne5+ Drawn.I conclude that with 9.e5 or 9.Qb3 White achieves little from plans aimed primarily at trapping the black queen on a1. Where 9.e5 is played Black appears to hold his own with counterplay on the queenside: where 9.Qb3 is played Black may also confront White
on the queenside or simply escape to f6. In particular, Black’s ...Bb4+ can also have a strong disruptive effect. I have therefore turned to 9.Bd3 as a better solution to these
problems. After 9.Bd3 a number of options are available to Black, but only two are playable.
A/ 9…bxc6 or B/ 9…Qf6
9…Nf6?? loses to 10.Qb3 and 11.Bb2 9…dxc6 loses as White can revert to the queen entrapment idea 10.e5! Ne7 11.Qb3 a5 12.Bb2 a4 13.Qc2! Qxa2 14.Bc4 wins the queen.
9...Bb4+?! The problem with this move is
that White can generate multiple threats - to
the Bb4, to g7, to b7 and to the queen. 10.Ke2
bxc6 11.Qb3 Rb8 12.Bb2 Bd6 13.Qxb8 wins.
If 9…Bb4 10.Ke2 Ne7 11.Qb3 Nxc6 12.Bb2
Na5 13.Qxb4 Qxa2 14.Qc3 wins as Nbd2 &
Ra1 will follow;
If 9…Bb4 10.Ke2 Qf6 11.Bb2 Qg6 (11...Qe7
12.Bxg7) 12.Qa4! Be7 13.Ne5 Qh5+ 14.g4
Qh6 15.cxd7+ wins.
9…Ne7 is dubious due to 10.cxb7 Bxb7
11.Qb3 Nc6 12.Bb2 Na5 13.Bxa1 Nxb3
14.axb3 where White has two pieces for rook
and pawn and a more advanced development.
A/ 9…bxc6 10.0-0 Qf6
Black escapes before White’s Nc3 and Bb2.
11.e5 immediately may be a better option.
11…Qd8 12.Nbd2 Rb8 13.Ne4 f5 What else?
Black is in a stranglehold and must get some
space 14.exf6 Nxf6 15.Neg5! and Black is in
trouble. For example if 15…d5? 16.Bf4! Ra8
17.Bg6+ hxg6 18.Qxg6+ and it’s all over.
Black has gained another pawn but White
has increased his development advantage and
will soon threaten e5 & Ne4 as shown in the
above note after 11.e5. The attacking prospects
should suit the gambiteer and there remains
plenty of scope for individual analysis here.
I regard 9…Qf6 as the critical move posing
win or lose choices to White.
Plan A (trapping the black queen on a1) is
now redundant and Plan B (driving the queen
out to the kingside) is underway.
White holds back playing e5 straight away
(Plan C) for the moment since one idea could
be to attack with cxd7+ followed by Ne5
sometimes sacrificing the bishop on g5 in the
If 10.e5 Bb4+! Black’s saving move 11.Ke2
Qe7 (threatens…Qc5!) 12.a3 Bc5 and Black’s
material advantage remains intact. If 13.c7
then 13…d5 leaves Black in an excellent
10.Bb2 I first thought this was the best move
here but Black can safely retreat with 10…Qe7
and if then 11.Ba3 then 11…Qf6 back again.
White is not looking for a draw here so I
favour 10.Bg5 forcing the queen to g6.
10…Qg6 11.cxd7+ Bxd7
Another position which is critical for the
viability of this whole line.
The attack on d7 has begun and, whilst
White would like to further his development
with Nc3 and/or 0-0, a greater priority is to
maintain the initiative and frustrate Black’s
Targets b7 and allows the possibility
of following up with Bb5 and Ne5 with
tremendous pressure against d7. The move
also prevents …Bb4+. A possible drawback
with this plan is the possibility of …Qxe4+
(when White relinquishes the defence of e4)
and …Qxg2 whilst White remains uncastled
and we will examine these options.
12.Qc7 is tempting in view of such trappy
possibilities as 12…Bc6 13.Qxb7!? Bxb7??
14.Bb5! which mates. However, 12…Bb4+!
thwarts White’s plans and should allow Black
at least a draw.
I have examined five responses for Black
after 12.Qb3. Whilst I have carried out quite
detailed analysis of possible continuations,
it is now time to be more selective; firstly,
this is because of space limitations; secondly
because it is totally unsatisfactory to rely
upon theoretical speculation in the absence of
empirical evidence from master tournament
practice, or even my own game experiences.
I shall therefore limit myself to providing a
few illustrative examples of typical positional
idea’s and tactical opportunities.
An unsuccessful attempt to defend along
the sixth rank. 13.Nxe5 Qe6 (13…Qh5
14.Be2! Qxg5 15.Qxf7+ and mate to follow)
14.Qxb7 (If 14.Bc4?! Bb4+! 15.Qxb4 Qxe5)
Bb4+ 15.Ke2 Rc8 16.Bb5 Bxb5 17.Qxb5+
Kf8 18.Qxb4+ Ne7 19.Bxe7+ Qxe7 20.Qxe7+
leaves White wining easily.
This does not offer an adequate defence
in view of 13.e5! putting Black’s queen
offside. The rook does nothing to protect the
vulnerable d7 and c6 squares. For example:
13.e5! Qh5 14.Bb5! Be7 (If 14…Bxb5?
Qxb5 mate) 15.Bxd7+ Kxd7 16.Qb5+! Kc7
17.0-0 Rd8 18.Rc1+ Kb8 19.Qc4! and White
This appears to be Black’s best reply. It
looks weakening but it not only frightens the
Bg5, it provides an escape square for king or
queen on f7, and protects the e5 square. White
may now not be able to claim back his lost
material but his initiative is maintained. Lots
of scope for investigative analysis here!
A reliable response. 13…fxg5? loses to 14.Ne5 Qh5 15.Nxd7 Kxd7 16.Qxb7+. White must choose between all-out attack (14.Qxb7) and patient manoevreing (14.Be3) and the outcome of both options remains unclear. For example:- 14.Qxb7 Rc8 15.Rd1 Bc6 16.Bb5 Bxb5 17.Qxb5+ Kf7 18.Nbd2 White remains the exchange down but has a continuing initiative. 14.Be3 Nc6! (14…Bc6? is bad on account of 15.Nd4! Bxe4 16.Bb5+ Bc6 (16…Kf7 17.Qxe6 mate!) 17.Bxc6+ bxc6 18.Nxe6 and White is much better) 15.Rd1 Be7 16.Qxb7 Rb8 17.Qc7 again unclear.Whilst 13…Ne7 is a reliable response for Black, he has other options. E.g. 13…Bc5. This looks dubious on account of White’s attacking possibilities. A plausible continuation could be 14.Bb5! Qf7 15.Rd1 Rd8 16.Ne5! fxe5 17.Bxd8 Bxf2+ 18.Kh1 Bd4 19.Bc7 Nf6 20.Bxd7+ Nxd7 21.Bd6! The whole 12…f6 line is ripe for further analysis from both sides.
d/ 12…Be7 13.e5!
Looks best as it regains all White’s material deficit. 13…Qh5 14.Qxb7 Rd8 15.Bd2!
This calm re-routing of the bishop seems to leave White better.
15…Nh6 16.Ba5! 0-0
16…Rc8 17.0-0 and Black cannot castle without losing a piece.
17.Bxd8 Rxd8 18.0-0
Material now level again but White stands better with his active queen, the probability of Rc1 to come and Black’s vulnerable a-pawn. Instead of 13.e5, White can try to be more adventurous with either 13.h4 or 13.0-0.However 13.h4?! seems to be dubious on account of 13…f6! 14.h5 Qf7 15.Qxb7 Rd8 when White has insufficient compensation. 13.0-0 is tempting Black to take the bishop but he has a much better move 13…Bc6! Black bolsters his queenside weakness leaving White’s Bg5 still compromised. (13…Bxg5 is very risky and probably bad 14.Ne5 Qh5 15.Nxd7 Nf6!? [15…Kxd7? loses to 16.Qxb7] 16.Bb5 Nxd7 17.Bxd7+ Kf8 18.Qxb7 Rd8 19.Qxa7 where White looks better to me) After 13…Bc6! White has many options but there is no obvious solution to his problems and Black simply looks better. Although both 13.h4 and 13.0-0 look aggressive and tempting, the straight-forward
13.e5! seems to offer White the best prospects here.
As opposed to 12…Be7 or 12…Ne7 the diagonal g5-d8 diagonal remains open signaling the possibility of a mating attack.
13.e5! Other options are less attractive.
13.Qxb7!? Ne7! (13…Bxb7?? 14.Bb5+ will mate) (13…Qxe4+? 14.Bxe4 Bxb7 15.Bxb7 Bb4+ 16.Ke2 Rb8 17.Bc6+ Kf8 18.Ne5! Nf6 19.Bxf6+ gxf6 20.Nd7+ winning) 14.Qa6 f6 and Black seems safe. 13.0-0 is too slow in view of 13…Be7! Transposing to a variation already discussed where Black stands well.
13…Be4 An aggressive but risky reply. 13…Qh5? loses to 14.Qxb7! Ne7 (14…Bxb7 15.Bb5+ and mates. This should be a familiar theme by now.) 15.Bb5 Rc8 16.Qxc8+ Nxc8 17.Bxc6 mate. 13…Bb4+ 14.Kf1! Qh5 15.Qxb4 and White stands better.
14.Qxb7! Bb4+ 15.Qxb4 Bxd3 16.Nc3!
Just in time!
16…f6 16…Rc8 17.Qb7 Rxc3 18.Qb8+ Kd7 19.Qd8+ Kc6 20.Qc8+ Kd5 21.Qxc3 Qe4+ 22.Be3 Qc4 23.Qa3! Kc6 24.Nd4+ Kb7 25.Kd2! threatening Rc1; White is winning. 17.Qb7 Rd8 18.exf6 gxf6 18…Nxf6 19.Ne5! 19.Ne5! fxe5 20.Qc6+ Kf8 21.Bxd8 White has recovered all his material and stands better. Therefore, after 12.Qb3! only 12…f6! seems to offer Black any prospects at all. Given this variation is hanging by a thread, Black may have to return to the option of 9… bxc6 as the way to solve his difficulties. I hope my two-part article encouragesmembers/readers to try this gambit for themselves and delve into the analytical depths to try and uncover fresh resources for both White and Black. I believe I have demonstrated that the Portsmouth Gambit is sound from White’s perspective and I thankLeonard Barden for his excellent inspirational article of thirty years ago! My thanks also to our Editor, Neil Limbert, for bringing a much needed reorganization to my copious notes, for his diligent and painstaking checking of my analysis, and for a number of important analytical contributions of his own.
Many thanks to The BCCA for allowing us to use this article.
Reprinted courtesy of BCCA