Back to Contents Page

Chapter One

Discovering New Ideas in the

Opening
The first thing I was confronted with when I started
playing again was the state of my opening repertory. I
still had plenty of unused analysis from my time as a
professional, not of all of which was obsolete, so it
seemed quite obvious to just play that again. At some
stage I understood though that it wasn't enough. The
openings themselves were fine, but playing these
openings game after game felt like going through the
motions: 'I played these 10 years ago, so I'll play them
now...and in 50 years' time.' On the other hand, just
playing a random opening without having any original
thought about it is also a bit of a waste. I feel much
more motivated and interested in a game if I have the
feeling that I'm coming to the board with something new
of my own. So what I needed was - as the Dutch saying
goes - a sheep with five legs. I needed new openings I'd
never seen/played before, I needed to have fresh new
ideas in them (if not actual novelties, then something
that felt new and exciting for me!)...and I didn't have
much time to study either.
Well I worked out how to put aside a few hours a week
for opening study, so full of expectation I sat down in
front of the board the first evening...and felt completely
blank. I switched on the computer, fired up ChessBase,
downloaded the most recent TWIC games, picked a
game at random, played it through, discovered the
opening summary feature, started Rybka, gazed blankly
while it evaluated the position as +0.03...and soon my
two hours were gone. Lovely...and this was supposed
to be creative and fun?
After a couple more evenings like that, I couldn't take it
anymore. My biggest problem seemed to be that I
couldn't generate any ideas. If I didn't start up Rybka, I
might stare at a position for hours without finding
anything interesting. But...I used to be really good at
this. So what was I missing now?
After pondering things for a while, I decided that I'd
forgotten the difference between playing and analysing.
When you play a game you invoke all sorts of
filters/prejudices (consciously or unconsciously) to
reduce the amount of information you have to process.
Think of things like 'I never play positions like that',
'Someone told me that you shouldn't do that', 'I'll never
be able to calculate all that', etc. This is useful in a
practical situation as it speeds up your decision-making
process, but these filters can get in the way of the
process, but these filters can get in the way of the
creativity you need in analysis. The goal in (opening)
analysis is not to limit the game and make it
manageable. The goal is to discover new truths, and
those truths may well lie outside the boundaries of your
prejudices.
My solution had three components:
a) To remind myself explicitly before I started that I was
analysing, not playing. I was going for creativity and
quality, not expediency.
b) I drew up a list of six 'mind-enlarging' approaches that
I hoped would stimulate me to leave my comfort zone
when analysing and trigger my creativity.
c) I set the rule that I would only use Rybka to check
analysis, not to generate ideas. In practice this meant
that I followed this routine:
1) I set an alarm clock for 15 minutes.
2) Analysed for 2 x 15-minute stretches.
3) Typed my analysis into ChessBase (with Rybka
switched off). Somehow writing things down always
activates the checking part of my brain (I wish I was
allowed to do that during games).
4) Then I switched on Rybka and checked through my
analysis. I awarded myself points for the number of
ideas I had analysed correctly. And what do points
ideas I had analysed correctly. And what do points
mean...? Pints! So you can see that a good analysis
session tended to last shorter than a bad session.
Well that's enough explanation; let's take a look at the
themes:
1) Disturbing the material balance.
2) You can't do that! Not in this opening! Well, actually...
3) My goodness, you can play this for a win!
4) Crossover plans.
5) Acts of wanton aggression.
6) The spoilsport gambit (exchanging queens).

Theme 1: Disturbing the material balance

This theme speaks for itself. Don't automatically shy
away from lines where you might have to give
something up. Be constantly aware that you can
unbalance the material equilibrium in order to generate
activity or create compensating weaknesses in the
opponent's position. The following example is a
wonderfully subtle exposition of this theme.
Shirov-Mamedyarov

Theme 2: You can't do that! Not in this

opening! Well, actually...
The previous example was also a good demonstration
of an unprejudiced approach to an opening. Who would
expect that you could play such a sharp exchange
sacrifice in the solid Ruy Lopez? Once you start
entertaining the thought, however, you suddenly start to
see all sorts of possibilities...

Theme 3: My goodness...you can also play

this for a win!
During my professional career, I scored very heavily on
the black side of the Queen's Gambit Accepted (QGA).
Strangely enough, my most important discovery in the
opening came about due to a serious failing of mine: my
extremely rigid opening repertoire. I played the same
openings against weak and strong players alike which
meant that I needed to be able to play all my openings
for a win. That requirement pushed me to investigate
the winning chances from a number of unpromising
situations, among which the typical QGA symmetrical
pawn structures where both sides have two pawns on
the queenside and four pawns on the kingside. To my
delight, I discovered that these positions weren't as
dead as everyone thought: there were real winning
chances with the right approach.
A word of warning while reading through this game: I'm
switching quite regularly to other topics within this game
switching quite regularly to other topics within this game
(for example, a related concept in another opening),
which can make it a little hard to follow.

P.San Segundo Carrillo-M.Sadler

Euro Team Ch., Pula 1997
Queen's Gambit Accepted
(View in Game Format)

1 d4 d5 2 c4 dxc4 3 Nf3 Nf6 4 e3 e6 5 Bxc4 c5 6 0-0

a6 7 Bd3

7...b6!?
A little wrinkle that I used in all sorts of QGA positions. I
got triggered to look at the move while studying
Hedgehog structures from a completely different
Hedgehog structures from a completely different
opening (an example of applying our fourth theme to
find new ideas in the opening). With 7 Bd3, White
delays the development of his knight on b1 and
anticipates 7...b5 which he plans to meet with 8 a4.
After 8...b4 White can transfer his knight on b1 directly
to the comfortable c4-square.
7...b6 achieves the same development goals as ...b5 -
activating the bishop on the a8-h1 diagonal - but avoids
providing White with an early target. The only drawback
to ...b6 is that Black's position remains slightly cramped:
for example, the black queen's natural square b6 is not
available. However, that doesn't have to be a
permanent state. The remedy to this problem occurred
to me after examining a completely different opening
(another example of applying the fourth theme in this
chapter!). Just bear with me as we wander off into a
Modern Benoni for a few seconds...

M.Sadler-E.Hummel
Amersfoort 2011
Modern Benoni
(View in Game Format)

1 d4 Nf6 2 c4 c5 3 d5 e6 4 Nc3 exd5 5 cxd5 d6 6 e4

g6 7 f3
A favourite scheme of mine against Benoni structures.
The pawn's placement on f3 is temporary. By shoring
up the e4-pawn, White takes the edge off Black's typical
Benoni counterplay which is normally based on a
combination of ...b5 and tricks against e4. It does give
White a bit of development malaise (try and get the
knight on g1 to somewhere decent...), but once he's got
himself sorted, White can get going with what he really
wants: playing f4 and e5.
7...Bg7 8 Bg5 0-0 9 Qd2 h6 10 Be3 a6 11 a4 Nbd7 12
Nh3 Re8 13 Nf2 h5 14 Be2 Rb8 15 0-0 Nf8 16 Kh1
Bd7 17 a5 N6h7 18 f4!
18...f5 19 e5 dxe5 20 Nd3
...and White enjoys a strong initiative.
In that line of the Modern Benoni, White plays his
desired plan f4 in two steps. The first step is inserted in
order to stifle any early Black counterplay attempts. So,
returning to San Segundo-Sadler, why can't I do the
same in the Queen's Gambit Accepted?
I play my desired plan ...b5 in two steps, and so avoid
presenting White with a target for attack at an early
stage.
8 Qe2 Bb7 9 Rd1 Nbd7 10 Nbd2 Be7 11 b3 0-0 12
Bb2 Qb8
I felt the time still wasn't right for ...b5, so I carried on
developing behind the lines. The text prepares to bring
a rook to the c-file so that I can ease my cramped
position with an exchange of rooks.
13 Rac1 Rc8 14 Bb1 cxd4 15 Rxc8+ Qxc8 16 Nxd4
So here we have it: the standard QGA symmetrical
pawn structure. Dull, but not dead! I played these
positions for a win based on two premises:
1) Nerves play a part. The white player can fall prey to
all sorts of nervous thoughts:
a) Disappointment and the urge to get away from that
disappointment. 'Curses!, I've got absolutely nothing
from the opening. I just need to force the draw and get
out of here. I can't seem to find a way to liquidate to a
clear draw...'
b) Indecision about the best course of action. 'Shall I
offer a draw, or do I play on? I don't want to be a
coward, but I'm not sure what I'm doing as White. Is he
playing for a win? Or am I still a bit better? I am White
after all...'
c) Carelessness. 'I can do anything. Nothing is ever
going to happen to me.'
d) A misplaced desire to 'show' Black that White knew
precisely what he was doing. 'So you think you've
equalized? Well, I'm going to show YOU who's the
better player...' This last frame of mind leads inexorably
to the second premise:
2) White plays e3-e4.
First things first: this isn't necessarily a bad move. White
blocks the diagonal of Black's light-squared bishop on
b7. If he can hold that blockade, then he's on the way to
a really big advantage. Just switching again to a few
examples from my games:

M.Sadler-J.Lautier
Elista Olympiad 1998
(View in Game Format)
After a long tactical sequence, I got the chance to shut
out the light-squared bishop and remove every last drop
of activity from Black's position:
21 e4! Ba8 22 Ne5 Nxd7 23 Nc4!
The dark-squared bishop is the one I want.
23...Ra7 24 Nxb6 Nxb6 25 Bb5
This stops Black regrouping with ...Rad7.
25...Nd7 26 Rfd1 Nf8 27 f3!

All I want now is to swap off all the major pieces and the
a-pawn is mine.

M.Sadler-N.Pedersen
M.Sadler-N.Pedersen
Dutch League 2002
(View in Game Format)

21 Nd3 a5?
21...Bd5 was Black's last chance to stay in the game.
Now it gets very painful.
22 f3! Ng3 23 Nxb4 cxb4 24 Qf2 Nf5 25 e4!
The queenside weaknesses render Black's position
hopeless.
However, white players often don't quite appreciate how
fundamentally e3-e4 changes the position. In one fell
swoop, Black gets central dark squares to play with: the
a7-g1 diagonal is no longer blocked, while e5 becomes
a lovely step-on/step-off square for the queen or a
knight. A nice example of this is my game against Mark
Hebden:

M.Hebden-M.Sadler
Hastings Challengers 1991/92
Queen's Gambit Accepted
(View in Game Format)
1 d4 d5 2 c4 dxc4 3 Nf3 a6 4 e3 Nf6 5 Bxc4 e6 6 0-0
c5 7 a3 b5 8 Ba2 Bb7 9 Nc3 Nbd7 10 Qe2 Qb6 11
Rd1 Rd8 12 e4?! cxd4 13 Nxd4 Ne5

It starts! Note that Black's pieces are already ideally

placed: the queen is on b6 and the knight is already
making use of the e5-square. Note also that White's
light-squared bishop on the a2-g8 diagonal does not
combine well with the plan of advancing e3-e4 in the
centre.
14 Bf4 Ng6 15 Be3 Bc5 16 Nc2 0-0 17 Bxc5 Qxc5
In principle, White really does want to exchange off the
dark-squared bishops and leave Black with a
light-squared bishop biting on granite. In the current
position, however, White is finding it hard to cope with
position, however, White is finding it hard to cope with
Black's pressure against e4 and his entry possibilities
on the dark squares. The knight on g6 is particularly
annoying - remember that knight!
18 g3 Ne5

Got my weakness, back to the transit square. The

knight is now looking at both f3 and c4.
19 Rac1 a5
Looking for ...b4 to step up the pressure on e4.
20 Ne3 Rxd1+ 21 Qxd1 Qb6 22 Bb1 Rc8
22...Rd8 was awesomely strong as 23 Qe2 Rd2 is
curtains. However, I do like what I did: there's no rush,
so let White collapse all by himself!
23 b4 axb4 24 axb4 g6?!
Again, 24...Rd8!.
25 h3 Kg7?!
And, again, 25...Rd8! was indicated.
26 Rc2 Rd8 27 Qe2 Nxe4

and Black won shortly:

28 Ng4 Nxg4 29 Nxe4 Bxe4 30 Qxe4 Rd1+ 31 Kg2
Nf6 32 Qe5 Qb7+ 33 f3 Rxb1 34 Rd2 Qa7 0-1

We have seen both extremes in the examples above. In

my games against Lautier and Pedersen, Black lost all
control over the central dark squares and so couldn't
make any use of the drawbacks of e4. In my game
against Hebden, Black had everything he could ever
dream of. After 16 Nd4 in San Segundo-Sadler White's
dream of. After 16 Nd4 in San Segundo-Sadler White's
pieces are better placed and Black's pieces are more
passive than in the previous Hebden-Sadler game.

Black won't be able to stop White going e4-e5.

But...that's not bad either! Once White plays e4-e5, the
a8-h1 diagonal is reopened for the b7-bishop which
gives Black hope again for fun and games on the
kingside.
16...b5!
This takes c4 away from the white knight. Moreover,
a2-a4 isn't dangerous any more.
17 e4 Nf8!
Heading for g6, tickling the f4-square.
18 N4f3 Qc7 19 a3 Rd8 20 b4 Ng6! 21 g3
Now the next step is to somehow give White the urge to
play e4-e5.
21...Ba8 22 Rc1 Qb7
Eyeing the e4-pawn.
23 Re1 Ne8!
Another typical manoeuvre, bringing the knight to d6 to
eye the outpost on c4.
24 Nb3 Nd6 25 Nfd2 Qb6 26 h4
Yum! White's coming forward. My knight doesn't have
all that much to do on g6 so I don't mind it being chased
away.
26...h6 27 Bd4 Qb8 28 h5 Nf8 29 Qg4 Ne8
Classic. I've got a number of teasing little ideas now:
...Bg5, ...Bf6, ...Nf6 and ...e5.
30 e5
There it is!
30...Nh7!
Heading for g5.
31 Ne4?! Qb7!
That's annoying. The knight is pinned now.
32 f3
Another little weakness appears.
32...Qd5 33 Ned2 Ng5 34 Kg2 a5!
The beginning of the end. White's position is collapsing.
35 bxa5 Bxa3 36 Re3 Bb4 37 Rd3 Qc6 38 Bb6 Rxd3
39 Bxd3 Bxd2 0-1
Black wins a piece after 40 Nxd2 Qc3.
This game just goes to show that something new
doesn't always mean new moves. Just deciding that you
can play a certain type of position for a win instead of
for a draw can also unlock a range of hidden
possibilities.

Theme 4: Crossover themes

The most rewarding technique for unearthing new ideas
in a particular opening is to borrow ideas from other
openings. It's both an effective way of discovering new
openings. It's both an effective way of discovering new
paths during home analysis and an efficient way of
finding new opening ideas at the board. We have
already seen two examples of this theme in the 7th
move notes to my game against San Segundo.
However, I first made use of this technique when I
discovered some unexpected similarities between my
two favourite openings: the Najdorf and the Queen's
Gambit Accepted.
I started playing the Najdorf when I was about 10 or 11.
I was very heavily influenced by Polugaevsky's games,
probably due to his sublime book Grandmaster
Preparation. I took away four lessons from studying
Polugaevsky's Najdorf games:
a) You castle as late as possible (preferably not at all).
b) You can survive anything if you calculate precisely
enough.
c) You should delay development in order to achieve
positional goals (for example, getting rid of White's
remaining central pawn - the e-pawn).
d) Sacrifice for the initiative!
Looking back, it's quite scary how much a single book
can shape your conception of a certain subject. For me,
these rules had the immutability of God. Still do, come
to think of it. I get this terrible feeling of blasphemy
when I see black players castling early in the Najdorf!
Thank goodness for my chess career that Polugaevsky
knew what he was doing! You could do worse than
follow his example:
Tal-Polugaevsky
Another Najdorf theme that grabbed my attention was
the fight for the e5-square, and I developed a number of
home-brewed variations based around this theme. For
example, take this very old game of mine:
Cullip-Sadler
This may remind you of my emphasis on the e5-square
in the QGA positions that we saw earlier, and that's no
coincidence. I don't think that I would have cottoned on
so quickly to this idea in the QGA if I hadn't been a
Najdorf player. The Najdorf was also a huge inspiration
for one of the most creative bits of analysis I did in the
QGA. Unfortunately, I never got to play it myself! Joel
Lautier (whom I seconded for many years) did, but even
that wasn't an unqualified success: his opponent, Jan
Timman, found the most critical and dangerous idea at
the board and even improved significantly on my
analysis. Still, it was a great concept!
Timman-Lautier
This dancing on the dark squares around a central
pawn on e4 has become one of my favourite motifs, and
I've applied it since to a variety of other openings. Take
I've applied it since to a variety of other openings. Take
a look at these two recent games; both examples of
using this theme to find ideas at the board.
Van Oosterom-Sadler
Without that game, with its positional theme of fighting
for the central dark squares with an advance of the
g-pawn, I might never have played this game:

F.Elburg-M.Sadler
Dutch League 2011
English Defence
(View in Game Format)

1 c4 b6 2 Nc3 Bb7 3 e4 e6 4 d3 c5 5 g3 Nf6 6 Bg2

Be7 7 Nge2 d6 8 0-0 Nc6 9 h3
I wasn't thrilled with the way I had developed. White has
two obvious plans to gain space:
a) Expanding on the kingside with f4, g4.
b) Expanding in the centre with d4 leading to a
Hedgehog type position after ...cxd4; Nxd4.
The first plan is the most annoying. White avoids
central conflict and asks Black to think up something
active himself. Black's natural plan is to gain space on
the queenside with ...b5, but I haven't made a good start
on that by playing ...b6 and ...Bb7 at such an early
stage. If I castle kingside, then I castle into the area
where White is expanding. Castling queenside ...hmm.
Well, it's an acquired taste.
None of the normal stuff looks that great. So I started
looking around for different development schemes. The
normal random moves like ...h5 pop into your mind of
course, but I couldn't really wrap a plan around that. I
started thinking about why castling queenside seemed
such a bad idea. It looked much easier for White to
open lines on the queenside than for Black to open
lines on the kingside. I don't have any convincing
targets to get started there. It would be easier if White
had played f4 for example: I could look for something
like ...g5 opening up the g-file, even at the cost of a
pawn. If I had that, then I could consider castling
queenside. I would still give White a big target, but at
queenside. I would still give White a big target, but at
least I would have something in return...
Lightbulb! How about playing ...h6 and ...g5? If White
just lets me do it, then I can meet f4 with ...gxf4 which
gives White a dilemma: gxf4 opens up the g-file which
gives me an avenue of attack against the white king,
but if White takes back with a piece, then I have the
e5-square! White does have the f-file, of course, so f7
might get a bit sensitive, but hey ho, you can't have
everything.
The big question is: what does Black do if White avoids
f4, and just switches back to gaining space in the centre
with d4. Doesn't ...h6 and ...g5 just look silly then? Well,
then you need to know your classics! Take a look at
these two games:

R.Fischer-U.Andersson
Siegen 1970
Nimzowitsch-Larsen Attack
(View in Game Format)

1 b3 e5 2 Bb2 Nc6 3 c4 Nf6 4 e3 Be7 5 a3 0-0 6 Qc2

Re8 7 d3 Bf8 8 Nf3 a5 9 Be2 d5 10 cxd5 Nxd5 11
Nbd2 f6 12 0-0 Be6
13 Kh1 Qd7 14 Rg1! Rad8 15 Ne4 Qf7 16 g4
Beginning the attack, but just what is Black up to?
16...g6 17 Rg3 Bg7 18 Rag1 Nb6 19 Nc5 Bc8 20 Nh4
Nd7 21 Ne4 Nf8 22 Nf5!
Fischer went on to win a beautiful game:
22...Be6 23 Nc5 Ne7 24 Nxg7 Kxg7 25 g5 Nf5 26 Rf3
b6 27 gxf6+ Kh8 28 Nxe6 Rxe6 29 d4 exd4 30 Bc4
d3 31 Bxd3 Rxd3 32 Qxd3 Rd6 33 Qc4 Ne6 34 Be5
Rd8 35 h4 Nd6 36 Qg4 Nf8 37 h5 Ne8 38 e4 Rd2 39
Rh3 Kg8 40 hxg6 Nxg6 41 f4 Kf8 42 Qg5 Nd6 43
Bxd6+ 1-0

M.Taimanov-A.Yusupov
USSR 1982
English Opening
(View in Game Format)

1 d4 Nf6 2 c4 c5 3 Nf3 cxd4 4 Nxd4 b6 5 Nc3 Bb7 6

f3 e6 7 e4 d6 8 Be2 a6 9 Be3 Nbd7 10 0-0 Be7 11
Qd2 0-0 12 Rfd1 Rc8 13 Rac1 Qc7 14 Bf1 Rfe8 15
Qd2 0-0 12 Rfd1 Rc8 13 Rac1 Qc7 14 Bf1 Rfe8 15
Kh1 Qb8 16 Qf2 Bd8 17 Nb3 Bc7 18 Qg1

18...Kh8!
In this typical Hedgehog position Yusupov borrows
Fischer's plan.
19 Rc2 Rg8! 20 Rcd2 g5! 21 Bd4 Rg6 22 Nc1 Rcg8
23 Nd3 Qf8 24 Re1 g4!
Black has a big initiative and Yusupov won after:
25 fxg4 e5 26 Be3 Nxg4 27 Nd5 Bd8 28 Nf2 Bh4 29
Ree2 Nxe3 30 Nxe3 Bxf2 31 Qxf2 Bxe4 32 Nf5 Nc5
33 Ng3 Ba8 34 Rd1 Ne6 35 Qxb6 Nf4 36 Rf2 Qh6 37
Kg1 Qh4 38 Qb3 Rh6 39 Rxf4 exf4 40 Qc3+ f6 41
Nf5 Rxg2+ 42 Bxg2 Qxh2+ 0-1

Summarizing the position after the 9 h3 in Elburg-Sadler:

Summarizing the position after the 9 h3 in Elburg-Sadler:
White has two clear plans: gaining space on the
kingside (f4, g4) and gaining space in the centre (d4).
I'm most worried about the plan of f4 and g4 because
then I need to come up with something to do myself.
None of the standard stuff looks convincing: for
example, castling kingside puts me in the line of fire of
White's kingside expansion.

I want to fight White's kingside expansion by trying to

gain space there myself. If lines get opened there, then I
can castle queenside. That's risky, but at least I've
already made a start attacking White's kingside. And if
White switches back to a standard plan of d4, I can
claim to be following the example of Fischer and
Yusupov!
9...h6!? 10 Be3

10...a6!?
Camouflaging my idea. I wanted to be able to answer
d4 with ...cxd4; Nxd4 Ne5, but I wasn't 100% happy
about doing it immediately. I don't think that was
necessary, though. Indeed, 10...g5 11 d4 cxd4 12 Nxd4
Ne5 13 f4 gxf4 (13...Nxc4 14 Qa4+was my worry of
course) 14 gxf4 Ned7 is pretty much what I was looking
for!
11 Qd2 g5
12 f4?!
Better was 12 d4, although actually it's not clear that
Black even needs to take on d4 immediately. After 13
d5 Ne5 14 b3, Black can try 14...g4 when the threat of
...Nf3+ is quite annoying. So we have:
a) 12...Qc7!? 13 b3 e5 14 dxc5 dxc5 15 Nd5 Nxd5 16
cxd5 Nd4, although White is probably better after, say,
17 b4!?.
b) 12...e5!? 13 dxc5 dxc5 14 Nd5 Nd4 (14...Nxd5 15
exd5 Nd4? 16 Bxd4 exd4 17 d6 wins) 15 Nec3 should
be a bit nicer for White.
c) 12...cxd4 13 Nxd4 Ne5 14 b3 and then:
c1) After 14...g4 15 h4 h5 I'm not quite sure how Black
will follow up, but if instead 15...b5 16 cxb5 axb5 17
will follow up, but if instead 15...b5 16 cxb5 axb5 17
Ndxb5 Nf3+ 18 Bxf3 gxf3 19 Rfd1 Ra6 20 e5 or
15...Nfd7 16 Bxh6 Bxh4 17 gxh4 Qxh4 18 Bf4 Nf3+ 19
Nxf3 gxf3 20 Bxf3 Ne5 21 Bg2, and White wins in both
cases.
c2) 14...Rg8! and I'll follow in the footsteps of
Fischer/Yusupov and put the king on h7!

White has two entry points into Black's position:

1) The f4 break pushing back the black knight on e5
and aiming to follow up with f5 or (rather unlikely) e5.
This is, however, very risky for White due to the open
g-file. If it works, it's going to be due to a one-off specific
thing.
2) An attack against the weak b6-pawn. However, the
obvious Na4 will fail due to the weakness of e4.
obvious Na4 will fail due to the weakness of e4.
Therefore a more modest attack with a4-a5 seems quite
logical, holding White's position while introducing an
extra threat.
The quiet Kh1 is another useful move that will be
played very soon, hinting at a future f4. Play might
continue 15 a4 Rc8!? 16 a5 bxa5 17 Qa2 Rc5!? with
...Qa8 to follow, and if 18 Nc2 g4!?.
12...gxf4 13 gxf4?!
If 13 Rxf4 Nd7.
13...Qd7

14 Kh2
14 Na4 Bd8 holds the b-pawn.
14...Rg8 15 Bf3 0-0-0 16 Rg1
The direct 16 Na4 Qc7 17 b4 Nxb4 18 Rab1 sort of
worried me a bit...
16...Na5! 17 Bf2 d5! 18 cxd5
18 e5 was much sharper. I've done lots of analysis of
this position, but I'll just show you a couple of variations
here to give a flavour of things. Note especially how the
weakness I've created on the white kingside by opening
the g-file comes back in all the tactics. If I hadn't done
that before castling queenside, then nothing would have
worked. After 18...d4 19 Bxb7+ Qxb7 20 Na4 Nxc4! (the
cunning idea behind 16...Na5; 20...Nd7 gives White
some respite) we have:

a) 21 dxc4 Ne4 22 Qe1 Rxg1 23 Qxg1 b5 (if 23...Nd2

24 Qg2! Nf3+ 25 Kh1) 24 cxb5 Qxb5! 25 Nac3 dxc3 26
24 Qg2! Nf3+ 25 Kh1) 24 cxb5 Qxb5! 25 Nac3 dxc3 26
Nxc3 Nxc3 27 bxc3 Qc4 is good for Black.
b) After 21 Qc2 Qf3 22 Ng3 I was sure during the game
that something would be winning here, but it's not as
clear as I thought: 22...Qxf4 (22...Ng4+ 23 hxg4 Rxg4
24 Rg2 Ne3 25 Bxe3 Qxe3 - threatening ...Rh4+ - 26
Qf2 is dangerous for White, but is not conclusive after,
say, 26...b5 27 Nxc5 Bxc5 28 Qxe3 dxe3 29 Rc1) 23
dxc4 Nd7 24 Qe4! Qxe4 (otherwise, 24...Qxf2+ 25 Rg2
Qe3 26 Qc6+ Kb8 27 Nxb6 Nxe5 28 Qa8+ Kc7 29
Qa7+ Kc6 30 Qxa6 leads to a draw, and White is better
after 24...Qxe5 25 Nxb6+ Kc7 26 Nxd7) 25 Nxe4 Nxe5
26 Nxb6+ Kb7 27 Rxg8 Rxg8 28 Na4 Kc6 29 Bg3! Nf3+
(29...Nxc4 30 Rc1 Ne3 31 Nexc5 isn't so clear) 30 Kg2
f5 31 Kxf3 fxe4+ 32 Kf2 and I like Black.

White's key difficulty is that the bishop on g3 cannot

White's key difficulty is that the bishop on g3 cannot
leave its post: ...Bh4+ is very awkward for White. Then
Black forces entry to the seventh rank and takes control
of the queening square of the passed e-pawn.
18...exd5 19 e5 d4 20 Bxb7+ Qxb7 21 exf6 dxc3 22
bxc3
Otherwise:
a) 22 Qxc3 Qf3! 23 Bg3 Rxd3 24 Qe5 Rd2 wins.
b) 22 Nxc3 Rxg1 23 Rxg1 Bxf6 24 Qe2 Nc6 is very
pleasant for Black.
22...Rxg1 23 Rxg1 Bxf6 24 Qe3 Qd5 25 d4 Nc4 26
Qd3 Kb7 27 Qg3?
A slip, but 27 Rb1 Kc6 is very nice for Black.

27...Nd2! 28 Rd1 Nf3+ 29 Kg2

Black wins too after 29 Kh1 Nxd4+ 30 Kh2 (30 Kg1
Nxe2) 30...Nf3+.
29...Nxd4+ 30 Kf1 Nxe2 0-1
31 Rxd5 Nxg3+ wins.

Theme 5: Acts of wanton aggression

In this section, we consider a number of standard ways
of upping the tension in a position.

An unreasonable 0-0-0
My all-time favourite method for sharpening up a
position. I was enormously influenced by this opening
idea in the 5 Bf4 variation of the Queen's Gambit
Declined:
Gurevich-Sokolov
I was impressed by the idea of course, but I was even
more impressed by the unsettling effect the idea,
combined with the sharp follow-up 12 g4, had on
Gurevich's very strong opponent. 12...b5 (a pawn
sacrifice played on general grounds, but not properly
calculated) and the panic reaction 14...Nb4 (facing
White's sneaky idea of Nc4 winning the queen) show
how confused and vulnerable the black player felt.
So I had to try that too! One of the games that have
So I had to try that too! One of the games that have
stuck in my memory is my game against Keith Arkell,
played towards the end of the British Championship of
1995. I was playing really appallingly, as I tended to
around that time (I was so distressed by the way I was
playing, I was one game away - one time-trouble
second away in fact - from stopping chess in 1995), but
this game started me off on a 3/3 finish and my first
British title.
10 0-0-0 wasn't a great chess move in this particular
game, but it had the right psychological effect. I really
felt that I'd forced Keith into an unfamiliar type of
position and that his play in the middlegame suffered as
a result.
Sadler-Arkell

Early kingside thrusts

Thanks to the efforts of Alexei Shirov, the early kingside
pawn thrust g4-g5 has become a standard weapon in
many different openings. g4-g5 is effective because it
disrupts the black knight on f6 which is normally a
stable factor in Black's position: the knight both
supports the central squares e4 and d5 and helps
defend the kingside. Take a look at this somewhat
famous example:
Shirov-Thorhallsson
Recently Aronian came up with a slight twist to this
idea: an early kingside pawn storm aimed at a pawn on
h6 even when the Black knight has already left f6! As
we can see from the following game, this idea is
working a treat:
Ernst-Harikrishna

Theme 6: The spoilsport gambit

(exchanging queens)
It's very easy to get carried away during analysis, only
considering the complicated, attacking stuff and
forgetting all the options that simplify the position. After
all, are you going to be impressed with yourself if you
spend four hours on a position and all you come up with
is an exchange of queens?
I like to think back to an experience I had while
analysing an interesting position with some strong
players. Kramnik was looking over our shoulders. Every
weird, 'creative' idea we had got beaten down by him
with comments like: 'No, but you don't need this. Qd4,
offering the exchange of queens is an easy advantage.
Simple play.' Simple can be best!
One of the best guys for these sorts of opening ideas is
Yasser Seirawan. You see it happening time and time
again in his flank openings. For example:
Seirawan-Nikolic

Last thoughts
Opening analysis can be a difficult skill to master,
especially when you're working on your own. It's really
important to get past all the instincts that are quite
healthy in practice - 'I don't want to waste time on this; it
might well turn out to be nothing', 'No one ever plays
like this in this opening; don't waste time on that' - and
open your mind up to non-standard possibilities.
Remember, at home you've got all the time in the world
to waste on bad ideas, and no effort is ever wasted.
If you do have the opportunity to work with another
player, then the following method of working is a very
effective way of investigating openings:
a) Agree before a training session on a number of
openings you want to practice. The more specific you
are, the more valuable the training session.
b) Analyse the variations you have selected
independently, looking seriously for ideas with both
colours.
c) Play a series of games with the openings (for
example, 30-minute games). Each player should try
playing both colours.
d) After the games have concluded, analyse them
d) After the games have concluded, analyse them
together.
I've done this a number of times in the past couple of
years and it's always worked extremely well.
So this was all about generating new ideas. In the
following chapter we look at the next step: feeling
comfortable when playing your new ideas/openings in a
tournament game for the first time.

Back to Contents Page