Another “Best game ever” — so far

I’ve been out playing chess again. This particular game appeals particularly to me, for various reasons: it was a win, against a much stronger player; it was a calm and steady win, not decided by some fluke error in either direction, just a steady pressure which gradually increased the positional advantage up to five pawns’ worth, while the material was still equal; and I managed to play through the whole middle game without throwing the whole thing on the floor, as I have a tendency to do.

As before, I present it here not as a demonstration of brilliance, but as an account of how a medium-class chess player thinks. The comments up to move 5 are mainly a synthesis of theory remarks, included mostly for my own sake.

So here we go: Faxe Chess Club Individual Tournament, round 4: White: yours truly, Black: Knud Hansen.

1.d4 d5 2.Bf4

Meet my new best friend: the London System.

Considered to be a dull, predictable opening, and hated by the likes of Kingscrusher for that reason. Or as a less judgemental friend said: it may be a comfortable opening for white, but it is quite comfortable for Black too.

Be that as it may, I like it for three reasons:

1. The piece development and play is very fluent; no minor pieces blocking each other in, and the central control is formidable.

2. Great attacking possibilities. Common lines involve an open h-file after …Bd6, Bg3 BxB, hxB or …Nf6-h5, Bg3 NxB, hxN; or a hack-attack-type of h-pawn advance; or the Queen-and-Bishop battery Qc2/Bd3, which together with the Bishop on f4 may cause great distress for the Black king. And

3. many of the things one usually fears, such as losing a bishop to a knight or getting double pawn, are nothing to fear here, since they are actually part of the advantage….

This is not to say that it’s a winning machine, but it’s comfortable — very comfortable!


By far the most common reply, at least judging from the games I’ve played with this opening. At club-level, 1.d4 d5 players are probably so accustomed to the QGD lines that they automatically (after some thought, usually) go into this.

It should be said that there’s nothing wrong with the 2…e6 response; the advantage for white is that, given the predominance of this response, one can narrow down the number of lines one has to study…

This is considered to be the most precise move order. 3.Nf3 is an alternative, but it allows Bd6! after which there are several alternatives, all of which seem to, if not directly favour black, then at least be on Black’s premises:  4.Bxd6 cxd6 gives Black good control over e5, and 4.Bg3 loses a tempo, but is probably better, thanks to the opened h line should black take the bishop. 4.e3 is a good alternative, after which Bxf4 5.exf4 gives White a double pawn, but also excellent control of e5.
3. … Nf6 4.Nd2
This move order is better than c3 first; there is no need to defend the centre with c3 before it is actually attacked, and b6 and Ba6 can now be countered with c4 directly — one of the few instances where c4 and not c3 is correct in this opening.
In this line, with black’s light-square bishop blocked in, an eventual Qb6 can be comfortably met with Rb1, since there is no Bf5 to harrass the rook. Also delaying Nf3 to prevent any funny business with the f6 Knight.
Again, 4. … Bd6 is the logical attempt at taking advantage of the delayed Nf3, since there is no knight that can block the bishop exchange with Ne5, owing to the delay of Nf3.
5.Ngf3 h6?
Too passive. g5 isn’t where white is heading anyway, so no need to cover it yet, and the Qc2/Bd3 battery is a distant future. If anything, the move creates a target for the dark-square bishop, or eventually — as we shall see — other tactics.
5. … Be7, 5. … c6, or  c5 are all better
6.Bd3 Be7 7.c3
The London setup is complete. The double Bishop/Knight batteries are great, the Queen is ready to go to c2 or work along the d1-h5 diagonal, and the Knight on f3 — well… we’ll see.
…O-O 8.Ne5
This is a thematic Knight move in the London. If the knight is taken, white recaptures with the pawn, forcing away the Knight on f6 thus weakening the defense of the black king. If left alone — well, I personally never objected to a fairly secured knight on e5…
…Nxe5 9.dxe5 Nd7 10.Qg4
I wasn’t quite sure about this move, but the computer confirms that it’s correct. The immediate threat is Bxh6, so
is logical (and was played quickly)
I thought for a long time about this one. I was considering Bg3, g3, and Nf3 too. The move I chose may not have been the objectively strongest, but at least it’s not losing (although the advantage diminishes strongly), and it’s consistent with my long-term strategy: queenside castling, pawn storm, and open lines on the kingside.
…Bxf4 12.exf4

12.Qxf4 was possible too, and the engine likes it better than what I played. It gets the queeen out of the way of the g pawn to advance more quickly than in the game, and it allows the queen to swing quickly to the queenside if needed, but it also leaves the e5 pawn slightly vulnerable, at least temporarily.

At least that’s what I thought during the game. On second thought (and with support from the engine), there is no danger of the queen ever being chased away from the defense of e5, and black really doesn’t have much more to attack that pawn with anyway.

12. … Qe8
I kinda liked this move, but the engine isn’t crazy about it. It does strengthen the defence of the king position, and the Queen is ready to swing to the queen-side after Nc5.
Given that Black has just signalled that he is ready to counter-attack on the queenside with his last move, this may seem more courageous than sound. I figured I have enough pieces left to defend the king, though, and as pawn storms go, I have a head start.
As expected.

The bishop is going to stay here for the rest of the game. At the moment, it just controls the black Knight’s forward movement, but it is going to be a decisive player in the end.

Another nice thing about the London System: the light-square bishop.

This move seemed a bit over-ambitious, but potentially dangerous. I therefore decided to encourage simplification with
15.Nf3 would probably have been better: the knight gets to d4 anyway, and a knight exchange doesn’t seem to benefit me, although I’m still half a pawn up after it, according to the engine.
15. … Nb7
This is probably too passive. It supports various pawn advances (a5 and c5), but I wasn’t too afraid of them.

Clears the way for the g pawn, and takes part in the queen-side defense, even with some attacking possibilities (Nc4, for example).

After a deep thought, the engine agrees with me, to my delight (hence the “!”).

I reckoned that it would take several moves for Black to get a serious attack going, so I wasn’t too concerned: any further pawn advance would either have to be supported by pieces (which are not in position yet) or would allow me to halt the attack. Most specifically this goes for c4. My next few moves therefore involved considerations about how to force c5-c4.
17.Qd3 was the other move I considered, and perhaps it is better. It would virtually have forced 17…g6 after which I intended to play Qe3, threatening c5, or Qg3.
My opponent thought for a long time about this one, and whenever I do that, I ususally come up with a bad move. The same thing happened here:
I’m clearly ahead in the race, and it seems like a good idea to react on the kingside, but the engine actually prefers pawn advances on the queenside for black anyway. This move, while temporarily blocking my bishop’s view, allows me to break open some lines against the king, and gives me concrete targets after the exchange.
18.exf6 was apparently better; I didn’t like the idea of opening the f file for the rook, but it seems that Rxf6 19.g5 Rxf4 20.gxh6 gxh6 21.Rhg1+ Kf8 22.Qe5 would have been crushing.
I chose to capture with the g pawn, because it gave a more clear position: the g file is semi-open, the black f pawn is under attack, and the black pieces are under control (i.e.: blocked in).
18. … exf5

We had reached a “stage” here, so time to sum up some factors:

I have an open g file against a hanging pawn and a shaky king’s position, I have a passed e-pawn, I have some pressure on the d pawn, and I have options against black’s f pawn.

Black in return has … uh … I guess some pawn pushes on the queenside. IF he ever gets to do them. Mwahahah.


I could have taken the d pawn here, but I was worried that there might be trouble with two exposed rooks on the long diagonal should the bishop get to b7 (I actually thought I saw a way for black to force that, but now I can’t find it. Just a ghost, probably).

Anyway, I was quite happy with this “quiet move”, actually. It attacks c5 (thus virtually forces the next black move, which gives my knight a wonderful spot on d4), and gets the queen ready to swing to the g file. I was quite happy with this position…

19.Rxd5 c4 20.Nd4
19. … c4 20.Nd4
“On a Knight like this…”, to quote Dylan.
Seems a bit desperate, but the alternative still gives med a solid edge.
20. … Nc5 21.Rhg1 a5 22.Rg2 Rf7 23.Rdg1 b4 24.Nxf5 Bxf5 25.Qxc5 Bxc2 26.Qxd5 Kh8 27.Kxc2 Rxf4 28.Rxg7 Rxf2+ 29.Kc1

Probably not the best move, but I’d make it again any day.

The reasoning was: “If I can advance the e pawn, I will win black’s f pawn. The only piece that can block that advance is the bishop (since 21. … Ne6 22.Nxf5 Rxf5 23.Bxf5 Qf7 24.Bxe6 Bxe6 25.Rhg1, and my lead is up to four pawns), and if the bishop goes to e6, I won’t have to fear anything along the long diagonal anymore. On e6 the Bishop is going to be staring up the asses of two pawns, and that’s the last we’re going to see of that bishop for a LOOONG time.

I also need some firepower on the g file. The ideal setup would be Rg6, Qg3 or g2, and Rg1. In any case, i don’t want the queen to be the avant guard in that trio; it would have to be rook.”

Which one? I wanted to advance the h pawn first. Qf3 would support that, AND attack the d pawn, but it also seemed kinda slow. So I decided to keep the h rook where it was and send the d-rook first into the fire, followed by Qg3 (possibly Qg2 from there, again attacking the d pawn).

And if it could send a bishop into oblivion on its way to the fireworks, I’d be all for it!

… and here it comes …
… and here IT comes!
Dual-purpose: might advance to the second rank, defending g7, but also supports a b-pawn advance.
Inexact, since it allows…b4, which is still better for white, but gives black some counterplay.

23. … b4 24.cxb4 Rxb4 25.Rg6 Rb7 26.Rhg1 Kh8

Rb7 would have been better as well. It’s the second rank that needs to be defended.


Stage one completed.
Bc8 25.Rhg1
25.e6 Bxe6 26.Qe5 Qd7 27.Re1 would have been stronger, but it’s a tricky win to find, especially when there’s just twenty minutes left on the clock…
Another possibility was 25.Rxb6 axb6,
which would effectively have ended any hopes of serious counterplay on the queenside, but I was much more keen to increase the g file pressure, hence the game continuation.
25. … Rxg6 26.Rxg6
Interesting: Material is still equal, but my advantage is up to five pawns…
…Rf7 27.Qg3(?)
The move isn’t disastrous, but I actually “loose a pawn” in evaluation terms. In any case, it forces the king to move out of the way, and I get a second attacking shot for free. I learned that from Kingscrusher: look for forcing sequences. I learned it from William Ockam too: the easiest path is usually the best…

27.e6 would have been the quicker win, but I miscalculated the number of attackers and defenders… Fatigue.


27. … Rf6 28.Rxf6 gxf6 29.e7 Kf7
29. … Nc6 30.Nxc6
30.exd8=Q Qxd8 31.Nxb5 Qe7
28.Nxf5 Rc7 29.e7
27. … Kh8 28.Rd6! Nb7?

Here, my opponent offered a draw. I would have done the same in his shoes.

I declined, however — I was wearing my own.

28. … Rd7 is the best, but it still allows 29.Nxf5

The club player’s dilemma: I want that pawn, I even moved the rook to d6 to get it, but can I actually take it? What if the rook is trapped out there? And the clock is ticking.

The only piece that could attack it lethally would be the bishop, and that would have to be from b7, and if the bishop left the defence of f5, I would probably manage to crash through even at the price of sacrificing the exchange. Besides, black would more or less have to play a6 next, I thought – otherwise I’d win the b pawn (turns out the engine disagrees
with me — and with my opponent…)

So it seemed safe enough. But man is it tough to move the rook to a spot like that with little time on the clock, little sugar in the blood, way too much caffeine in the system, and a lot at stake in the tournament.

29. … Bd7 30.Qg6 Qg8 31.Qa6 Nd8 32.Rd6 Qe8 33.e6 Nxe6 34.Nxe6 Rf6
29. … Rc7 30.Nxb5 Rd7 31.Qg2 Re7 32.Qg6 Qg8 33.Qc6 Be6 34.Rd2 Qf7 35.Nd4 Nd8 36.Qa8 Re8 37.Nb5 Nb7 38.Qxa7 Bd5 39.Nd6 Nxd6 40.Qxf7 Bxf7
29. … Rf8 30.e6 Bxe6 31.Re5 Nd8 32.Bxf5 Rxf5 33.Nxf5 Qf7 34.Qg6 Qd7 35.Nxh6 a6 36.Nf5
Chicken. I know. But at least my rook has a safe retreat, should it become necessary.
…Qc6 31.Rd1 Be6?
Allows the knight to return to d4 with tempo and threats.


32. … Qe8 33.Qg6!
This move is probably too obvious to merit an exclamation mark, but I don’t care. Now, the white bishop, which hasn’t really done much throughout the game, becomes a deadly threat: all the possible continuations involve threats of a mate on h7.
Nd8 34.Nxf5
and black resigned.

White threatens Nd6, Ne7 (if black moves the queen to g8), and the queen/bishop battery mate on h7 wherever the knight goes. Some possible continuations:

34. … Qf8 35.Ne7 Bf5 36.Bxf5 Rxf5 37.Nxf5 Ne6 38.Qxe6 b4 39.cxb4 c3 40.bxc3 a5 41.bxa5 Qa3+ 42.Kb1 Qf8 43.Ne7 Kh7 44.Qf5+ Qxf5+ 45.Nxf5 g5 46.hxg6+ Kxg6 47.Rd6+ Kxf5 48.Rxh6 Kxf4 49.a6 Kg5 50.a7 Kxh6 51.a8=Q Kg6 52.Kc2 Kf7
34. … Bxf5 35.Bxf5 Kg8 36.e6 Nxe6 37.Bxe6 Kh8 38.Qxf7 Qxf7 39.Bxf7 g6 40.hxg6 Kg7 41.f5 Kf6 42.Rd7 b4 43.Be6 bxc3 44.bxc3 Ke5 45.g7 Ke4 46.Rd4+ Kf3 47.g8=Q
34. … Qg8 35.Ne7 Bf5 36.Bxf5 Rxf5 37.Qxf5 Qe8 38.Qc8 Qxe7 39.Rxd8+ Qxd8 40.Qxd8+ Kh7 41.e6 b4 42.e7 bxc3 43.e8=Q cxb2+ 44.Kb1 a5


There it is. Not a flashy display of brilliance, but a slow crush.

I was quite happy to receive the praise from the big guns in the club afterwards. One of them said: “I would have loved to get a position like that!” The answer I should have given was “One doesn’t get a position like that — one creates it.” :)

2 thoughts on “Another “Best game ever” — so far

  1. I’m a fan of Dylan and a long time reader/user of Dylanchords and also an amateur chess player. The last four months I’ve been trying to improve my game a little.
    So now I look up some tabs and for the first time clicked on a link to your blog and imagine what I see: an annotated chessgame! I’m almost suspecting there to be a deeper link between Dylan and chess, but it’s probably a coincidence ;-)

    Nice win you’ve played there: first cramping his space and then slowly pushing black with his back to the wall. I’ve especially enjoyed your explanation of the London system; I’ve never tried it but now I will.

  2. Hey:

    I love Dylan and the wonderful website you have put together with good tab. If you ever want to get a game I’d be happy to play with you. I’m not a bad player!


Leave a Reply

Your email address will not be published. Required fields are marked *