This move by black sets up 2 ideas that I can see. The first is ...Ba7. This frees the queenside rook to get into the fight, say on c8. It also lines up a bishop+queen battery aimed at my king. In order to activate this, however, Black would have to move his c-pawn, and I just don't see that happening soon. So to deal with this threat, I should just make sure I keep that c-pawn cemented in place and maybe someday play Kh1.

The second idea is ...Bb5, to attempt to exchange off some pieces and maybe give the c-pawn some mobility. I see 3 ways of addressing this:

1) 20. Qd3 Bb5 21. Nxb5 axb5 22. Bxb5 c4+ seems like a problem as my queen is en prise and my king is in check. I seem to lose at least piece in these lines. And if I don't take the bishop on move 21, then I'm not sure why my queen went to d3.

2) 20. a4 seems simple and effective enough. Black may plan something like ...Bd6, ...Qc7, and ...b5 -- hmm, no that's bad as after axb5 his a-pawn is pinned due to the rooks. Therefore, he may plan ...Ba7, ...Qc7, and ...b5. This gives me time, though, to play moves like Qd3 (and then Bb2 to connect my rooks?), Kh1, and/or a5, which should take the sting out of such a plan.

A downside of 20. a4 is that it makes b3 a backwards pawn, but it is currently protected by my queen and LSB. It almost makes my bishop look like a big pawn as it's not doing much a pawn can't do, but if it ever does make a capture on b5 then it aims at e8, and also if I ever play d5-d6, then the bishop is aiming at f7, so I'm still happy with it.

3) Play something else, as after something like 20. Bb2 Bb5, I don't have to take on b5. What I don't like about this is that if Black captures on c4 (...Bxc4 bxc4), my pawn on c4 seems weak. If Black can play ...b5 at some point, that also weakens my d5 pawn.

So I'm leaning towards #2, 20. a4. What will Black play next? Maybe 20...Bd6 (and now with 19...a6, there's no way I'm getting Nb5 in ever). And then what for me?? Maybe 21. Bb2 Ng4 planning ...Ne3 is annoying. Maybe 21. Qf3 planning Qg3 (though it puts my queen behind a pinned f-pawn, I think it makes Bb2 more effective by coordinating my queen and bishop on the kingside).

I'm not entirely sure still, but as 20. a4 best deals with the immediate issues brought about by 19...a6, I find it hard to fault it as my choice.

21. a4

I don't really mind this pawn structure change. White also gets the b-file, and b6, and is better dynamically.

Imo there's no straight way to crash through black's defences anymore. So this loo sabout right. White is just going to improve his pieces across the board, and a4-a5 is about the best he could improve his queenside for now.

As white builds up his position, the rook on a8 is not going to be too happy for a while.

I mean after 21. a4 Bxb5.

You mean after 21. a4 Bxc4!

haha oops

The b8-h2 battery is coo.n's model in a Christmas hat. As long as it's in his head, he won't want to settle for the dumpy, run-of-the-mill 20.-Bd6, the most obvious and logical move.

One wonders how he talked himself into thinking it held promise. It's far too slow. Investigating sample variations, as RT did in his own game, would make that obvious. But he has also talked himself into the notion that variations are secondary to ideas and general principles.

I haven't managed to think of a better move than this possibly stupid Qd6, which does increase the pressure on f4, but maybe just g3 is quite safe for White after it. Bg4 is also an option, but I don't want to wake his queen up that early - it won't benefit me, as he can move h3 later with a tempo and g4.

However, it looks like I can prepare the b5 break by Ba7 and Rb8 because the a6 pawn will be protected by the queen.

This confused me; 21. Ø Ba7 22. Ø Rb8 23. Ø b5 24. axb5 axb5 25. Rxa7. I guess he anticipates White playing the Benoni-style a4-a5 (and then axb6 ep upon b7-b5), but it would be odd for him to omit that from his commentary.

Ok, let's figure out what 20...Qd6 has accomplished. The first thing I see is that it opens up for ...b5. However, this still just loses the pawn, not to mention that after ...b5 axb5 Black can't recapture with his a-pawn due to the rook pin. Hmm, so that's not likely what he's planning.

What makes more sense is that Black has just created a Q+B battery aimed at my h2-pawn. If my f-pawn were to disappear, my king would be in a lot of trouble (especially with the e-file blocked off due to his e8-rook). On a side note, I'm thinking now that Black played 19...a6 in order to be able to play 20...Qd6 without being harassed by Nb5. Back to the current position, Black may be planning 21...g5 to try to get my f-pawn out of the way. If I did nothing, this actually wouldn't be a threat as after ...gxf5 Bxf5 and Black's not getting to h2.

Something else in Black's arsenal is 21...Ne4 22. Nxe4 Rxe4 gets his rook positioned to attack f4. At first I thought this and the preceding paragraph meant I couldn't play a move like 21. Bb2, but I believe that to be wrong now: 21. Bb2 g5 22. g3 Ne4 23. Nxe4 Rxe4 24. Be5, and I don't think Black does well with 24...Rxe5 25. fxe5 Qxe5. Also in this line, Black could try to capture on f4 with 22...gxf4 23. Rxf4 and now he can't even play ...Ne4. However, I don't see why I want to play Bb2 in such a position anyway. The bishop now is doing more important work holding onto f4.

Another way for Black to get at f4 (and also g3 in case I play my pawn to g3) is ...Nh4. That currently loses to Qxh4, but perhaps Black can play ...Bg4 first. For instance, 21. g3 Bg4 22. Qd3 Nh4 and I just don't see anything here for Black, especially with my queen on the 3rd rank to guard g3. Another try for him in this example line is 22...g6 23. Bb2 (after Black plays ...g6, my bishop seems better on b2 than before) Bf5 24. Qd2 Ne4 25. Nxe4 Bxe4 26. Be5 Qb6 27. e6 and this seems nice for me. Of course, there are some deviations for both sides to consider, such as 26...Rxe5 27. fxe5 Qxe5 28. Rxf7, which looks good for me, or 24...Nh4 which can be met with 25. Qg2 perhaps with Rae1 to follow soon.

Overall, I like the idea of 21. g3 as it immediately blunts the Black Q+B and avoids me getting into trouble. But it does open up the pawns in front of my king, so I should consider other options. Only issue is that I'm not sure what other options I have. One thought is 21. Ra2 to get that rook into play without having to move my DSB. But I don't know where I'd want that rook, so it doesn't seem like a very useful move.

Right after making my move and actually seeing it on the board, I noticed something else. Black may now play 21...Bh3 but I don't think I have anything to fear after 22. Re1.

As a general comment, I would say that ganstaman is too focused on his opponent's possibilities to the point that he plays too defensively, while **** is too focused on his own potential and plays too aggressively, ignoring his opponent's resources.

I think White is still a bit better, but g3 does nothing for his game.

I think cooon is significantly higher rated so this partly explains it, psychologically.

I don't understand this position at all anymore.

All right, it looks like I've scared White and he's decided to cement the b8-h2 diagonal; but of course he's weakened the light squares around the king, and my LSB now has a natural square on h3; as my king is not on the kingside, I also have the h7-h5-h4 option.

Before I launch a kingside attack, though, I need to coordinate my pieces. As my DSB is out of play anyway, I'd better put it in ambush on the a7-g1 diagonal, keeping c7 available for the king in order to connect the rooks, so that I keep the e-file under control even after White plays Re1. I'm also not losing hope for the b5 break (maybe by means of Kc7 and Reb8 to keep a7 protected after axb5 axb5); the idea is to move the c5 pawn out of DSB's way; but a big downside is that I'll need to exchange the LSB on b5 then, so it looks as though there's no way I can use both bishops in the attack... that's a bit too tricky for me now, seems a suboptimal way because the light squares around his king are easier to access.

Qe7 would be tempting if White couldn't connect the rooks and then kick my queen out by Re1. I guess the most natural way to activate the queen will be along the 6th rank if I move the knight out of her way at some point.

Oh well, I want to see first what White does. But I'm sure my DSB belongs on a7 anyway, otherwise my development would be too disrupted.

Haha, I didn't take f4-f5 into account, it's actually a pretty dangerous move because it enables a bishop outpost on f4. Maybe I'll even need to castle kingside artificially, e.g. 22. f5 Ke7 23. Bf4 Qb6 24. Re1+ Kf8. Ewww, that's ugly

Probably won't work. 22. f5 Ke7 23. Bf4 Qb6 24. d6+ Kf8 25. Nd5, and everything looks awful for Black. 24.-Kd8 25. Nd5 Nxd5 26. Qxd5 Qc6 27. Qxf7 Qe4 28. Rad1 looks dismal too.

Hmmm, so he's not trying to go down the b8-h2 diagonal. Instead, it seems that his plan is something along the lines of (if I just pass my move now) 22...b5 23. axb5 axb5 24. Nxb5 Bxb5 25. Bxb5 c4+, with a discovered check. However, I don't think this works for 2 reasons:

1) After 26. Kh1, Black is already a pawn down and his e8-rook is under attack by my bishop. If he saves the rook, I can follow with bxc4 so that I will be up 2 connected passed pawns. If instead 26...cxb3 27. Bxe8 Kxe8 28. Qxb3 Qxd5 29. Qxd5 Nxd5 (not 28...Nxd5 29. Rd1, I believe), I'm up the exchange rather cleanly, if I see this correctly.

2) If I'm wrong above, I imagine I could try 24. Rxa7 Rxa7 25. Nxb5 Bxb5 26. Bxb5 Rg8 27. Bc4, going down the exchange for a pawn. Maybe worth a try. If I don't pass on move 22 and instead play 22. Bb2, this would also set up for Be5 soon (which is possibly more awkward for Black if I play 27. Bc6 to almost trap the queen).

So then if this doesn't work, could Black be planning something else? Perhaps he is just freeing his rook, preparing ...Rc8. Oh, maybe it's ...Rb8, which would make the move ...b7-b5 much more annoying for me. Well, maybe, as it also leaves the defense of the bishop on a7, making it vulnerable to my rook. Let's investigate 22. Bb2 Rb8 23. Kh1 (just as a placeholder move, not sure I like it) b5 24. axb5 axb5 25. Rxa7 bxc4 26. bxc4 Rxb2 loses me a piece. I'm struggling to find ways to improve this, however. A move like 23. Qc2 should help but after 23...g6 planning ...Bf4, I'm not sure where my queen should go. Maybe instead (22. Bb2 Rb8) 23. Qd2, and if 23...Ne4 24. Nxe4 Rxe4 25. Bxg7 wins a pawn that I hope is safe. Also, 23. a5 might help so I can capture the b-pawn e.p.

Overall, it seems I like 22. Bb2. If my knight moves, the bishop will be well placed on that diagonal. Also, it connects my back-rank finally. I don't think I still need my bishop defending f4. Should Black replace the bishop on b8, maybe I'll need to put my bishop back on c1, but I don't think that's actually necessary. I can probably play Qd2 and R(a or f)e1 in the upcoming moves if Black plays ...Bb8 or ...Bg4.

Phew, the 22. f5 Rc8 23. Bf4 Qf8 24. d6 Bb8 (planning Rc6) line would give me more headache.

Now his DSB is away from c1-h6, which has created a happy opportunity for me to penetrate on e3 by the knight. He can either retreat the DSB to c1 and exchange it on e3 (losing the bishop pair, which will be a bit unpleasant for him) or move a major piece away from the fork, but even then I'll fork another major piece and the LSB, grabbing the latter, which is awesome because 1) I'll retain a decent bishop pair for the endgame, 2) as said above, the light squares of his kingside are weak and he'll lose an important defensive bishop.

A downside is that his passed pawn will become protected (unless he recaptures on c4 with the queen), but the a-pawn will become isolated and possibly up for grabs, though I can't play a5 soon because I don't want his knight to come to b5, which also means that c5 will become my weakness because he'll play a5 himself. However, I can play b6 then, creating a (remote!) passed a-pawn for myself.

So basically, I'm going for a dubious tranformation of strengths - an exchange of my slight pawn structure advantage for his bishop pair.