Anyway, might post some questions in here.

Starting with:



Played g3 because I didn't want to get back-ranked*. Analysis tells me this is a lost position, but other moves such as equalise. What's the logic?

*Re1 also works but unless I move a pawn the back-rank threat will remain.

Perhaps you can compare the two lines and see what the computer does and what causes the differences.


Sent from my iPhone using Tapatalk

Moving the g or h pawn makes sense to avoid back-rank mate threats now and in the future. g3 weakens all of the light squares around your king. Black only has a LSB. So that doesn't seem ideal. h3 strengthens the light squares around your king and accomplishes the same goal of avoiding a back-rank mate. h4 does that and also gains space, hampering the position of the black king. However, there may be a concrete reason h4 is better than h3. But both make more logical sense than g3.

My guess is h4 is followed by h5 attacking the King, and if Kxh5 then Rxg7 keeps the king out of the game forever and leaves 3 isolated pawns for black. So you are down a pawn but black can't win with the king trapped on the H file.

After h5 the only way to defend g7 is Kg6 which can be attacked by the dark square bishop

Note that black's light square bishop cant defend the h5 square, since Be2 loses the bishop.


Disclaimer: I also suck at chess. ~1300 Rapid and <900 Blitz on chess.com

Edit: Analyzed it for 5 min or so. Looks like h4 h5 is the general plan for white here. Black responds to h4 by playing h5 himself but that still leaves a weak, backward g7 pawn that the king has to give up to get into the game.

Black still has a passed pawn and the promotion square is the right color so black is still pushing for a win here, but its only like -0.5

Pretty much this. The idea of creating space for your king to avoid back rank mate is fine. The problem is the only square you opened for your king is g2. Your opponent has a piece that can still control g2 that you have no way to contest, so you haven't really created any air. A move like h4 create air on h2, which your opponent cannot control, so thats a safe spot.

As for sucking at chess, trust me - everyone here knows they suck at chess, including the master level players. It's just a question of relativity.

The general advice above is good, but here g3 gives black a forced winning tactic that doesn't exist after h4 or h3. Can you find it?

Nice catch, I didn't even look at it with that level of detail. Tim is right - there's a very specific reason g3 loses beyond just the positional considerations described already.

Thanks. Is it best to put pawns on the same colour as your opponent's bishop? Putting them on the opposite colour is intuitive to protect them from capture.

p.s just had this nice checkmate, haven't seen the pattern before.

Can you post this game? How are no pieces developed?

It depends. In that position you have opposite colored bishops. You are a pawn down and trying to draw. In that case, putting all the pawns on dark squares makes sense. It's only dangerous here because there are also rooks on, and your king can get in trouble. 1.g3 loses to:



So 1.h4 h5 2.Bc5 controls d4 preventing the pawn advance, and protects f2. g2-g3 can be played later if rooks are exchanged, or if you can get your king centralized so that it won't get mated as in the spoiler. Another danger is black getting his bishop to f3 or e4 and renewing the back rank threats. With B on f3, h4 and g3 played by white, Rb1+ Kh2 Rh1+ is mate.

In other kinds of positions you don't always want to put your pawns on the same color squares as your own bishop, because it might restrict your bishops mobility. Also, there are times where you can use your pawns to control squares of one color, and use your bishop to control the other color squares. If you had put all the pawns on the same color as your own bishop, there would be nothing to control the other color squares and maybe the opponent could invade on those squares with the king or other pieces.

1. e4 e5 2. Nf3 d6?! { (0.18 → 0.85) Inaccuracy. Nc6 was best. } (2... Nc6 3. Bc4 Nf6 4. d3 Bc5 5. c3 a6 6. a4 d6 7. O-O) 3. Bc4 { C41 Philidor Defense } c6 4. O-O d5?! { (0.60 → 1.48) Inaccuracy. Bg4 was best. } (4... Bg4 5. d4 Nd7 6. Qd3 Bh5 7. dxe5 dxe5 8. a4 a5 9. Nbd2) 5. exd5 cxd5 6. Bb5+ Bd7 7. Bxd7+ Qxd7?? { (0.97 → 3.54) Blunder. Nxd7 was best. } (7... Nxd7 8. Re1 e4 9. d3 Ngf6 10. dxe4 dxe4 11. Ng5 Nc5 12. Nd2) 8. Nxe5 Qb5?? { (3.70 → 9.05) Blunder. Qf5 was best. } (8... Qf5 9. Re1) 9. Re1 f6 10. Ng6+ Kf7? { (9.82 → Mate in 1) Checkmate is now unavoidable. Be7 was best. } (10... Be7 11. Nc3) 11. Nxh8# { White wins by checkmate. } 1-0

After Rb3, white has Re3 blocking the check and defending the bishop. You dont move the king.


I was wondering what tactic your were talking about that won on the spot. Looking at the position with a computer black is still winning after g3 but just because its up a passed pawn, there isnt some clean tactic that wins a piece or anything.

Oh, duh. I was thinking with the pawn on d4 already. OK so 1.g3 d4 or 1.g3 Rb1 2.Kg2 d4 threatens that, with an enduring initiative. But I don't see a clear win, and the engine doesn't either. While it gives a -2.60 evaluation, this is not increasing with search depth even out to over 40 ply (20 moves). If there were a win it would show within that depth and the evaluation would skyrocket. Basically the engine is using repetitions and long lines of defensible threats to push the lack of a win past its search horizon.