Why did he did so?

[IM Jack Rudd]

It would, but white would have been a pawn up. If black wanted to retain winning chances, 2...Rxf2+ would probably have been best.

[Roger de Coverly]

2..Bg7 would have stopped White's perpetual check.

Count the pawns. White retakes on f1 and is a pawn up.

[soheil_hooshdaran]

What's this combination based on?
[Event "?"] [Site "?"] [Date "????.??.??"] [Round "?"] [White "?"] [Black "?"] [Result "*"] [SetUp "1"] [FEN "5rk1/R3p2p/3bq1pB/1p6/1P1PNnP1/2P2Q1P/8/6K1 b - - 0 1"] 1... Qxe4 2. Qxe4 Nxh3+ 3. Kg2 (3. Kh1 Nf2+ 4. Kg2 Nxe4 5. Bxf8 Kxf8 6. c4 bxc4 7. b5 c3 8. b6 Ke8 9. b7 c2 10. Ra1 Kd7 11. Rc1 Kc7 12. Rxc2+ Kxb7) 3... Rf2+ 4. Kxh3 (4. Kh1 Rh2#) 4... Rh2# *

[IM Jack Rudd]

Some poor analysis, for starters: 3...Rf1+ is a rather stronger response to 3.Kh1 than 3...Nf2+ is.

[Christopher Kreuzer]

Some poor analysis, for starters: 3...Rf1+ is a rather stronger response to 3.Kh1 than 3...Nf2+ is.

I was about to say that... It is a nice combination, that wins a piece. White can't recapture on e4 as mate is forced. If the White queen wasn't undefended, then White might have time to capture on f8. I don't think any of the moves to save the queen really help (because of the mate on g2), so White is a piece down with further material loss to follow. White should just resign. Not much else to say.

[Roger de Coverly]

What's this combination based on?

Given Jack's line, a forced mate or loss of the Knight on e4. The ending was the two minor pieces is presumably good for Black, but not one you would want to play if short of time and without a 30 second increment.

[soheil_hooshdaran]

What's this comination based on?
[Event "?"] [Site "?"] [Date "????.??.??"] [Round "?"] [White "?"] [Black "?"] [Result "*"] [SetUp "1"] [FEN "r1b2rk1/1p3pb1/2pp1n1p/2nPp1pP/1qP1P1P1/pPN1BQ2/P1KNBP2/2R4R w - - 0 1"] 1. Bxg5 hxg5 2. h6 Bh8 3. h7+ Kg7 4. Qh3 *

[soheil_hooshdaran]

And what about
[Event "?"] [Site "?"] [Date "????.??.??"] [Round "?"] [White "?"] [Black "?"] [Result "*"] [SetUp "1"] [FEN "8/2Q2bkp/8/5pp1/pP1r4/P5P1/1KR2PBP/4q3 b - - 0 1"] 1... Rxb4+ 2. axb4 a3+ 3. Kxa3 Qa1+ 4. Ra2 Qxa2# *

[soheil_hooshdaran]

Some poor analysis, for starters: 3...Rf1+ is a rather stronger response to 3.Kh1 than 3...Nf2+ is.

Ha!ha!
true...I am a poor patzer!

[Matt Mackenzie]

And what about
[Event "?"] [Site "?"] [Date "????.??.??"] [Round "?"] [White "?"] [Black "?"] [Result "*"] [SetUp "1"] [FEN "8/2Q2bkp/8/5pp1/pP1r4/P5P1/1KR2PBP/4q3 b - - 0 1"] 1... Rxb4+ 2. axb4 a3+ 3. Kxa3 Qa1+ 4. Ra2 Qxa2# *

Well that line is completely forced, so not much debate surely?

[soheil_hooshdaran]

What is the justification for 15...e5 in:
[Event "?"] [Site "?"] [Date "????.??.??"] [Round "?"] [White "?"] [Black "?"] [Result "*"] [SetUp "1"] [FEN "r2q1rk1/pp2pp1p/2np2p1/3N4/2P1P1b1/1P2QN2/P4PPP/R4RK1 b - - 0 15"]
*

[Roger de Coverly]

What is the justification for 15...e5 in:
*

My first thought is that the intention is to follow with .. Bxf3 and .. Nd4. But perhaps not, as this leaves the Knight unchallenged on d5. Maybe then it's to get f5 in, with play against the Kingside in the style of a Kings Indian.

It's one of those difficult decisions as to whether or when to incur the backward d pawn particularly against a half open d file. The success of the Sveshnikov suggests that it's frequently a plausible idea.

(edit) It's actually Sefc v Petrosian from 1957. Sefc went 16. Ne1 whereupon Petrosian played .. Nd4. Against an engine suggestion that 16. h3 should be played, the engine suggests .. Bxf3 17 Qxf3 f5.

The Soviet players of the 1950s knew things in these Kings Indian like positions that later became common knowledge.
(/edit)

[soheil_hooshdaran]

What'd give White the advantage in:
[Event "?"] [Site "?"] [Date "????.??.??"] [Round "?"] [White "?"] [Black "?"] [Result "*"] [SetUp "1"] [FEN "1k1r3r/p5pp/Qppq1p2/5b2/8/3Nn3/P2NBRPP/1R4K1 w - - 0 1"] 1. Rxb6+ axb6 2. Qxb6+ *

[soheil_hooshdaran]

What makes 1...Qa4 winning in:
1k1r3r/ppp4p/6p1/1q3p2/2P5/3PQ1P1/b1P3BP/K3R2R b - - 0 1

[Roger de Coverly]

What makes 1...Qa4 winning in:
1k1r3r/ppp4p/6p1/1q3p2/2P5/3PQ1P1/b1P3BP/K3R2R b - - 0 1

It defends the Bishop on a2 and removes the Queen from the threat of the c4 pawn. I'm not sure I see why it's better than .. Qa5 or .. Qa6. Perhaps being closer to the King, it works better for constructing a mating net.