The other 2 options lose immediately:
a) 18..Bg5 19.Rxg5+ Kh8 20.Nxe6+ f6 21.Bxf6+ R(N)xf6 22.Nxc7, White is a piece and two pawns up b) 18..Kh8 20.Nxe6+ f6 21.Nxc7, White is a piece and a pawn up

Back in the main line:
19.Nxe6!! Q is on threat, so either Black moves the Q or takes the N. A) If Black takes the N with:
i) 19..Rxe6
20.Rxg7+ Kh8
21.Rxf7+ Nf6 <(21..Rg8 22.Kg7+ Kh8 23.Rxd7+ and Rxc7 1-0)> 22.Rxc7, White 1 piece, 2P up!
ii) 19..fxe6
20.Rxg7+ Kh8
21.Rxd7+ and Rxc7, White 2 pieces up!

Your Chess knowledge led you to find and feel the validity of 17. Qxh5 being a winner for White. If getting the entire puzzle is a Home Run today, then simply feeling/finding the correctness of 17. Qxh5 gets you on base; as if you get an Ichiro-patented base hit!

Excellent stuff and here's to you hitting one out of the park next Sunday!!

Cheers,

White has the opportunity of opening lines to attack the black king with 17.Qxh5 gxh5 18.Rg3+ (18.Nb5 axb5 19.Rg3+ Bg7 20.Rxg7+ Kh8 21.Rxf7+ e5 - +):

A) 18... Kh8 19.Nxe6+ f6 20.Nxc7 + - [N+P].

B) 18... Bg7 19.Nxe6 (19.Rxg7+ Kh8 unclear)

B.1) 19... fxe6 20.Rxg7+ Kh8 21.Rxd7+ + -.

I still stuck with <19.Rxg7+> instead of an immediate <19.Nxe6>. I guess the real beauty of the puzzle lay in the various pins along the a1-h8 diagonal, the d-file and the seventh rank once Black's rook was pinned on <e5>.

click for larger view

Fritzie says that there are four sensible queen retreats from this position - Qc6, Qc5, Qb8 and Qc8. All score between +1 and +1.25.

Let's stay with the game for a little while longer. After 23...Qc5 24. Ne3, Fritzie offers 24...Bc8 (+1.08), 24...Ba8 (+1.49) and 24...Qc8 (+1.98)

Two more game moves 24...Bc8 25. R7d5, and here again Fritzie is offering a multitude of choices - Qc7, Qa3, Qe7, Qc6 and Qb4 all occupy the ground between +1 and +2.

In other words, it's a veritable thicket of possible variations. White doesn't haven't many moves to force black's reply. That in turn means that black has lots of choices. Which makes it very very difficult to calculate all the way to move 32.

across after hint it sac in h5 arc, rook over in sat I light rookg3+

up bishop i net in covering g7 hole nest hustle in takee6 to release

bishop rook might along axis ally together broken chain gang rook up

im seventh heave now king to corner and rook is an animal it seem in

mopping up ply it arrow for f7+ either in have re5 to juxtapose

again bishop flight enact in b2xe5+ re ignoble defeat staring down

the barrel it yeah in dread rook sweep chances to dream away it h5

in delight cog rook now has c7 ape go at peg back house queen

dressing down c5 looks like plus for white of course I cant be sure

i gnaw angle for opening position (pawns could flow) freely it d5 in

god we true street in dfiles fcolomn promote again c6 ar good in

ghosting along here h5 wins on the spot re5 as on apost in blocking

b2 however, i neap tide flowing box inc. At rook gg rescind coats

In regards to everything else, as a fellow Chess player and friend in the game, I want to say that I hope you continue learning, winning more games, and improving your rating!

That goes for myself too!

Keep up the good work!

<GauraNitai: Finding Qxh5 and Nxe6 and then it's quite a smooth ride. You're done if you fairly see till Rxd7>

These are just two examples of posts that grossly underestimate this puzzle. The game continuation is a brilliant win, but I question whether the winner or any of our puzzlers could see through all the immense complications to the win. Even our best engines could not do it in an OTB time control - the game line is 29-ply deep and more to prove the win.

I believe this is a psychological puzzle: why, in a position did white decide to go "all in" with the Q sac? Did he need the win, instead of just a chance to win? Did the clock situation suggest that white expected to find better threats than black could find defenses? Or did he just, "feel lucky, punk?"

White to play (17.?) "Insane"

I think the first move is fairly obvious, but maybe there something tricky about the follow-up.

17. Qxh5! gxh5

Obviously forced, or Black is just a piece down.

18. Rg3+ Bg7

Or 18...Kh8 19. Nxe6+ followed by 20. Nxc7.

19. Nxe6 Rxe6

Or 19...fxe6 20. Rxg7+ Kh8 21. Rxd7+.

20. Rxg7+ Kh8 21. Rxf7+ Re5

Or 21...Kg8 22. Rg7+ Kh8 23. Rxd7+ and 24. Rxc7.

26.Qxh5 to open the g file
26.............gxh5
27.Rg3+ Bg7
28.Nb5
Bishop 'eyes' on the pinned Bishop on g7 and Knight attacks the Queen <if...axb5 29.Rxg7+ Kh8 30.Rg5#>

28............f6
29.Nxc7
White falls ahead by at least a Knight.

If on move 26, Queen sac is declined:
26.Qxh5 Bg7
27.Qd1
White has gained a Knight and a possible continuation could be:

I owe <ChessGames.com> an apology, and everyone who has ever read my solution posts and trusted in their integrity.

When I saw the puzzle position, at first glance so simple, I failed to examine it in sufficient depth, failed entirely to consider that, for example, Black might capture on e6 with the rook and later interpose on e5 with that rook, or that he might *not* accept the exchange offer on g7. These omissions were sufficient to make of what I called a trivial march to victory a veritable _Aeneid_ of trials to undergo on the way to a longed-for but elusive triumph. Over the board, they might have sufficed to turn a promising position into a hopeless wreck of White's game.

This, in itself, could be excused. Anyone can be careless; anyone can fail to see all the variations in a complex of thrusts, parries and ripostes.

But my sin was not carelessness. It was dogmatism. I set about "solving" this puzzle with all the earnest certitude of a half-educated whelp setting out to prove that the Earth really *is* flat after all. My thesis: that <CG.com>'s puzzle picker was incompetent, using this (to my incorrect analysis) simple puzzle in conjunction with some other recent offerings that have proven easier than billed as evidence.

In the end, all I proved was that it does not do to approach any undertaking with prejudice.

Although this puzzle was perhaps not as complex as some Sunday puzzles past, it was certainly not simple. There really were "tricky variations," and had I not been blinded by a false thesis, I might have found them.

Three different posts, but all questioning whether we need to see all the way "to the end".

With the greatest of respect I beg to differ. POTDs are not puzzles in the traditional sense. Instead, they are positions from real games.

This means that there isn't always one "correct" solution. This is not "white to play and mate in 15". We are challenged to find the best move in the position. Sometimes we can chase that best move down to a resignable position, a draw or a forced win. But not always.

Today's puzzle is a case in point. The combination starting with 17. Qxh5 does not lead to a forced mate or a certain draw. By move 23 we get to a position where both players have many choices, but white has a small material and positional plus. There is no point in looking for a clear "solution" from that point. As I said, this is not a puzzle.

I recall reading a letter to a hifi magazine. One of the hifi hobbyists was a fan of early punk rock, but was disappointed with the quality of the sound reproduction he was getting fro his oh-so expensive equipment. No matter how much he spent or fiddled with the settings, the music always sounded muddy, imprecise, raw.

The magazine's reply has stuck in my memory ever since. They said: "The early punk records were not recorded with much fidelity themselves. That was part of the appeal. So don't go looking for a precise sound. It just isn't there."

When it comes to a saturday/ sunday POTDs, somtimes the same thing applies. From time to time we get a forced mate (like yesterday). But more often we shouldn't go looking for a definitive solution.