(+9.71) 13... Nb4+ 14. Kc4 Bxd1 15. Na3 Bg4 16. Kb3 Kxd8 17. Nb5 Be7 18. a4 a5 19. h3 Be6+

Black to play (10...?) "Medium/Easy"

Black is down a pawn, which makes sense, since this was a From's Gambit (noticed the header when it popped up). It looks like White's last move must have been 10. Bf2-h4 after 9...g4.

One of the first things you want to look for here are captures, even ones that apparently give up the queen! Especially when there are interesting follow-up moves! In this case, 10...gxf3 11. Bxd8 is a sham sacrifice, since Black wins the queen back with interest.

10...gxf3! 11. Bxd8

White obviously has no choice now. There's no turning back!

11...f2+ 12. Ke2 Bg4+ 13. Kd3

These moves have all been forced. Now White can win a piece with 13...Bxd1, since Black's bishop on d8 is hanging (14. Kxe4 Bxc2+ 15. Bd3 Bxd3+ 16. Kxd3 Rxd8), but he has even better! (When you see a good move, look for a better one!)

13...Nb4+!

This greedy move wins more than a piece! If now 14. Kxe4 f5 mate! So White's next move is forced.

14. Kc4 Bxd1

It's time to finally take the queen, since 14...Be6+? 15. d5! and not only is there no mate, but Black would have to play like a computer to win this! I assumed that Black was just losing a piece after 15...Bxd5+ 16. Qxd5 (forced to stop the mate) Nxd5 17. Kxd5, but I just put this on Fritz, and the silicon "beast" came up with 15...Nxd5! 16. Kd3 Rxd8! and Black is slightly better, despite having only two pieces for the queen, due to his active pieces.

15. Na3 Bg4 16. Bxc7

Or 16. Bh4 Be6+ 17. Kb5 a6+ 18. Ka4 b5+ and White will have to give up another piece to stop the mate.

John, I think I mostly agree with you. In a real game, we often don't need to calculate everything. Indeed, I made much the same point in Nezhmetdinov vs Kotkov, 1957.

But there are situations when you do need to calculate a line which is 4 or 5 moves deep, or sometimes more. And then I find that the horizon effect kicks in ... the longer the variation goes on, we all start to lose track of where pieces are. Some players can also lose track of the material balance (what Dan Heissman calls "counting"). How many games are lost because of the move that comes right at the end of a combination? Or forgetting that you are actually a piece or pawn down?

Granted, in this puzzle you don't need to see all the way to mate because grabbing the white queen is enough to win. But as an exercise to practice a vital skill it can be quite useful to counter the horizon effect and improve counting.

I suppose it comes down to your reason for being here. Some folks like to analyse to the nth degree to expose the "truth" of a position. These guys go for the long lines and exhaustive analysis. Some folk like to extract just enough to work out what they would play in a game. And for these exhaustive analysis is not so important - once they have got to a winning position, they stop.

And some folk come to learn, which probably puts them halfway between the two extremes. I tend to flip between all three modes, depending on mood, day of the week and how much time I have before jumping in the car for the commute to work.

Flocking stupid game title