Move 16 - Candidate Discussion

16. ... Qd8 is no different than playing it in move 15. It will drop a pawn. See the 15. ... Qd8 line i posted in that thread.
I agree with Estragon that 16. ed can be disregarded.
16. a5 attempting to trap our knight. If 16. ... Nxa5 17. Rxa5 Qxa5 18. Nxe7+ Kh8 19. Nxc8 and white is up a bishop.
16. ... Nxd5
17. Nxd5 Bxd5
18. Qxd5 and we're left with a few choices, but they all still end up losing a pawn.
18. ... b6
19. ab Nxb6
20. Qd3
wins a pawn. 18. ... Rc5
19. Qxb7 Rxa5
20. Rxa5 Qxa5
21. Rd1 Bf6
22. Be2 will win a pawn.

IMO the following is a sharp, forcing line worthy of in-depth consideration (please also read the comments in the diagram)
{edit]: 19.a5 variation has been added

dsarkar, we should also consider the same diagram but replace 19.Qb3 with 19.a5. Then at the end, we no longer have the a4 pawn as a target. Regardless, this line is actually making me take another look a Bxd5, which I had pretty much thrown out already. It doesn't look that bad!
I think somebody already mentioned it but another thought I had was tha even though an eventual Bg4 threatens to trade our good bishop, after Bxe6 fxe6, white has one less defender of d5 and we have one more (and a cheap pawn nonetheless!). I think that the advantage is definitely worth giving up our "good" bishop, especially with the added bonus of the half-open f-file. So Bg4 isn't a move to be feared.

My argument that 16. exd5 can be disregarded is not simply a positional one. It is that the possible good plan to attack our kingside on the light squares will not work after the move 16. exd5 here (to open up the b1-h7 diagonal, because we would be able to play ...g6 and/or ...f5 with no problems and stop the plan dead.
After the expected 16. Nxd5 all moves other than the two already mentioned lose material in an unhelpful way, so our candidate moves are unusually easy to see.

In dsarkar's analysis, I consider the line 20. c3 Nc5 21. Qc2 Ne6 the mainline, and that black is worse at the end, especially after 22. g3! which restricts our knight and prepares Bg2. White has the two bishops, no vulnerable weaknesses. We have a bad bishop and a backward d-pawn (which white can ensure with Rd1 soon). White has the possibility of a well-timed f4, maybe f5 after, creating new problems for us. I believe the mainline after 16. ... Qd8 is preferable as we keep two bishops and can probably play g6 and make an f5 break ourselves later, a plan that has a good track record.
[Both dsarkar's and my comments should be in a 16. ... Bxd5 thread, but it does not exist yet!]

Since we are pretty sure that Nxd5 is Natalia's next move, I propose that for this move, we accelerate our voting significantly this round, and try to have our final move picked by the time voting opens so that we can all vote right away. This will give whatever move we vote for a huge opening lead, and then we can do the usual tactic of posting the vote results so that we continue to get more votes. We can then also spend the next 6-8 hours continuing to plug whatever our choice is in the main forum so that it gets most of the support.
Also, on the topic of actual moves, Qd8 looks better to me because after the exchanges, we just can't do anything in the final position that works that well. With Qd8, however, we may be able to either A) force a white pawn to d5, which cleans up a lot of mess or B) create a more dynamic position in which we are actually doing something rather than waiting for white to pile up on us.

D3V1L - That looks pretty solid, and I am not sure that we are even down a pawn in any variation that I can find after we play Rxc2. Our pawn on e5 is hanging, but after 21. Rxe5, Bf6, 22. Re2 (to hold the b-pawn, though I guess something else is possible here.), Rxe2, 23. Bxe2, Rd8, I think we get the pawn back with a playable (and possible a bit drawish maybe) position.

D3V1L - After Rxe5 (which I think has to occur on move 21 or it is not going to happen because we can defend e5), and then Bf6, white has to move the rook somewhere, I just picked e2 because it defends the b2 pawn, but it certainly could go elsewhere.

Holy smokes D3VIL, that looks like a great line! It's not even a pawn sacrifice because of Qxc2, which works because of the Bc5 response to Qxb6. The one improvement I would suggest is in the 19.a5 variation, I think Nxa5 is stronger than Bd6. I strongly suggest that everyone looks at this line; right now my support is 96% behind Bxd5 because of it. And this morning, I was 100% behind Qd8.

Noodlehead - In looking at your diagram, Qa5 (after Qxb6, Bc5) actually looks like it more or less kills this line, unless someone has a good idea of how to handle it. Qxf2 does not do anything, and Bxf2 gets us the e-rook, but then we will have a rook for two bishops, which basically guarantees that we will lose and badly.
We have only 4 candidate moves:
15...Nxd5 confirmed
16.Nxd5
16...Bxd5
16...Qd8
16.exd5
16...Bf5
16...Bd7
Should we create the 4 threads right now?