So the question in the title is certainly very position dependent, but here's the general set-up: let's say I'm in a certain position in a game and I feel I have the advantage (either materially or positionally). I narrow down my next move to 2 choices.

One is safe and solid, should maintain my advantage without really doing much. While it doesn't throw away my advantage right away, it might be too passive and give my opponent the opportunity to get some play against me. The other move looks riskier, but tries to get me more of an advantage. It looks like it should be good, but I simply can't calculate it well enough to see if it really works or instead throws away my advantage and maybe loses. So I'm torn about which to play.

Here's the example that sparked this question. It's from our Ruy Lopez tournament on chess.com. This game is ongoing, but we're at the endgame 20 moves beyond the position diagrammed below (so be sure to comment only the below and not where we're at now).



Black to move. The last 2 moves were 22...Bxe4 23. Ng3. I just won the pawn on e4 and now need to move my bishop. I can play either:

1) 23...Bb7. Gets the bishop to safety while maintaining it on a good diagonal and protects the a6-pawn. But it doesn't do much else for me and I'm not entirely sure what I'll do to make progress after that. I do have the bishop pair to his knight pair, but I don't know if I'll be able to open up the right diagonals to make my bishops useful.

2) 23...Be3 with 24...e4 most likely being played next (and ...f5 soon?). This tries to get more than a pawn out of the position. I place my bishop more aggressively and grab space with my pawns. I'm opening up for my DSB on the kingside, where my queen is already well placed. But, if nothing comes of this, I may end up losing my a6-pawn, allowing his rook to infiltrate my position. And I hope he won't be able to trade one of his knights for my LSB, which likely won't be in my favor.

So, which would you play here? Any general thoughts on such a dilemma?

There's a lot to analyse here but it would take me something specific to reject Bd3. The worries you mention don't seem rational

"I may end up losing my a6-pawn" - Rd6 or Re6 look like they defend it pretty well and improve your position

"I hope he won't be able to trade one of his knights for my LSB" - both are 3 moves away even if you didn't have massive threats for White to deal with. There is no way White will ever be making this exchange without concessions (massive time concession + you establish a passed pawn on d3 or you penetrate on the d-file).

On the other hand, Bd3 is a forcing move that comes with tempo. You might even be winning material immediately, what happens after Re1 Bc2!? (or Bc4!?). After Re3, you have f5 or c4 with serious threats against the exposed R.

Clearly Black is better after either move but the title of the question presents a false dichotomy; in general you should play more conservatively when you are winning but Bd3 doesn't take real risks, it's just a slightly more active choice.

Generally your idea should be to keep control of the position and prevent counterplay. At first I thought Bb7 was good, but White can start to unravel with Qc2 and Ne4 and the c5 pawn can become a target. So probably Bd3 is best. Maybe Bd3 Re3 Qg6 to get out of the way of Nf3 and then follow up with f5-f4 (Qg6 is also defending a6).

Actually I guess Bb7 Qc2 Bd6 is simple enough. White has no real way to create counterplay. So I'll say that Bb7 and Bd3 are probably equally good.

Both moves look OK, but my assessment is that Bd3 is just better, and you already have the correct continuation in mind. Who cares about that a6 pawn, anyway? White can have it. With e4/Bf4/{e3,f5} coming, it won't do him any good. That's just a reason to talk yourself out of taking a risk.

I think RoundTower has it right.

If your advantage is so great that you can win without permitting any counterplay, do that (minimize your effort to convert). But here, I feel like you need some more work to do...

Personally, and I don't know that this is really commendable, I tend to take the strength of my opponent into account. The stronger the player, the more likely I am to try to play the principled continuation. (Against weaker players, I have a better chance to outplay them with a minimal advantage, and minimizing my errors is more important than maximizing my opportunities.)

Slightly off-topic as usual but one thought that came into my mind when thinking about doubling cube decisions in backgammon:

How many objectively suboptimal decisions can we take based on minimizing risks against weaker players before we actually play too weakly for the whole principle to apply? Part of the reason that we are stronger should be the capability to evaluate risks instead of avoiding them on principle.

Anatoly Karpov's answer:

That is his answer to a different question

Agree with Sholar. Bd3 is what i would play no doubt. RE1 looks forced then BC4 looks good, if RE3 then E4 is a monster.

Both moves leave black much better, white simply blundered with Re2.