To Take or Not To Take? Gambits

What should you do in a gambit? Should you take the gambit or not? For example, the Albin Countergambit? If fased with a gambit, should you accept? Take a look at this one:

It depends, most e4 e5 gambits if black accepts and than gives back the material by getting in d5 that normally means it is equal.  Against the wing gambit in the sicilain I would probaby accept against the Morra I would probably decline.

It depends on how booked up you are.

Declining a gambit is more practical.Choose one line and chances are you will be more exprienced with it than your opponent.

On the other hand,accepting the gambit can reap big rewards.The risk is higher though.You will be playing into your opponent's prep.So you will have to be booked up.

In general, you should accept every gambit, except the queen's gambit.
But of course, if say your opponent plays an evans gambit against you, and you don't know that gambit very well, the only way to survive is to declaine it. You can lose soon if you accept a gambit you don't know well. 

Great insight. It helps to know that it does not matter as much as I had thought. But, in specifics, what would you do if you did accept a gambit not to have a disadvantage? What would you do to have an advantage? I am faced with the Albin Counter-gambit quite often, but I don't know how to gain the upper hand as white... any suggestions? Usually, I don't take. Please see the following and read the comment at the end:

Any tips for further play would be much appreciated for black and white!

The short answer is 'it depends on the specific gambit'. In general, gambits fall into one of three categories:

1.) Those which aren't real gambits, per se, meaning if one side tries to take and then keep the offered material, the other side can forcibly get it back, often with advantage (the Queen's 'Gambit' is the most salient example).
2.) Those which are real gambits, but legitimately offer the gambiting side decent compensation if the other side accepts (even if in theory, the compensation shouldn't be worth an entire pawn). The Volga/Benko Gambit comes to mind.
3.) Those which are real gambits and are potentially dangerous against inaccurate play, but typically don't offer the gambiting side anything more than an interesting game (objectively-speaking). With the amount of chess theory there is these days, I think it's fair to say that most gambits fall into this category nowadays, and things like the Albin Countergambit or Latvian Gambit are in it.

As for the Albin Countergambit specifically, capturing the pawn via 3.dxe5 is considered best by theory. There are some traps to be wary of (such as the Lasker Trap), but aside from that, white should be ok.
It's also worth noting that the move order 1.d4,d5 2.Nf3 and then 3.c4 avoids both the Albin Countergambit and the similar Budapest Gambit.

Also, if playing black you don't reply 1...e5 to 1.e4, that eliminates a lot of white's potential gambits that are recognized by modern opening theory.

Thanks, that makes more sense. I've been playing chess for seven years but have not taken the time to consider the benifits of my actions. I think experience would help a ton too. Thank you for the help!

It mostly depends on how good the declined lines are, really. Just ask jourself if, besides offering a pawn, the last move your opponent made archieved some huge strategic goal. In that case, you really have to grab the pawn to show that you have some material compansation for what the opponent just archieved.

For example, let's consider the king gambit. after 2.f4 white is not threatening anything in particular, since fxe5?? would just lose to Qh4+. Therefore it's entirely possible to decline the gambit with something like 2...Bc5. In the evans gambit, it's entirely possible to decline the pawn with Bb6, since it's not clear that the inclusion of b4-Bb6 really helps white. 

Things are quite different if we consider the albin countergambit. With 2...e5!?, black has just archieved a really huge strategic goal; pushing the e pawn to e5 is one on black's main goals in a d4 d5 opening (but it's normally almost impossible to archieve, and so back usually settles for the less ambitious alternative break c5). The only problem with the move is that white can grab the pawn and ask black to prove that having played e5 is such a big archievement that it con fully compensate the loss of a pawn. If left in peace with something like 3.e3?, there is no reason to think black can have the slightest problem, on the opposite we are left wondering whether white, after having allowed such a desiderable pawn break for free, is still equal or already a bit worse. So in this case you just have to take the pawn.

@bresando

That was very helpful.  Thanks!