How do you count/calculate in a multiple capture scenario?

I'm a beginner, attempting to develop the skill to quickly calculate the loss/gain in value that results from a series of captures, but can't seem to settle on a method that works for me.

When assessing a situation a like the one below, assuming you can't just look at the position and immediately see the answer, what does your internal monologue sound like? 

Are you doing a total tally for each side? or a piece by piece tally? 

For the purpose of this example, it's white to move, and all possible captures will be made.

How do you quickly work out if it's worth making the first capture? 

What will probably help you is to total up the number of attackers, and the number of defenders.  In your current example, white has 3 pieces attacking the e5 pawn.  Black has 4 defenders of the e5 pawn.  

(I typed this before anyone replied, so this isn't against godspawn)

Some people like to count attacks and defenders... but that doesn't tell you if it's safe in situations where the value of the pieces attack and defending are different. For example two rooks are attacking a pawn, and the pawn is defended by another pawn, that's 2 attackers to 1 defender but if the rook captures then that player loses material!

The counting method also fails to find in between moves like checks.

The best way to do it is visualize the sequence yourself. This is not easy at first, but gets easier over time.

---

But here are some tricks that might help.

After either white pawn captures and the black pawn recaptures, because that's an equal trade, my mind in a sense goes back to zero. I'm not remembering 1 pawn for 1 pawn, it's just at zero again. Then I calculate the next two captures (white plays pawn takes pawn, and black recaptures with the knight). Again that's a pawn for a pawn, so I go back to zero. I made black recapture with the knight because typically you capture with the least valuable piece.

So then I calculate bishop takes knight, and rook takes bishop. Bishops and knights are worth the same, so again I'm back at zero.

---

The other trick I'll mention is if you capture first (and the pieces are worth the same), then you can't lose material... because the best your opponent can do with his 1 turn is play one re-capture and make it even again.

So if you're capturing first in a sense you're safe... of course as long as your pieces aren't more valuable like the first example I mention with a rook capturing a pawn that'd defended by another pawn.

So this gives us another useful observation -- if you're the attacker and you're capturing first, then the only way you can win material is if you make the first and the last capture.

No offense taken :-)

My reply was just keeping it simple.

The method I use in situations like this may be totally confusing... but in case someone finds it useful...

After an initial calculation, seeing there is a lot of valuable material coming off, I would actually ignore all pawn captures until the end. Especially when it's just 1 in the beginning.

So you say first is knight takes pawn... I'd just pretend it's not even a capture. Then after _xN, RxN I'm at 0 again. Then QxR fork NxQ and I lose the knight I can go to the common R+minor for queen. Then add the pawn at the end for R+minor for queen+pawn.

And because of it you may have made the most useful reply for the OP 

Lets not get carried away.  We are here to be helpful :-)

The position isn't important, the OP is asking for the method, tips, and tricks.

 You mostly mocked him. The specific position (e.g. kings and such) isn't important. The general position is.

Here is a "massive piece advantage" and white wins material anyway:

For what it's worth, that's the system I've been using most recently, too. I guess people come up with the same system independently. I used to count material points, but I found it was too time-consuming and error-prone.

As an example, here is a game posted today in the Game Analysis subforum (https://www.chess.com/forum/view/game-analysis/any-comments-or-suggestions-welcome), with Black to move...

(Material starts as even; tally is zero.)

11...Nxe5: Black is up a queen

12. Bxd7+: Black is down a queen (the tally cancels to zero)

12...Nexd7: Black is up a minor piece

13. White has nothing else to capture (safely), therefore the last tally stands unaltered (Black is up a minor piece).

This method gets more complicated with different piece types getting captured (minor pieces, rooks, queens, pawns), but the tallies tend to cancel quickly, so there is not much that must be kept in memory.

Yes, those were my lables because we disagreed that a position was important. I said it was important to convey the general idea, and the specific position wasn't important.

Same goes for my diagram. It's not meaningless, it's the general idea.

I mean, I know you already understood this from post 1, and you're trying to be cute or whatever, but you're not even making a good fake semantic BS teenager argument, sorry to let you know lol.

Maybe it would be useful to list some common types of trades... equal is the most obvious.

But even things like rook for minor piece. Queen for rook. That sort of thing.

This probably follows the natural progression you get from experience. You don't have to think "rook for a knight" you can just think "I won the exchange."

Then after more experience you can add more layers e.g. during calculation you run into "I lost the exchange, but won a piece" and you would know this = "I have two minors for a rook."

It's the idea that you (black in that position) can optically have a "massive piece advantage" but still lose material after some exchanges.

Obviously that specific position will never happen. But e.g. I expect people to generalize that when one side has two minor pieces, and the other side only has one + heavy pieces, then material can be lost even if there are many more defenders than there are attackers.

A person can generalize this even further to all cases where there are lower cost attackers vs higher cost defenders.

---

I expect this just like the OP expects people to realize he is not interested in that specific position, but it is meant to help convey the general idea.

Actually, I already developed such a system, but I haven't been using it in real games, only holding it ready to use in my repertoire file for annotation notes someday.

Below is my system. I ignored the (0.25) point difference between knights and bishops, so the types of pieces involved in exchanges are:

pawn = 1

minor piece = 3

rook = 5

queen = 9

Now take all the possible differences of these (i.e., all the ways to "win an exchange") and you'll get mostly unique integers:

9 - 5 = 4

9 - 3 = 6

9 - 1 = 8

5 - 3 = 2

5 - 1 = 4

3 - 1 = 2

The difference values are: {2, 2, 4, 4, 6, 8}. From there you can make up a term like "lox" (loss of exchange) or "wox" (win of exchange) to produce terms like...

lox-2

lox-4

lox-6

lox-8

...to describe exactly the type of uneven exchange that just occurred. You'll have to figure out how to differentiate between the two "2" values and two "4" values, though. If you ignore pawn-for-piece exchanges, though, the problem partially solves itself. As I mentioned, I haven't yet used this system, only came up with the basics of it for future use.

I did think about adding a phrase along the lines of "obviously, the position I've just thrown together as an example", but figured that was obvious from the rest of my question.  Some posters at least, managed to get my drift and provided some very helpful  information. Thanks : ) 

That's pretty thorough

Also it seems you have two 2s: knight for pawn, and the exchange.

The pawns ones are perhaps impractical though. In a game I think it's more efficient to just consider a knight for a pawn being up or down a piece.... especially a queen for a pawn, it's just winning (or losing) a queen.

The ones that are more likely to lighten calculation loads are the exchange and e.g. queen for two rooks, queen for 3 minor pieces, and queen for rook and minor piece. "The exchange" is the only one with its own term... I know you like terms, I think this is a place where terms could help... functionally that's what I do during a real game. I don't think "queen for two rooks" because it's sort of its own (nameless) entity.

Yeah that's a good answer. I think we take for granted though that we can tell when something is adequately defended. In a game if something is safe, then you can move on to strategic considerations. But lets say the OP has the defender's side, and has no idea whether or not his opponent is threatening to win material (because he's really bad at calculation). In that case the tip to add up attackers and defenders (like godspawn said in post #2) may be the most helpful advice for him (at least in the short term).

That's a good point I hadn't yet considered. Yes, I love new names. You can see how this is another case where existing chess terms ("win the exchange") are woefully ambiguous.

By the way, I fixed my (numerous) typos in my previous post. Also, in case there are any math pseudo-wizards out there who complain that a set can't have multiple instances of the same value in it as I have shown, I suggest you look up bags (= multisets) and their representation. (https://en.wikipedia.org/wiki/Multiset). Those are bags, not sets. 

We know that (I assume), but we're looking for practical methods that work for most games. You made a good point, though: if the method a player uses is only addition of piece values, then at the end when he mentally reports the result of his calculations he won't have enough useful information from a single number. For example, getting two rooks for a queen is a good trade *if* the rooks are connected and have a fairly open board. Similarly, two minor pieces for a rook and pawn (a commonly seen trade at KB2 in beginner's games) total 6 - 6 = 0, but that zero doesn't tell you that the side that lost the two pieces is going to have a rough time of it in the middlegame. That's why keeping a mental tally of the *types* of exchanges can be important.

Not true. A bishop with open lines is more powerful than a knight in most situations. A blocked bishop, is however less powerful than a knight.