There really are a lot of ways one can study the endgame. Rene's method of asking questions (why this move? why not that move?) is one of the most effective, and not just for endgame study.

Regarding the initial question, I think a common reason for a weak endgame is a lack of practice. In the OP's case it was that he was usually winning before the endgame and hadn't built up the necessary experience to adjust from 'middlegame mode' to 'endgame mode', but you could also take the case of young players who fall into the habit of offering a draw because they find the endgame boring or don't have any idea how to win.

For that reason, playing out the Chess.com endgame drills would be a logical step - I know it helped me in this area.

If one is still stuck in the endgame after following some of the suggestions in this thread, it may be that you are overcomplicating things. Then I can suggest a simple approach:

Is the endgame dynamic or strategic?

If it's strategic, the priorities are:

- Piece activity
- Creating/fixing/liquidating weaknesses. Note that a passed pawn of ours counts as weakness for the opponent, as they will have to place a piece to stop our pawn queening.
- Stopping the opponent's counterplay
- Exchanging (if you think the more simplified position is winning)

If it's dynamic, then concrete calculation is the most important, so we may add the following:

- Forcing tries
- Pushing/creating passed pawns
- King safety (if there's more than 8 pawns worth of non-pawn material on the board)
- Prophylaxis (anticipating the most unpleasant idea of the opponent)

I didn't do solitaire chess with them, but what I did do was annotate them. I wrote in the book every question I had and  investigated without a computer  until I had the answer, which I wrote down next to it. What's the point of this move? What if x.... But why not y?

This worked for me. Plus when I reread the book I can see what I was thinking and how far I've come (what was a question then is obvious now).

You won't regret it. You become unafraid of simplifying; you become sensitive to the lurking endgame even as you pursue middlegame plans; your attack need not be pursued to mate but can be dissolved into a won ending that untrained players just wouldn't win. And you leave many opponents who thought they were as good as you or better wondering what happened to their wins or draws.

My latest blogarticle discusses practical endgames: http://chess-brabo.blogspot.be/2014/08/practical-endgames.html

In answer to the question about endgame study - any concentrated sort of study on chess (where you put in effort!) is good.

Endgame study is probably a good way to train pure calculation as the positions are often clean enough to really get a long way into, and to punish minor innacuracies quite hard. That all gets much less clear in most sorts of middlegames.

We're having a gently amused/pointlessly pedantic debate about the given position just because we're not sure if it really relates to endgames rather than something else I have to say that my impression of 'pure' B&B vs R endings is that the poor rook gets totally outgunned, so I'd have stayed well clear of anything like that on principle.

As an weak player I get confused. The main question is still: Is endgame study waste of time or not?

I think doing endgame study and execises will improve your visualization and calculating ability. And not necessarily only the technique of delivering mate. Those players which have reach 2000+ has learned this in other ways. But will endgame study be a shortcut or tardy route?

I still have hard to belief that Capablanca and Tarrasch were wrong.

Jupp53 wrote on 06/25/14 at 12:02:44:

But in this case, are you sure that the lack of endgame knowledge was really the reason you exchanged rooks in this position? In other words, did you aim for the 2B vs R+2P endgame, or did you not foresee that white could also force the swapping of queens? Was Rab8 one of your candidate moves?
I agree with brabo that to me this looks more like a middlegame problem. As I said, I don't know much about these endgames apart from the very vague fact, that I've always suffered with a rook against the bishop pair. I'd have no firm idea how to try and use those two pawns extra to my advantage. But to me, even this vague idea is enough for me do decide that if possible I'd try to keep the heavies on board for a long as possible. That's by the way not saying that I would have found Rb8...

What were your feelings during the game once you found out that this endgame would certainly arise (after the exchange of Q's)?

Exchanging one pair of rooks this is an endgame for me with middlegame elements. But maybe this is one of my weaknesses to judge it this way. After the exchange of the rooks white forced the queen exchange, which lead to an endgame r+2p vs 2B.

The road to an endgame is often a small one. That's what I wanted to say and the game position has to do with different choices.