I Need to Get Better at Endgames

Everyone could be better at endgames - this is one of the common themes of chess improvement. In my case, however, I am starting to see that I'm drawing *completely won* endgames. This is one of downsides, I think, of studying just theoretical endgames, or endgame puzzles that only have one correct move. You don't have the experience to overwhelm your opponent's desperado counter-play and convert the position. Especially when I'm more used to defending a losing endgame! This is all rather new to me, to be honest.

Here are some recent examples. In all cases, it's "simply a matter of technique" - which I am sorely lacking.

Opposite Colored Bishops Don't Always Draw

I reached an endgame up 2 pawns and somehow figured out how to draw instead of win:

In the game, I for some reason moved my king back instead of pushing pawns and the king up the board.

Knight vs. Knight

Again, up to pawns - but with knight vs. knight - I snatched a draw from the jaws of victory:

Instead of controlling key squares and forcing the knight away, I tried to break the pawns through too early.

Simplify to Win

The Engine has the cleanest win, but Humans actually have to calculate. Fewer pieces on the board makes it easier:

For some reason, I left my rook on the board instead of trading down to an overwhelming pawn majority. This allowed White too much counterplay under time pressure.

Hold the Draw

The Engine has multiple moves drawing, but one draw is easier to hold than others.

Conclusions

A big part of my training regimen is Lichess endgame puzzles:

https://lichess.org/training/pawnEndgame
https://lichess.org/training/rookEndgame
https://lichess.org/training/bishopEndgame

(I've been neglecting knight endgames - d'oh!)

I've been doing these three regularly, along with puzzle streak (to over-train simple tactics) and opening-themed puzzles. However, these endgame puzzles always have just one correct move! What are you to do when you have multiple moves that work, but one that makes winning the easiest? This is the issue I think I been having, and I need to figure out what to do about it.

It's about finding the line that simplifies, is easiest to calculate, and causes the most complications or increases pressure for my opponent. This is tournament chess against a human, over-the-board, so finding the objectively best move is not always the absolute best one (i.e. capture the queen or mate-in-7?)

I suspect as my play has improved, I'll be in this situation (completely winning endgame) more and more.

Anyway, more work to do.