Week 16: chess tactics for beginners, taught by three knight moves in my games

Week 16 (April 22–28, 2024) pulled me back toward 1000: 14 games, 7 wins, 7 losses, rating 976 → 997 (+21), with an average accuracy of 83.6%, my best week so far, and only 17 blunders. Most of the climb came on Sunday, when I played nine games and won six. The positions the engine flagged were all about one piece, the knight, which makes this week a neat little lesson in chess tactics for beginners: the fork, from both sides of the board.

The week in numbers

The low point came after game #402 on Saturday (961). Then Sunday: games #403 to #411, six wins and three losses, up to 1005 after game #410, and back to 997 when I abandoned the last game of the night. That's the strange part of this week: four of my seven losses were abandoned games, and one of them, game #398, was a position where I was better (-1.45) when I left it after 23 half-moves. Only one loss (#406) came after I had been at least 80% winning. When I finished games, I mostly played them well.

Three knight moves

A fork is a single move that attacks two (or more) things at once; the knight is the classic forking piece because it jumps and can't be blocked. Evaluations are from White's side.

Game #400: the knight check I played one move late

With White against llSerGHioll (984), in an Old Benoni, I was clearly better (+1.98). Black's king was still on e8. The engine's move was knight from c4 to d6, with check: from d6 the knight attacks the king and the bishop on c8, and moving it also opens my queen's line on the c-file to the knight on c6. I played my bishop to c7 first, and Black answered by pushing a pawn to b5, attacking my knight. The evaluation was back to equal (-0.07).

r . b . k . . r
q p . p b p p .
p . n . p n . p
. . . . . . . .
. . N P . B . .
P . . . P . . .
. P Q N . P P P
R . . . K B . R

I played the knight check on the next move anyway, but by then it won nothing. Later my queen walked into a knight capture, and I abandoned the game.

Game #401: a knight move into the center that achieved nothing

With Black against Rothria (1128), in a Cozio Ruy Lopez, White's pawn on d5 was attacking my knight on c6 and I had to move it. I went forward to d4, attacking White's knight on f3. The engine preferred the modest retreat to a7. After my move White simply traded on d4, won my pawn, and was clearly better; the engine's own line after my knight move is even worse for me (+3.15).

r . b q k b . r
. p p p . p p p
p . n . . . . .
. . . P p . . .
. . . . P . . .
. . . . . N . .
P P P . . P P P
R N B Q K . . R

Not every knight move toward the center is good: this one just got traded and cost a pawn. I resigned on move 11.

Game #402: the fork I walked into

With Black against hermlax (1099) I was already worse (+2.10) on move 26, but still fighting. I moved my rook from f8 to c8 to attack the pawn on c2. White's knight on f5 had a check on e7, and from e7 a knight attacks c8. The engine's line, and the game: knight to e7, check, attacking my king and my rook. +7.43 for White.

. . . . . r k .
R . . n . p p p
. . . . . n . .
. . . . . N . .
. p . B p P . .
. . . . . . P B
. . P b . K . P
. . . . . . . .

I resigned right after the fork.

The lesson of the week

Chess tactics for beginners start with one question about knights: where can a knight give check next move, and what else would it hit from there? I asked it for neither side this week. In game #400 I didn't look for my own knight check, which was the best move on the board. In game #402 I didn't look for White's, and put my rook on the one square the checking knight would attack. Game #401 is the footnote: a knight jump only works if it attacks something that can't just trade it off.

The other lesson is about finishing: 83.6% accuracy is wasted if four losses come from abandoned games.

All games this week

Previous: week 15, early queen attacks. Next: week 17, back above 1000. Every week is on the journey page, and Chess Coach can find the forks in your own games, including the ones you never noticed.