Game Review 18 Jul 2026

Slow Down, Past Me

Puzzle Rush Survival this morning finished on 28, personal best is still 35, so a way off, but the three misses on the way down are worth more than the number itself. Same evening, two rated games, one draw I should have won, one loss that got out of hand fast. Different formats, same problem showing up five separate times, picking the first move that looks like it works and not checking whether there's a better one sitting right next to it.

Puzzle Rush misses

Puzzle 1
Puzzle 1 position
I was thinking that Black had no choice but to trade rooks and accept losing, my selected line was Rb1 Rxb1 axb1=Q. I rushed this more than I should, and again, in my usual style, picked the first move that looked okay and went with it. My candidate generation OTB is just not strong enough.
The correct move was to sacrifice the rook instead, letting the pawn promote with check. Black can either interpose with the rook and lose it instantly, or move the king back to the seventh rank, which just walks into another check, this time a double attack that picks up the rook next. The actual line runs Rf8+ Kxf8 a8=Q+ Kg7 Qa7+ Kg6 Qxb6.
Puzzle 2
Puzzle 2 position
My queen is under attack, but I can obviously trade it with check, this puts White's f-pawn into a pin, allowing me to capture their e-pawn. Again this move seems to work, so I went with Qxd1+ Kxd1 Nxe4. I'm now looking at the correct solution on screen to type up my notes, and instantly I see it, the white knight has a fork. When I did this OTB I never saw I was in a fork, this is again a board vision issue. I need to slow down and check every piece in a new position, what are they all seeing? Disappointed that I didn't spot my rook being under attack OTB.
The puzzle is actually very simple. Qxd1+ is forced, since anything else eventually leads to mate or losing the queen. Kxd1 is obvious too, the rook is under attack with two attackers and only one defender. The actual solution continues Bxf7 Bxf7 Kxf7.
Puzzle 3
Puzzle 3 position
I immediately spot there's a discovered attack if I move my light squared bishop, so if I can play Bg8 it's over. I try to play it and it won't let me, because I'm in check, so I actually failed this puzzle twice, but you can't play illegal moves online so it doesn't count as a fail. I continue trying to find a way to get my bishop to g8, and see that Kf1 gets my king out of the way with no more checks, game over. Again I picked a move I thought worked and didn't fully check Black's responses, Black can respond with Bg4, completely destroying my attack and putting me in a losing position. What I didn't account for is that the bishop doesn't need to block the king's escape at all, I saw it was protected on g8 by the pawn, but didn't take into account that the square being protected stops the king going to g8 anyway. Such a stupid miss.
Seeing the puzzle again in review, I accidentally clicked the solution button before looking properly, but the actual solution is fairly obvious once you see it, Be4#. I focused too much on needing to get the bishop to g8, and not on the discovered attack itself, the bishop interposes the check from Black while simultaneously revealing checkmate.
The same thing happens in every one of today's misses, not taking enough time to look for a better move once I've found a good one, not running the line out to the end of my opponent's response, and not fully assessing the position. Slow down.

Game 1: v Ouelletjude, Colle System, Draw

I played White. Made an inaccuracy in the opening that levelled the game, slowly built up an advantage from there, then completely collapsed in the endgame. I had many chances to win it. Where's that Silman Endgame Course book at, I think I should spend some more time on endgames after this.

10. O-O Inaccuracy, lost 1.32 pawns
The engine says this is a mistake and I lose my momentum, it prefers to take the knight. I did look at the line it recommends, but in the game I didn't want to lose my bishop.
Bxe4 was the top engine choice, but I avoided it purely because giving up the bishop felt uncomfortable in the moment, not because O-O was actually the safer option. Discomfort with a trade isn't the same as the trade being wrong.
35. Rf5?? Critical moment, lost 10.88 pawns, hung material
Here I saw the option to trade off the rooks, I know this drops a pawn but I think I can still push my f-pawn up the board. Again I did see the best move, Rf7, attacking the two pawns, but I was confident I could convert the pawn endgame.
Seeing the better move and talking yourself out of it because you fancy your chances a different way is its own kind of mistake, separate from just missing it outright. Rf7 kept real winning chances alive, trading down removed the safety net and put everything on pawn endgame technique instead.
Moves 38 and 41 Two more critical moments, lost 6.74 and 9.73 pawns
There are many mistakes in this endgame, I only pick out the biggest ones to look at, the eval bar swings wildly on each move, as we both clearly don't know how to successfully convert pawn endgames. This continues the same thread of not being able to calculate the pawn endgame.
Pawn endgames are pure calculation, no tactics to fall back on, no pieces to bail you out, just counting squares correctly under time pressure. Both sides missing the accurate lines here isn't really about board vision, it's a specific gap in endgame technique that the earlier mistakes don't share.

The position eventually simplified all the way down to bare kings after both sides queened and traded everything off, 73 moves and a draw. A frustrating result given how winning the position was earlier, and a clear signal about where the next block of study time should go.

Game 2: v baldeepdeol, Rapport-Jobava System, Loss

I played Black. A poor game overall, focused on protecting a pawn rather than getting my opponent's queen away from my king. Only two real mistakes in the whole game, but their severity meant it was all downhill from there.

10...Nxd4?? Critical moment, lost 3.46 pawns, hung material
This is quite a significant first mistake, I didn't give enough credit to my opponent's attack and how severely it can damage my king's protection. I saw a line where I check the king and attack the knight at the end, I stopped too early and didn't realise the knight can save itself by blocking the check.
Same failure as the puzzles, stopping the calculation one move before the position was actually resolved. The line looked winning right up until the point I stopped reading it.
16...Be5? Critical moment, lost 6.61 pawns, positional
My opponent makes a mistake bringing their queen up to h6, I don't feel like this move is really scary, the engine agrees, however I didn't spot Bf8 to chase the queen. My priority here should have been chasing the queen away, instead I focused on protecting my pawn since it's under a double attack. The engine line actually ends in me taking the knight and being completely winning. Same issue as always, not generating enough candidate moves, and playing the first move that looks good.
The opponent had already handed back the advantage with their own mistake, and there was a move on the board that punished it outright. Getting fixated on defending the pawn meant the actual opportunity never even got considered as an option.

From there the attack kept coming and never let up, ending in checkmate on move 30. A tougher one to review than the draw, since there's less to point to beyond the two moments, but those two moments carry the whole game.

Five mistakes, one cause

Three puzzles, two games, five separate moments where the story is basically identical, find a move, feel confident about it, and stop looking. The pawn endgame collapse in game one is its own separate issue and points at needing more endgame study specifically. Everything else, the puzzles and the tactical miss in game two, comes back to the same root cause. Generate more candidates. Run the line to the end. Slow down.

Follow the journey

Join Inter Mezzo Sporting Club

Join the club, play in the league, and follow the Road to NM journey as it unfolds.

Join