Chess, explained.
Practical writing about analysis, improvement and the positions that keep coming back. No engine worship, no ten-step secrets.
How to analyse your own chess games (without wasting the review)
Most players run the engine, see a red blunder, and move on. A useful review starts before the engine and ends with one change you can actually make.
Read the guideStockfish depth vs nodes: what is the difference?
Depth is how many plies the engine looks ahead. Nodes are how many positions it visits. They are related, but they are not interchangeable.
What is a giga node? 100 GN explained with practical examples
A giga node is a billion positions searched. It sounds enormous, and it is, but chess analysis spends nodes quickly. Here is the practical arithmetic.
How to read a Stockfish evaluation (cp, mate and +0.30)
An evaluation is a compressed opinion, not a verdict. Learn what +0.30, -1.20 and M5 actually mean before you change your opening repertoire.
Chess accuracy scores explained (and why 100% is not the goal)
Accuracy is a summary of how much your moves lost compared with the engine's best move. It is useful for finding patterns, not for ranking yourself.
Blunder, mistake or inaccuracy? How chess engines classify moves
The labels are useful shorthand, but they are thresholds, not verdicts. A move can be a blunder in one position and the only practical try in another.
Why a full-game review beats checking only the blunders
The worst move is rarely the whole story. The move before it, and the plan behind it, usually explain more.
How to review a lost chess game without tilting
Losing feels like evidence. It is not. A loss is a set of decisions you can inspect, and the review should be about the decisions, not the score.
Opening preparation with an engine: plans before moves
An opening repertoire is not a list of moves. It is a set of plans you can recognise when the opponent leaves the script.
How to use engine lines correctly (without following them blindly)
The engine is a calculator with no clock, no nerves and no opponent. Its line is a proposal, not a prophecy.
Endgame training with Stockfish: depth, tablebases and practical drills
Endgames are where deep analysis pays off most. The goal is not to memorise every tablebase position; it is to recognise the few structures that decide games.
How to use infinite analysis without wasting credits
Infinite mode is not a magic button. It is a long search you have to steer. The skill is knowing when to stop.
Puzzle rating vs game rating: why they diverge
Puzzles measure pattern recognition in a clean position. Games measure everything else: opening, clock, nerves, and the ability to create the pattern.
How to solve hard Lichess puzzles (without guessing)
Hard puzzles are not solved by seeing further. They are solved by asking better questions before calculating.
How to create puzzles from your own chess games
Generic puzzles teach generic patterns. Your own blunders teach the patterns you keep missing under pressure.
Chess time controls explained: bullet, blitz, rapid and classical
A time control is not just a number. It changes the opening choices, the tactical density and the kind of mistakes you make.
How to choose a time control for chess improvement
The best time control is the one that exposes the skill you are working on. If you only play one format, you only train one skill.
What bot depth should you play against?
A deep bot is not a strong human. It is a predictable calculator. Choose the depth for the skill you want to train.
Why engine assistance ruins improvement (and why it is banned)
The point of a game is not the result. It is the feedback. An engine removes the feedback and leaves only the result.
PGN explained: how chess games are stored and exported
PGN is the lingua franca of chess games. It is simple, portable and worth understanding if you keep a game library.
How to build a permanent chess study library
Most players lose their best analysis in chat windows and temporary tabs. A study library turns scattered work into a reusable asset.
Centipawn loss and win probability: why chess engines use both
Losing 100 centipawns from +0.20 is not the same as losing 100 from +8.00. Win probability captures the difference.
How online chess clocks work (latency, increment and flag falls)
An online clock is a server-side timer, not a local stopwatch. Understanding the difference explains most clock disputes.
How to prepare for a chess tournament with online tools
Tournament preparation is not about learning a new opening. It is about making your existing chess reliable under pressure.
Review a game, not just a position.
Choose a depth, see the maximum credit cost, and get accuracy plus move-by-move markers for both sides.