Chess Exam And Training Guide
Rating chess skills and progress
Description
Most players who take chess seriously carry a number in their heads. It might be an official rating from a federation, it might be a guess, it might be the memory of a good tournament three years ago. What almost nobody carries is an honest picture of where that number comes from — which parts of their game hold it up, and which parts are quietly dragging it down. Igor Khmelnitsky, an International Master who grew up in the Soviet chess system before settling in the United States, noticed that gap and built a book around it. Chess Exam and Training Guide, published in 2004, is exactly what the title promises: an exam first, a study plan second.
The idea sounds almost too simple. Khmelnitsky assembled a hundred positions, each with questions, and turned the whole thing into a scored test. We sit down, we solve, we tally the points, and at the end we get a rating — not a vague sense of being "pretty good at endgames," but a number attached to twelve separate skills. Tactics, calculation, openings, defense, attack, the endgame, the middlegame plan. Each one gets measured on its own, against the performance of players at every level from beginner to master.
What comes out is often uncomfortable. A player rated 1600 discovers their tactics are worth 1900 and their endgame is worth 1200, and suddenly the flat number they'd been quoting for years turns into a jagged profile. That profile is the whole point. Khmelnitsky isn't interested in flattering anyone; he's interested in showing where the next hundred rating points are actually hiding.
The question we’re asking : How can a chess player replace a vague sense of their own strength with an honest, measurable picture — and use it to improve?What we’ll see : How a scored self-exam turns a single rating into a detailed map of strengths, weaknesses, and the work that follows.
Table of contents
01Chapter 1 — The problem with knowing how good we are
Ask a club player how strong they are and you'll usually get a rating and a shrug. The rating is real enough — it comes from wins and losses over the board — but it's a single number standing in for something wildly uneven. A rating averages everything: the games we won on a cheap tactic, the ones we lost in a drawn endgame, the openings we know cold and the ones we bluff through. It tells us the sum. It tells us almost nothing about the parts.
Khmelnitsky's starting observation is that this averaging hides exactly the information a player needs to improve. Two people rated 1700 can be completely different players. One survives on sharp tactics and collapses the moment the queens come off; the other grinds patient endgames but never sees a combination coming. They share a number and share nothing else. If both of them study "to get better," they'll waste time on the wrong things — the tactician drilling more tactics, the grinder polishing endgames they already own.
02Chapter 2 — A test with a number at the end
The mechanics are straightforward, which is part of why the book works. Khmelnitsky presents a hundred positions drawn from real games. Each one comes with a question — sometimes find the best move, sometimes assess who stands better, sometimes choose a plan. We write down our answer before turning the page, because the whole system depends on honesty, and it's easy to convince ourselves after the fact that we "basically had it."
The scoring is where the design gets clever. Answers aren't simply right or wrong. A position might award full points for the strongest move, partial credit for a reasonable one, and nothing for a blunder — with the point values calibrated so that the total maps onto a rating scale. Khmelnitsky spent years gathering data on how players at different levels actually solved these positions, which lets him say, with more authority than a guess, that a given score corresponds to roughly master strength, or roughly 1500, or somewhere in between.
03Chapter 3 — The map of a chess mind
Once the scoring is done, the profile reads like a topographic map of a player's mind. Khmelnitsky breaks the game into around a dozen components — tactics, calculation, standard positions, the opening, the middlegame, the endgame, attack, defense, counterattack, sacrifice, strategy, and the harder-to-name skill of practical decision-making under real conditions. Each gets its own line, its own rating, its own place on the terrain.
What the map reveals is that improvement is rarely uniform. A player might be a strong attacker and a poor defender, sharp in calculation but shaky in quiet strategic positions where there's nothing to calculate. These imbalances are usually invisible from the inside, because we steer our games toward the ground we like. An attacker unconsciously creates messy positions; a positional player trades into calm ones. The self-exam strips away that steering and asks the questions we'd never volunteer to answer.
04Chapter 4 — When the mirror is the coach
Step back from the sixty-four squares and Khmelnitsky's real subject comes into focus: the distance between the skill we imagine we have and the skill we can prove under test conditions. Chess is unusually honest here, because it keeps score and pairs us with opponents who don't care about our self-image. But the illusion the book fights — the confident, averaged, memory-flattered sense of our own ability — is not a chess problem. It's a human one, and chess simply makes it visible.
In most skills we never get the equivalent of a rating. We assume we're a decent writer, a competent manager, a careful driver, and nothing forces the assumption to meet evidence. When feedback does arrive, it usually comes as a single verdict — a promotion, a rejection, a compliment — that averages everything the way a chess rating does, hiding the components underneath. Khmelnitsky's method is a small argument against that averaging: break the skill into parts, measure each one honestly, and let the weakest part set the agenda.
05Conclusion
The player who opens Chess Exam and Training Guide expecting a puzzle book closes it holding a profile of themselves. That was Khmelnitsky's design from the first position: not to entertain but to measure, and then to convert the measurement into a study plan aimed at the one line that scored lowest. The single flat rating we walked in quoting turns out to have been an average concealing a dozen separate truths, and the book's whole value is in pulling those truths apart.
