How to Explain Chess Ratings to Parents Honestly

A chess rating measures one thing: how a player's results compare with the results the system expected against rated opponents. It is a relative estimate of strength inside a specific pool of players, not a percentage score, not a grade, and not a measure of effort or talent. When a parent asks what the number means, that is the sentence to give them first.

The oldest and most familiar version is the Elo system, named after the physicist Arpad Elo and used by FIDE and many national federations [1]. Elo works like a prediction market with exactly two participants. Before a game, the rating gap implies an expected score: a player rated 200 points higher is expected to win clearly more often than not. After the game, the rating moves by a slice of the difference between expectation and reality. Beat someone much stronger and you gain a lot; beat someone much weaker and you gain almost nothing, because the system already expected it.

Most online platforms use a refinement of this idea. Lichess uses Glicko-2, a descendant of Mark Glickman's Glicko system, which adds a second number called rating deviation [2]. Rating deviation is the system's honesty about its own uncertainty. A brand-new account has a high deviation, so the system says, in effect, I have no idea yet, and moves the rating in big jumps. A player with hundreds of recent games has a low deviation, so each result nudges the number gently. Parents do not need the mathematics; they need the translation: the rating is a guess, and the system itself tracks how confident the guess is.

Because each rating lives in a different pool of players, under a different formula, with a different starting point. A Lichess rating, a Chess.com rating, and a FIDE rating for the same child routinely differ by hundreds of points, and none of them is lying. They are answering different questions: how strong is this child relative to Lichess users, relative to Chess.com users, and relative to FIDE-rated tournament players.

The comparison parents instinctively reach for, my child is 1600 online but only 1100 FIDE, so which is real, has a clean answer: both are real, and neither converts to the other. The pools barely overlap. A casual blitz pool on a free website and a hall full of tournament regulars are different populations, and a rating only ever ranks you inside your own population. The practical advice for families is to pick one number per context and track it against itself over time: the Lichess rapid rating against last quarter's Lichess rapid rating, the FIDE rating against last season's FIDE rating, never one platform against another. Our comparison of Lichess and Chess.com for academy use covers how the platforms differ for coaching workflows beyond the rating question.

Normal progress for a child is a noisy line that trends upward in steps, not a smooth slope. The early months look dramatic in both directions because the rating system is still uncertain about a new player, and because the player is genuinely changing week to week. After a few hundred games the line calms down, and from there progress arrives in plateaus punctuated by jumps, usually after a new idea finally clicks.

The early volatility deserves its own conversation with families, because it is the single biggest source of misread signals. A new Lichess account starts at 1500 with a high rating deviation, which means the first twenty games can move the number by a hundred points or more in either direction [2]. Parents who watch daily see a child gain 120 points in a fortnight and conclude the child is a prodigy, or lose 90 and conclude the coaching has failed. Neither is true. The system is calibrating, and the sensible coaching position is to treat any rating younger than a few dozen games as provisional and say so out loud before the family ever sees the chart.

On the shape of progress by age and hours, be careful to present what you have: coaching observation, not statistics. In our experience across academy batches, a primary-school child training twice a week with some home practice typically shows visible movement across a season, while a teenager starting fresh often climbs faster at first because they read and study independently, then meets the same plateaus everyone meets. A child who only plays in class and never plays at home progresses more slowly than the family expects, and that is a practice-volume conversation, not a talent verdict. We state these as patterns we observe, we never dress them up as percentages, and we tell parents the honest baseline: improvement correlates with games played and games reviewed far more than with any trait visible in the first month. For students stuck on a long flat stretch, our guide to the chess rating plateau covers what flat periods mean and what actually breaks them.

What should worry a coach is not a flat line but a flat line with falling activity. A plateau at thirty games a month is a player consolidating. A plateau at four games a month is a player quietly disengaging, and the rating chart will not show the difference unless game volume is plotted next to it. That is why our progress conversations always pair the rating line with the games-played count for the same period: the two lines together tell a story that neither tells alone.

You talk about a rating drop by zooming out before you explain anything. Open the six-month chart, not the one-month chart, and let the parent see that this month's dip is a wiggle on a line that is still higher than where it started. Most rating-drop conversations end right there, because the anxiety was about the last three weeks and the chart is about the year.

When the parent needs more than the chart, move from the number to the games. A rating is a lagging summary; the games are the evidence. Pull up two or three recent losses and show what actually happened: time trouble in winning positions, a new opening that is not comfortable yet, stronger pairings this month. Each of those is a concrete, fixable story, and concrete stories calm families in a way that rating theory never does. Our guide to reviewing a student game in fifteen minutes shows the fast version of this workflow.

Two things make this conversation easy to have honestly. First, never have promised a number in the first place: commit to process, games played, openings learned, endgames mastered, and let the rating follow. Second, have the history ready before the question comes. A coach scrambling to reconstruct three months of results looks evasive even when nothing is wrong. The structure in our chess progress report template keeps the rating section honest by default, and the broader reports guide covers how the rating chart fits into a full progress report.

An honest rating chart shows the full available history, labels which platform and time control the line comes from, and plots game volume alongside the rating. The test is simple: if the family opened the child's Lichess or Chess.com profile themselves, would they see the same story the chart tells? If yes, the chart is honest. If the chart starts at the season's lowest point and stops at its peak, it is marketing.

• Full history, or at minimum the last six months, never a window chosen after looking at the data
• Platform and time control on the label, because a Lichess rapid line is not a FIDE line
• Games played per month next to the rating, so a quiet month explains a flat line
• Provisional periods marked or mentioned, since early numbers carry high uncertainty [2]
• The same chart every report, so families learn to read it instead of relearning a new format

The cheapest way to keep charts honest is to remove the human from the charting. Numbers retyped into a slide deck drift toward flattery, not because coaches lie but because everyone prefers the better-looking window. A chart drawn directly from synced platform data has no window to choose: Lichess publishes each account's full rating history through its public API [4], Chess.com profiles show per-format histories publicly [5], and a FIDE rating is checkable by anyone on the federation's own list [3]. When the source is the platform, the chart and the profile cannot disagree.

A child should pursue a FIDE rating when they are playing over-the-board tournaments regularly and holding their own against rated opponents, not as a milestone to rush. A FIDE rating only comes from games against rated players in FIDE-rated events [3], so the practical question for a family is whether the child is ready to play several such events a year and absorb the early losses that getting on the list usually involves.

Set the expectation that the first published FIDE number will likely sit far below the child's online ratings, for the pool reasons covered above, and that this is normal rather than a setback. The early FIDE number is a starting line, not a verdict. For families in India, the path runs through state association events and national federation registration before FIDE rating becomes possible; our guide to getting a FIDE rating in India walks through that sequence step by step. For most academy students, the right order is: stable online rating first, local unrated and rated events second, FIDE list third.