Why use this calculator
Official rating lists are published at the end of each rating period, so after a weekend tournament there is a gap of weeks between knowing your results and knowing your rating. That gap matters if you are chasing a threshold, deciding whether to enter a rating-restricted section, or working out whether a last round draw is worth taking.
This calculator closes it. It applies the expected score formula game by game, respects the 400 point rating difference cap that a naive Elo implementation leaves out, and lets you set the K-factor that actually applies to you rather than assuming one. The per game breakdown is the useful part: it shows which games earned points and which lost them, and it makes the central insight of Elo visible, which is that only the difference between what you scored and what you were expected to score moves your rating. Beating people you should beat is worth almost nothing, and that is by design.
How to use it
- Enter your current published rating from your FIDE profile, making sure you use the list that matches the event, because standard, rapid and blitz are rated separately and a player can hold three quite different numbers.
- Choose the K-factor that applies to you: 40 while you have fewer than thirty rated games or if you are under eighteen and rated below 2300, 20 if you are established and have never crossed 2400, and 10 once you have reached 2400 at any point.
- Enter each opponent's rating exactly as it appeared on the pairing sheet or the rating list in force for the tournament, not their current rating from the website, since the calculation uses the list that was live when the event started.
- Select your result in each game, scoring one for a win and a half for a draw.
- Leave the second and third opponent ratings at zero if you are rating fewer than three games.
What each result means
- Estimated new rating
- Your rating after these games, rounded the way a published list would show it. It is an estimate, not an official figure.
- Rating change
- The total movement across the games entered. It is the only number that matters, because everything else is intermediate working.
- Games counted
- How many games were included. Opponents left at a rating of zero are skipped rather than treated as very weak players.
- Points scored
- Your tournament score in the usual chess notation, where a win is one point and a draw is half a point.
- Points expected
- The score the Elo model predicted against this particular field. Compare it with your actual score to see how the event really went.
- Points above expectation
- Your score minus the expected score. Multiplied by your K-factor this is exactly the rating change, which is the whole model.
- Game by game
- Each game's opponent, result, expected score and rating change, so you can see which results carried the tournament.
- Opponent rating
- Shown when no opponent rating has been entered yet, because a rating change cannot be estimated without at least one game.
The formula
Expected score in one game = 1 / (1 + 10 raised to the power of d / 400)
Rating change for that game = K x (actual score - expected score)
The total change across a tournament is the sum of the individual game changes, which is identical to K multiplied by the difference between your total score and your total expected score. The new rating is the current rating plus the total change, subject to the published minimum of 1400.
How this number is used in practice
The practical use of this is deciding what a result is worth before you play it. Late in a Swiss tournament you often have a choice between a safe draw with a higher rated opponent and a sharp game with real losing chances. The expected score tells you the price of each outcome directly: against someone 200 points above you the expected score is 0.24, so a draw is worth about a quarter of a point above expectation and at K equals 20 that is roughly five rating points, while a loss costs about five.
Against someone 200 points below you the arithmetic reverses completely, and the draw that felt respectable is the same size of setback as the win was a gain.
The 400 point cap is the rule most online Elo calculators omit, and it changes results at the edges of a field. Without it, a heavily favoured player who loses to a beginner would drop by almost a full K-factor, and a beginner who beats a grandmaster would gain the same.
FIDE caps the difference used in the formula at 400 points in either direction, which limits both the reward for a giant killing and the damage from an upset. In an open tournament where a titled player meets a club player in round one, this cap is doing real work on both sides of the board.
Treat the output as an estimate, not a rating. FIDE calculates official changes at the close of a rating period using the list in force at the start of the event, applies its own rounding conventions, and handles special cases this tool does not model, including the first rating a player receives, games against unrated opponents, and the separate treatment of rapid and blitz events.
Title norms are a completely different calculation based on tournament category, opposition average and the number of federations represented. If a threshold genuinely matters to you, check the FIDE rating regulations in the official handbook or ask your national federation's rating officer rather than relying on an estimate.
Worked examples
An 1800 player at K equals 20 beats a 1750 and draws a 1900
Points expected 1.07 against 1.5 scored, a change of +8.7 and a new rating of 1809
A new player at K equals 40 loses to an opponent 600 points higher
The 400 point cap holds the expected score at 0.09, so the loss costs only 3.6 points
A 2450 player at K equals 10 draws three opponents rated 2300, 2350 and 2400
Expected 1.85 points against 1.5 scored, a change of -3.5 to 2447
An 1800 player at K equals 20 beats a 1400 and a 1500
Two expected wins earn just +1.1 points in total, ending at 1801
When people use it
- Estimating your rating change on the drive home from a weekend Swiss
- Deciding whether a draw or a win is worth more against a particular opponent
- Checking whether a result keeps you above a rating-restricted section threshold
- Explaining to a club newcomer why beating weaker players barely moves a rating
- Comparing how the same tournament result would land at K equals 40, 20 and 10
Tips
- Use the rating list that was in force when the event started, not today's website figure, because that is the list the official calculation uses.
- Standard, rapid and blitz are rated separately. Putting a blitz rating into a standard event calculation will give a change that never appears.
- The 400 point cap protects you far more than it costs you when you are the underdog, so playing up in a strong section is rarely expensive.
- If you are within a few points of a threshold, remember that a draw against a much weaker player is the same size of setback as a win against a peer is a gain.
- Check your K-factor before an event. Moving from 40 to 20 halves both your upside and your downside, which changes how much a single bad round matters.
- Only the gap between your score and your expected score moves a rating, so a good tournament against a weak field can still lose points.
What it assumes, and where it stops being accurate
- Uses the standard Elo expected score formula with the 400 point rating difference cap applied, which matches the published FIDE rating regulations for ordinary rated games.
- Assumes a single K-factor applies to every game in the set. FIDE can change a player's K-factor once a threshold is crossed, which this does not detect.
- Assumes every opponent is rated on the same list as you. Games against unrated opponents follow a separate procedure that is not modelled here.
- Does not apply the rules for a player's very first rating, which is calculated from a minimum number of games rather than by adjusting an existing number.
- Not accurate for tournament performance ratings, title norms, national federation rating systems, or online platform ratings such as Glicko, all of which use different mathematics.
- Rounding may differ from the official result by a fraction of a point, because FIDE applies its own conventions at the end of a rating period.
Terms used on this page
- Elo rating
- A number describing relative playing strength, where the difference between two ratings predicts the expected result.
- Expected score
- The average score the model predicts for a game, between zero and one, derived from the rating difference alone.
- K-factor
- The multiplier controlling how much a single result moves a rating. Higher K means faster movement and more volatility.
- Rating period
- The interval between published rating lists, during which results are collected and processed together.
- Performance rating
- A separate statistic estimating the rating level a player's tournament score would correspond to against that field.
Questions people ask
- Which K-factor applies to me?
- FIDE uses 40 for players with fewer than thirty rated games and for players under eighteen rated below 2300, 20 for established players who have never crossed 2400, and 10 once a player has reached 2400 at any point. Your national federation's rating officer or your FIDE profile can confirm which currently applies to you.
- What is the 400 point rule?
- If the rating difference between two players is more than 400 points, the difference is treated as exactly 400 for the purposes of the calculation. It caps both the reward for beating a far stronger player and the penalty for losing to a far weaker one, and leaving it out is the most common error in unofficial Elo calculators.
- Why did my rating barely move after winning every game?
- Because Elo only rewards the difference between your score and your expected score. If you were expected to score four and a half points out of five and you scored five, the half point surplus multiplied by your K-factor is all you gain. Winning a weak tournament is worth very little by design.
- Does this calculate my performance rating?
- No. Performance rating is a separate statistic derived from your opponents' average rating and the percentage you scored against them, and it describes how you played rather than how your rating changes. The two frequently disagree, especially over a short event with an unusual field.
- Can I enter an unrated opponent?
- Not meaningfully. FIDE handles games against unrated players through a separate procedure that depends on the opponent eventually receiving a rating from that event. Leave the opponent rating at zero to skip the game, or enter a realistic estimate if you only want a rough sense of the outcome.
- Why does my estimate differ from the official change?
- Official changes are computed at the end of a rating period against the list in force when the event began, with FIDE's own rounding and with any special cases applied. Differences of a fraction of a point are normal, and a larger gap usually means a different opponent rating list was used.
Use this with
FIDE Chess Rating Calculator answers one part of the question. These pick up where it stops: Probability Calculator, Percentage Calculator, Z-Score Calculator and Ratio Calculator.
Where these figures come from
This calculation is jurisdiction-neutral arithmetic. No country-specific rules are applied.
- FIDE Handbook. The rating regulations, including K-factor rules and the 400 point difference cap
- FIDE Ratings. Published rating lists and individual player profiles
- International Chess Federation. Rating periods, tournament regulations and title requirements
FIDE Chess Rating Calculator produces an estimate for general information. It is not financial, tax, medical, legal or engineering advice. read the full disclaimer. Spotted a wrong number? Report it and it gets fixed. The form picks up this calculator and the inputs you used.