Take two players with known ratings. The system does not ask who is better - it produces a number: the share of points the higher-rated player is expected to score in a long match.
A game is then compared with that prediction. If the result matches it, ratings barely move. If it contradicts it, the ratings move sharply. That is the whole mechanism, and everything else is detail.
Why beating a weaker player earns almost nothing
Because the system already predicted it. You confirmed what was expected, so there is nothing to correct. And a loss to that same player costs a great deal - the prediction was badly wrong.
The expectation depends only on the rating difference, not on the absolute numbers: 1200 against 1000 and 2400 against 2200 give the same prediction.
| Rating difference | Expected score of the favourite |
|---|---|
| 0 | 50% - a coin toss |
| 100 | about 64% |
| 200 | about 76% |
| 400 | about 91% |
| 800 | about 99% |
Two hundred points is the interval worth remembering: it means roughly three points out of four for the stronger player - noticeable but far from certain. Which is why upsets at that gap are routine rather than remarkable.
One formula, and it is simpler than its reputation:
change = K × (actual score − expected score)
The actual score is 1 for a win, 0.5 for a draw and 0 for a loss. The expected score comes from the table above. K is a coefficient described in the next section.
An example: you are rated 1600 and beat somebody rated 1800. The expectation for you was about 0.24, you scored 1, and the difference is 0.76. With K = 20 that is roughly +15 points - and the same 15 come off your opponent, because the system is zero-sum.
Which leads to a useful conclusion: a draw with a much stronger player gains you rating. The expectation was a loss, and half a point beats it.
K decides how strongly one game moves the number. It is not the same for everyone, and deliberately so: a new player's rating is a guess that needs correcting quickly, while an established player's is based on hundreds of games and should be stable.
So a beginner swinging fifty points in one event and a master moving three are not inconsistencies - they are the same formula with different K.
You cannot compute a rating from nothing: the formula needs both players to have one. So the first is obtained differently.
Practically: to get onto a national or international list you need events that report to it. A club evening does not produce an official rating, however well it is run - what club events do produce.
Classical, rapid and blitz are rated separately, and the numbers routinely differ by a couple of hundred points for the same person.
Three things get confused constantly, and separating them clears up most questions.
For an amateur club the practical answer is usually an internal rating: it gives players visible progress without anybody having to register events with a federation - and visible progress is what keeps people coming back - about retention.
By the formula Rnew = Rold + K × (S − E), where S is the points actually scored, E is the expected result given the rating difference and K is the development coefficient. The expected result is calculated as E = 1 / (1 + 10^((Ropponent − Rplayer) / 400)).
Because the rating responds not to the win but to the deviation from expectation. If the system predicted 90% of the points for you and you won, the result matched the forecast - so the gain is minimal. A draw with a stronger opponent, on the other hand, produces a noticeable rise.
The multiplier that sets how fast the rating changes. Under the FIDE regulations: 40 for new players up to 30 games and for juniors under 18 rated below 2300, 20 for players below 2400, and 10 for those who have reached 2400.
At least five games against rated opponents; they may be played in different tournaments. A first published rating cannot be below 1400. A Russian federation rating is easier to get: with no rating at all, a starting value of 1000 is used.
They are three independent lists, and a tournament goes into the one matching its time control. One player's figures usually differ: a blitz rating is often higher or lower than the classical one depending on playing style.
The FIDE lists are published monthly, on the first of the month. A tournament has to be submitted for rating before the period closes, or the results go into the next list.
The rating falls with poor results. If it drops below 1400, the player is shown in the list as unrated and gets a rating again after new games. Rating categories, unlike the rating itself, are not taken away.
Everything people usually keep in a spreadsheet is already in the app: pairings, points, tie-breaks, round order. Free, with no subscriptions and no cap on the number of players.
many players, one evening: four or five rounds are enough
the fairest format for six to ten players
not a tournament, just an evening of chess - who is free and who is playing is visible in the app
The bot is constantly improved based on real tournaments. If you need a feature that isn't there yet, message the developers right inside the bot with the /support command. Many features appeared exactly because organizers asked for them, and useful ideas get added quickly.
Reference - FIDE titles, tie-breaks, glossary, online ratings.
At the board since childhood: hazy about his rating, precise about every rook he has blundered. Counts an evening as a good one if the opponent stayed to talk afterwards instead of rushing off to check with an engine. Message →
