In an Elo-based pool, a 100-point rating gap gives the higher-rated player about a 64% expected score and the lower-rated player about 36%. That does not mean a 64% win probability or guarantee the result of one game. Expected score includes draws, which count as half a point, and becomes meaningful across repeated games in the same rating pool.
One game: expected points are 0.64 for the higher-rated player and 0.36 for the lower-rated player.
Ten games: expected totals are approximately 6.4 and 3.6 points.
Important: this is a long-run scoring expectation, not a prediction that the favourite must win the next game.
Judge each statement as correct or incorrect, then use the explanation to separate score expectation from one-game prediction.
1. Core Meaning
A 100-point Elo advantage gives the higher-rated player about a 64% expected score.
2. Win Probability Trap
A 64% expected score means the higher-rated player wins exactly 64% of games.
3. One-Game Certainty
A player rated 100 points higher should be expected to win every individual game.
4. Ten-Game Translation
Across 10 games, the higher-rated player's expected total is about 6.4 points.
5. Draw Value
A draw contributes 0.5 to each player's expected-score result.
6. Upset Possibility
The lower-rated player can win without proving the rating system is broken.
7. Cross-Pool Comparison
Any two ratings separated by 100 points are directly comparable, even when they come from different pools.
8. Rating Changes
The number of points gained or lost also depends on the applicable K-factor.
Do not read these as required results. Expected score is the average target implied by the rating model; a real match can finish above or below it.
That is why 64% expected score is not 64% win probability: many different combinations of wins, draws, and losses can produce roughly the same total score.
These examples use 0.64 for the higher-rated player, 0.36 for the lower-rated player, and the same illustrative K-factor of 20 for both players.
Illustration only: actual systems may round results and use different K-factors or update procedures. If the players have different K-values, their changes need not be symmetrical.
Expected score = 1 / (1 + 10-D/400)
Here, D is the rating advantage. With D = 100, the higher-rated expectation is approximately 0.6401, normally expressed as about 0.64.
Rating change = K × (actual score − expected score)
FIDE's official conversion chart places differences from 99 to 106 points at 0.64 for the higher-rated player and 0.36 for the lower-rated player.
Read the official FIDE Rating Regulations and scoring-probability rules.
In an Elo-based rating pool, a 100-point gap means the higher-rated player has about a 64% expected score and the lower-rated player about 36%. Test the distinction in case one of the 100-Point Rating Gap Quiz.
The higher-rated player's expected score is about 0.64, or 64% of the available points over time. Read the 1-game card in the Expected Score Cards section.
The lower-rated player's expected score is about 0.36, because the two expectations add to 1.00 for each game. Check the lower-rated value in the Expected Score Cards section.
No. The 64% figure is expected score, not pure win probability, because draws contribute half a point. Reject the win-probability claim in case two.
No. A rating advantage changes the long-run expectation but cannot guarantee the result of one game. Use case three to separate probability from certainty.
The higher-rated player is expected to score about 6.4 points from 10 games, while the lower-rated player is expected to score about 3.6. Confirm the multiplication in case four.
The higher-rated player is expected to score about 12.8 points from 20 games, while the lower-rated player is expected to score about 7.2. Use the 20-game card in the Expected Score Cards section.
Yes. The lower-rated player is an underdog, not a player with no winning chance, and individual upsets are completely compatible with Elo expectations. Answer case six.
Yes. The expected-score figure allows wins, draws, and losses; it does not prescribe one result distribution. Use case five before reading the Ten-Game Examples.
A draw counts as 0.5 points for each player, which is why expected score cannot be read as win probability. Apply the half-point rule in case five.
Yes. Six wins and one draw score 6.5 points, while five wins and three draws also score 6.5 points over 10 games. Compare both cards in the Ten-Game Examples.
It is meaningful but not overwhelming: 64% versus 36% is a real long-run edge while still leaving substantial room for draws and upsets. Use the direct-answer box as the practical benchmark.
Not reliably. One game contains too much tactical, psychological, and time-pressure variation to reveal a modest rating edge by itself. Use case three before judging strength from one result.
Not necessarily. Rating-class labels depend on the organisation and chosen boundaries, while the expected-score calculation uses the numerical difference. Keep the 64-to-36 interpretation from the Expected Score Cards section.
Not safely. Separate services may use different rating systems and player pools, so their 100-point gaps need not carry identical practical meaning. Reject the cross-pool claim in case seven.
Yes, if you want the clean Elo interpretation. The two ratings should belong to the same organisation, rating system, and time-control pool. Apply the Same-Pool Checklist.
They should not be assumed identical because online and FIDE ratings come from different pools, conditions, and sometimes different methods. Open the Online Versus FIDE Ratings card.
Treat rapid and blitz as separate pools. The arithmetic may look similar inside each system, but the ratings measure performance under different time conditions. Check time control in the Same-Pool Checklist.
Ratings are relative estimates built from results against the people in a particular pool, so the surrounding population gives the numbers their meaning. Complete case seven before comparing two gaps.
With an illustrative K-factor of 20 and a 64% expectation, the higher-rated player gains about 7.2 points before rounding for a win. Read the higher-rated-win card in the K=20 Rating Change Cards section.
With an illustrative K-factor of 20 and a 36% expectation, the lower-rated player gains about 12.8 points before rounding for a win. Read the upset-win card in the K=20 Rating Change Cards section.
A draw is below the higher-rated player's 0.64 expectation and above the lower-rated player's 0.36 expectation, so the higher-rated player normally loses points and the lower-rated player gains them. Read the draw card in the K=20 Rating Change Cards section.
The K-factor scales the update: a larger K produces a larger change for the same result and expectation. Use the K=20 cards as an illustration rather than a universal promise.
Not always. They are symmetrical in the simple example when both players use the same K-factor, but different K-values or rating procedures can produce different changes. Read the note below the K=20 cards.
The common expected-score formula is 1 divided by 1 plus 10 raised to the power of minus the rating difference divided by 400. Use the Formula Box to connect 100 points with approximately 0.64.
FIDE's conversion chart places rating differences from 99 to 106 at scoring probabilities of 0.64 for the higher-rated player and 0.36 for the lower-rated player. Open the official FIDE regulations link below the Formula Box.
No. Performance rating depends on the full set of opponents and the score achieved, not one isolated pre-game rating difference. Follow the Elo Formula card for the wider calculation context.
No. Ratings are estimates based on available results and can contain uncertainty, recent improvement, inactivity, or pool effects. Use the Same-Pool Checklist before treating the gap as precise.
Expect a competitive game with a modest edge to the higher-rated player rather than treating either result as predetermined. Complete the quiz, then choose opponents based on useful challenge as well as rating.
Next study expected-score calculations, K-factors, provisional ratings, rating pools, and the practical meaning of different rating bands. Choose the most relevant card in Continue the Rating Route.
Treat the rating gap as an expectation, then play the position rather than the number.
or create a ChessWorld username
Already have an account? Log in