Chess Elo Calculator

Calculate Elo rating change after a chess game. Inputs: your rating, opponent rating, result, and K factor. Shows expected score and rating delta.

New Rating

1512.8

+12.8 points

Class C

Expected score (vs opponent)36.0%
Actual resultwin (1)
Rating delta+12.8
Current → New15001512.8
Tip:winning against higher-rated opponents earns more points; losing to them costs less. The system is designed so winning "feels" meaningful regardless of skill gap.
Chess Rating Classifications
Rating RangeClassificationDescription
Under 800BeginnerLearning rules and basic tactics
800 – 1200NoviceRecognizes basic patterns and threats
1200 – 1400Casual / Tournament-readySolid amateur; ~50th percentile of rated players
1400 – 1800Class C-B (Intermediate)Consistent tactics, opening theory, endgame basics
1800 – 2000Class A (Advanced)Strong club player; ~95th percentile
2000 – 2200ExpertTop 1% of tournament players
2200 – 2400Candidate MasterSenior Master / Candidate Master title eligible
2400 – 2500FIDE Master~50,000 worldwide
2500 – 2700IM / GM~1,800 grandmasters in the world
2700+Super GMTop 100 in the world

How to Use the Chess Elo Calculator

  1. Enter your current rating from your FIDE, USCF, or Chess.com profile.
  2. Enter your opponent's rating. For exact accuracy, use the rating they had before the game, not after.
  3. Select the game result: Win, Draw, or Loss.
  4. Choose the K factor: 40 for new players or juniors, 20 for standard adult ratings under 2400, 10 for masters. Chess.com uses different K values per time control (rapid 32, blitz 24, bullet 16).
  5. The calculator shows your new rating, point change, and the expected score you would've been predicted to earn (e.g., 35% expected against a stronger player).

The Elo Rating Formula Explained

Elo (created by Arpad Elo in the 1960s) ranks players by predicting the probability that one player will beat another based on their rating gap. The formula adjusts ratings after each game based on the gap between expected and actual result.

Expected Score = 1 / (1 + 10^((Opponent − You) / 400))

Score: Win = 1, Draw = 0.5, Loss = 0

New Rating = Current Rating + K × (Actual − Expected)

Example: Rated 1500, playing a 1600-rated opponent, with K=20.

  • Expected = 1 / (1 + 10^(100/400)) = 1 / (1 + 1.778) = 0.36 (36% chance)
  • If you win (score = 1): change = 20 × (1 − 0.36) = +12.8 → new rating 1513
  • If you lose (score = 0): change = 20 × (0 − 0.36) = -7.2 → new rating 1493
  • If you draw (score = 0.5): change = 20 × (0.5 − 0.36) = +2.8 → new rating 1503
Every 100 rating points of difference equals roughly a 64/36 win probability for the higher-rated player. A 200-point gap is ~76/24. A 400-point gap is ~91/9.

Elo Rating Across Platforms: Why Your Ratings Don't Match

If you play on chess.com, lichess, and FIDE-rated tournaments, you'll notice your rating varies by 200-400 points across them — sometimes more. This isn't random; each platform calibrates differently.

Platform / SystemStarting RatingNotes
FIDE (official)Provisional 1400, then real after 20 gamesMost conservative; over-the-board tournaments only
USCFProvisional based on first 4 gamesUS tournament rating; runs ~50-100 points higher than FIDE
Chess.com Rapid1200Wide rating spread; can run 200+ above OTB
Chess.com Blitz1200Time pressure shifts results; separate rating
Lichess1500 with high uncertaintyGlicko-2 (not pure Elo); uses RD; runs ~100-200 higher than chess.com
Chess Engines (Stockfish, etc.)Not Elo-comparableEngines benchmark differently — Stockfish rated ~3500+ is not directly comparable to human ratings

Three things to know about online vs over-the-board ratings:

  • Online ratings inflate. Most players are 100-300 points higher online than in tournaments. Online games favor pattern recognition and time management; tournament games favor calculation and stamina.
  • Time control matters. A player rated 1800 in rapid (10-30 min) often falls to 1500 in bullet (1-3 min). Tactical accuracy degrades fast under time pressure.
  • Lichess and Chess.com use different math. Lichess uses Glicko-2 (which includes a rating deviation factor); Chess.com uses traditional Elo. The same player will rate differently on each by 50-150 points consistently.

Frequently Asked Questions

Elo rating changes by K × (Actual Score − Expected Score). Expected score is the probability of winning based on the rating gap: Expected = 1 / (1 + 10^((Opponent − You) / 400)). K is typically 20 for adult players under 2400, 10 for masters, and 40 for new or junior players. Beating a higher-rated player earns more points; losing to a lower-rated player costs more.

Related Calculators