Elo is the rating system used by basically every chess platform. Here's what your number means, how it changes, and why ratings categories matter.
Every chess platform — chess.com, Lichess, FIDE-rated tournaments, ChessLvl — uses some version of the Elo rating system to estimate player strength. The system was invented by Arpad Elo, a Hungarian-American physics professor, in the 1960s. It's so good at modeling competitive skill that it's been adopted across most competitive games beyond chess: Scrabble, Go, esports, even some sports analytics.
Every player has a single number representing their estimated skill. When two players play, the system predicts the higher-rated player wins more often. After the game, ratings update based on what actually happened: if the favorite won, both ratings barely move. If the underdog won, the underdog gains a lot and the favorite loses a lot.
The math: the rating change per game is calculated from the rating difference and the actual result vs the expected result. A 100-point favorite has roughly a 64% expected win probability. If the favorite wins, they gain ~6-8 points. If they lose, they lose ~24-32 points. The system self-corrects over time — players with too-high ratings lose them back; players with too-low ratings climb.
The 'K-factor' controls how quickly ratings change. New players have high K-factors (their ratings change fast as the system figures out their skill). Established players have lower K-factors (their ratings change slowly because each game is small evidence). Different platforms use different K-factors.
Rough rating bands and what's typical at each level:
These bands shift slightly across platforms — chess.com ratings tend to be higher than FIDE for the same player, Lichess ratings tend to be even higher than chess.com. If you have a 1500 chess.com rating, you're probably ~1300 FIDE. The relative skill is what matters, not the absolute number.
Same player can have very different ratings on chess.com, Lichess, and OTB (over-the-board, FIDE). Reasons:
First, the player pool differs. Lichess attracts more casual players; FIDE OTB pools are more serious. So a 1500 on Lichess (median of casual players) corresponds to a stronger player than a 1500 in FIDE (median of tournament players).
Second, time controls differ. Most online chess is played at 5-10 minute time controls (rapid or blitz). FIDE tournaments are mostly 60-120 minute classical. Players have different ratings at different speeds because faster chess emphasizes intuition over calculation. A grandmaster might be 2700 classical but 2400 bullet.
Third, K-factors and rating math differ slightly. Chess.com uses a custom Glicko-based system. Lichess uses Glicko2. FIDE uses traditional Elo. The differences are small but accumulate over many games.
Beginners obsess over their rating. They check it after every game. They feel devastated when it drops. They tell themselves they need to be 1500 before they can really enjoy chess. This is mostly counterproductive.
Your rating is a measurement, not a goal. The goal is to improve at chess — to play better moves, see more patterns, calculate more accurately. The rating is a side effect of those things. Players who focus on rating obsess over short-term results, take shortcuts (playing easy opponents to pad rating), and avoid challenging positions where they might learn but might lose. Players who focus on improvement gain rating naturally as a consequence.
There's also short-term volatility to consider. A 1500 player will have stretches of looking like a 1700 and stretches of looking like a 1300. Daily rating fluctuations of ±50 points are noise. Treat your rating as a rolling average over months, not a snapshot of today.
Instead of 'reach 1500 ELO,' try goals like:
These are skill-based goals. Hit them and your rating climbs as a consequence. Skip them and your rating stagnates regardless of how often you play. The work, not the number, is what matters.
"Chess, like love, like music, has the power to make people happy."
— Siegbert Tarrasch