Chess ratings confuse a lot of players — not just beginners. Your Chess.com blitz rating, your Lichess rapid rating, and your official FIDE rating can all show completely different numbers for the same player. Understanding why requires understanding how chess ratings actually work.
What a Chess Rating Measures
A chess rating is a numerical estimate of a player's skill level relative to other players in the same rating pool. It's not an absolute measure of chess ability — it's a relative measure within a specific system.
Two important implications:
- A 1500 on Chess.com doesn't mean the same thing as a 1500 on Lichess, or a 1500 FIDE rating. Different rating pools, different calibrations.
- Your rating only means something in comparison to other ratings in the same system. A 1500 is "better than X% of players in this pool" — not an absolute skill certification.
The Elo System
The Elo system, invented by Arpad Elo in the 1960s, is the mathematical foundation of most chess ratings. The core idea:
Every player has an expected performance level. Before each game, the system predicts the probability of either player winning based on the rating difference. After the game, the actual result is compared to the expected result, and ratings are adjusted.
The formula in plain English: If you beat someone you were expected to beat, you gain a few points. If you lose to someone you were expected to beat, you lose many points. If you beat someone much stronger than you, you gain many points. The further the result deviates from the expected outcome, the larger the rating change.
The K-factor determines how sharply ratings respond to individual results. Higher K-factor = faster change. Chess.com uses higher K-factors than FIDE for most players, which is why online ratings swing more dramatically after a few games.
Elo vs. Glicko vs. Glicko-2
Modern rating systems have improved on the original Elo:
Elo — the original. Only adjusts for wins/losses. Doesn't account for how long since a player last played (rating reliability).
Glicko — adds a "ratings deviation" (RD) value that measures certainty. A player who hasn't played recently has a high RD (uncertain rating). After several games, RD drops (rating becomes more reliable). Lichess uses a variant of Glicko.
Glicko-2 — adds a volatility parameter that measures whether a player's performance is consistent or erratic. Chess.com's current system uses a variant of Glicko-2.
Practical difference: If you haven't played for a few weeks and come back, your rating is considered less certain. A few strong results will move your rating more dramatically than if you'd been playing continuously.
Why Your Chess.com and Lichess Ratings Differ
The two platforms use different:
- Starting ratings (Lichess starts new players at 1500; Chess.com estimates based on early games)
- K-factors (how fast ratings adjust)
- Rating pools (different player populations)
- Calibration (what "1500" means on each platform)
In practice, most players find that:
- Lichess ratings tend to be 100–300 points higher than equivalent Chess.com ratings
- Neither accurately reflects FIDE Elo without adjustment
- Differences also vary by time control — a Lichess blitz rating vs Chess.com rapid rating aren't comparable at all
Don't compare ratings across platforms directly. Compare your own rating within the same platform over time — that's the meaningful number.
What Different Elo Ranges Actually Mean
On Chess.com (approximately):
| Rating | Skill level |
|---|---|
| Under 400 | Complete beginner — still learning the rules and basic piece movement |
| 400–800 | Beginner — understands the rules, makes basic tactics errors frequently |
| 800–1200 | Novice — plays complete games, misses simple tactics |
| 1200–1500 | Casual player — recognizes common patterns, has some opening knowledge |
| 1500–1800 | Intermediate — solid tactical player, understands positional concepts |
| 1800–2000 | Advanced amateur — rarely loses to simple tactics, strong positional player |
| 2000–2200 | Expert/Candidate Master — tournament-competitive, strong in all game phases |
| 2200+ | Master and above — top few percent, strong enough for serious tournament play |
FIDE ratings run approximately 200–350 points below equivalent Chess.com ratings. A 1700 on Chess.com might correspond to 1400–1500 FIDE, though this varies significantly.
Rating Volatility Is Normal
New players see their ratings swing 200–300 points in either direction over the first few hundred games. This isn't unusual — the system is still finding your actual level.
Even established players experience volatility of 100+ points depending on form, preparation, and which opponents they happen to face. Don't treat your current number as a fixed label. It's a running estimate.
The number to track: Not today's rating, but the trend over 3–6 months. A rating that's 100 points higher than six months ago, despite normal swings, represents real improvement.
Why Your Rating Can Drop While You're Getting Better
This happens and is disorienting. Possible reasons:
- Learning new concepts that temporarily disrupt established habits. Adding opening study sometimes leads to short-term losses in positions you've moved out of your comfort zone into.
- Playing at a harder time control. Switching from blitz to rapid exposes calculation weaknesses you didn't face under time pressure.
- Facing tougher opponents. If you've been matched against strong opponents for a stretch, the expected score is lower.
In all these cases, the skill improvement is real even if the rating temporarily drops. This is most common when players deliberately train their weaknesses — the weakness gets worse before it gets better as you rewire habits.
How to Use Your Rating Constructively
The rating's primary value is as a feedback loop over time, not as a status marker. Use it to:
- Identify training plateau regions (rating stuck at the same number for 3+ months signals a systematic weakness)
- Set realistic benchmarks (improving 100 points in 3 months is ambitious but achievable with focused training; 400 points in a month is not)
- Choose appropriate opponents for games (playing at your level produces more useful feedback than constant upsets or easy wins)
What the rating doesn't tell you: which specific skills to improve, whether you're training correctly, or how close you are to your ceiling. For that, you need game analysis.
Your rating is a useful signal, not an identity. Understanding how it's calculated helps you interpret it correctly — and stops you from either chasing the number obsessively or dismissing it entirely. Both extremes waste the genuinely useful information the number provides.
For actual improvement, focus on understanding your games: where the evaluation shifts, what you missed, what the engine saw that you didn't. ChessSolve shows Stockfish's suggestions in real time during your Chess.com and Lichess games, giving you that feedback in the moment it matters most.