Again not many posts but I have a 142 day playing streak on chess.com which means I've been at least one game a day for that length of time. My rating still ranges between 1300 and 1350 but many games have been lost through blunders caused by time pressure. If I can improve my time management, then my average rating should slowly climb.
I've started playing on lichess.org again after a long break. I'd forgotten the platform was completely free and relied on donations, unlike chess.com. I should at least make use of it. My rating on the two platforms is quite different and I asked Gemini why this was the case. Here was my query:
I play online chess on chess.com and my rating hovers around 1350. However, my rating on Lichess is about 1650. What is the reason for the disparity in my rating between these two platforms. Which one more accurately reflects my real level of competency?
 |
My rating graph on Lichess |
 |
My rating graph on chess.com |
Here was the response:
The 300-point gap between your ratings is completely normal and mathematically expected for your current bracket. The disparity is not a reflection of inconsistent play on your part, but rather the result of differing algorithms, starting baselines, and player demographics.
Reasons for the Disparity
- Starting Baselines: Lichess defaults every new account to a starting rating of 1500. Chess.com, on the other hand, allows new users to self-select their experience level upon account creation, often placing them at much lower starting points like 400, 800, or 1200. Because the baseline on Lichess is significantly higher, the entire bell curve is mathematically shifted upward for beginner and intermediate players.
- Mathematical Systems: Neither site actually uses the traditional Elo system. Chess.com uses the Glicko-1 rating algorithm, whereas Lichess uses Glicko-2. These systems calculate rating deviations and volatility differently, which naturally results in different numerical outputs.
- Player Pool Demographics: Chess.com has a massive, highly commercialized reach, attracting a huge volume of absolute beginners and casual players. This creates a dense compression of users at the bottom of the rating ladder. Lichess tends to attract a proportionally higher number of dedicated hobbyists. In the 15|10 rapid time control you play, you are facing different player populations on each site, altering how your relative strength is measured.
The More Accurate Reflection
To answer directly: your Chess.com rating of ~1350 is a much closer numerical reflection of your real-world competency.
While neither platform provides an official FIDE (World Chess Federation) rating, Chess.com's system generally tracks much closer to what a player's actual over-the-board FIDE rating would be. The Lichess rating scale sits hundreds of points higher than FIDE until players reach the master level (around 2200+), at which point the ratings across all platforms finally begin to converge.
It is important to note that your Lichess rating is not "fake" or "wrong." Online ratings do not act as an absolute, universal ruler of chess mastery; they simply measure your relative strength within a closed pool of active players. You are playing the exact same caliber of chess on both sites—the tape measure being used to score you is just calibrated to start at a different number.
One thing I discovered is that I've been mostly playing the London system with White. It's something that I settled upon because it suited my style and I gravitated toward it. Now that I know it's a well established system, I have access to a plethora of YouTube videos that discuss it. I'll be writing more about this system in future videos.
