Elo system
Mathleagues treats a submission as a match between a user and a problem. Both ratings react to the result.
the formula
R is the user rating, R_p is the problem rating, and S is marks awarded divided by marks available.user K-factors
| user Elo | K | purpose |
|---|---|---|
| below 1400 | 32 | Faster calibration for developing ratings. |
| 1400–2000 | 20 | Standard competitive volatility. |
| above 2000 | 12 | More stable expert ratings. |
partial credit
A score of 4/6 is S = 0.667, not a binary loss. The problem uses a fixed K-factor of 10 and updates from 1 − S, so it gains rating when solvers struggle and loses rating when they do well.
only your first attempt counts
Your first submission for a problem is the one that can move your rating and the problem's. You can try the problem again as often as you like to practise, but later submissions are graded without changing either rating, whatever the score.
user Elo = 1200 problem Elo = 1333 score = 4 / 6 = 0.667 K = 32 expected score E ≈ 0.317 user delta = round(32 × (0.667 − 0.317)) = +11 problem delta = round(10 × (0.333 − 0.683)) = −3
when Elo applies
On a first attempt, deterministic auto-graded results and non-flagged ai-graded results update both ratings straight away. Flagged work initially receives zero Elo impact. A human decision later applies the target effect from the stored submission-time context. A revised appeal applies only the difference between the revised target and the previously applied target. See reviews and appeals.
100 and 3000. New users start at 1200.