- Base
- — not yet computed
- Pre-game
- 97%
- Live
- At start
No live figure before the game starts; it updates from score and time remaining once it does.
Calibrated win, scoreline & player statistical probabilities — MLB, soccer & esports
Brier 0.200 · 60 games graded (last 7 days) · walk-forward validated · no odds, ever
0.200Brier score (lower is better)649games graded in public (all time)69.4%right so farsee the full report →Live games, upcoming matchups, and analytics — all in one place.
Data updated
No live figure before the game starts; it updates from score and time remaining once it does.
No live figure before the game starts; it updates from score and time remaining once it does.
No live figure before the game starts; it updates from score and time remaining once it does.
Predictions are pre-game and locked before start; the live column only shows how the model has moved since. The model breakdown lives on the full analysis page. Statistical research, not advice.
Every probability here is computed by our own engine, from event data rather than repackaged odds or someone else's projections. v2.0 makes soccer the core with per-league models refit weekly, and new simulators are validated in shadow before they count.
Most published probabilities are reverse-engineered from market prices. The engine never takes odds as an input; every number is computed from event data alone.
From v2.0, soccer result, totals, BTTS and margin all come off one scoreline distribution. Nothing is computed market by market, so two numbers cannot disagree.
A new engine such as the shot-level simulation runs and is graded daily, but stays at weight 0 with no effect on screen until it beats the existing model.
The first value shown becomes the graded record. A later model change never rewrites a past prediction.
Every finished game is scored and the calibration coefficients are refit. Losing predictions stay in the sample.
After v2.0, soccer anytime scorer, shot and shots-on-target props are also computed and graded daily on the shot-level simulator. They also run in shadow at weight 0 and do not appear until they beat the incumbent.
Injuries, form, xG, park and umpire adjustments are each capped at ±4-6% and thin samples are pulled toward the mean. An implausible live value is withheld rather than shown.
One email per day. No picks, no spam.
Free forever. By subscribing you consent to receive the daily probability email; unsubscribe any time from the link in every email.Privacy notice