Track record
Every prediction we make is graded against the real result — walk-forward, using only what was known before the match. No hindsight, no cherry-picking, and no "value bet" claims. This is the whole record.
57.1%
Pick accuracy
0.240
Brier score
479.958
Graded predictions
’ M
Tracking since 02 M
518.963 matches · 4.287 players tracked across 7 leagues in 2 sports.
Track record over time
48 months · 517.794 graded · avg Brier 0.219
Monthly Brier from each match's pre-match prediction (walk-forward, no hindsight). A flat line below the 0.25 coin-flip mark means the model stays calibrated month after month — no drift.
Jul ’22coin-flip 0.25Jun ’26
Jan
0.240
57% · 30.509
Feb
0.241
56% · 28.107
Mar
0.240
57% · 31.121
Apr
0.241
57% · 30.913
May
0.240
57% · 32.262
Jun
0.239
58% · 25.492
By league
LeagueAccBriernGrade
🏓Challenger Series74.6%0.1759.322Good🇺🇦Setka Cup Women65.5%0.21911.806Good🇵🇱TT Elite Series60.1%0.233118.603Good🇺🇦Setka Cup57.4%0.240114.577Good🏸Badminton62.2%0.2412.079Good🇨🇿TT Cup55.5%0.245114.557Good🇨🇿Czech Liga Pro52.6%0.249109.014FairWhat these numbers mean. A Brier score measures how close our probabilities are to outcomes — 0.25 is a coin-flip, lower is better. Accuracy is how often the favourite we name wins. The calibration grade reflects whether a stated 60% really happens 60% of the time. We tested our model against the closing market and found no reliable edge — so we sell transparency, not tips. How to read these numbers →