Loading statistics...

FootballOdds

AI Statistics

How often the AI predictions turned out right, measured on every pick published before kick-off.

Statistics are reloading…

Core markets

49.6%

580 correct from 1170 settled picks

Match Winner, Home/Away, Goals Over/Under and Goals Over/Under First Half

Matches
302
Correct
580
Incorrect
590

Best picks

+12.2 pts vs all

59.6%

56 correct from 94 settled picks

Matches
39
Correct
56
Incorrect
38

Opportunity picks

Not enough data

Only 24 settled picks match these filters. A rate is shown from 30 upward, so a handful of results cannot look like a track record.

Market leaderboard

Every market we publish, ranked by hit rate. Markets are not equally hard - Double Chance picks two outcomes of three, HT/FT Double one of nine - so a high place reflects difficulty as much as skill. The arrow is the comparable part: it tracks each market against its own past.

  1. 1
    Double Chance

    119 of 158 settled

    –
    75.3%
  2. 2
    Goals Over/Under core

    168 of 298 settled

    –
    56.4%
  3. 3
    Goals Over/Under First Half core

    154 of 295 settled

    –
    52.2%
  4. 4
    Asian Handicap

    146 of 285 settled

    –
    51.2%
  5. 5
    Handicap Result

    125 of 248 settled

    –
    50.4%
  6. 6
    Both Teams Score

    150 of 302 settled

    –
    49.7%
  7. 7
    Home/Away core

    126 of 276 settled

    –
    45.7%
  8. 8
    First Half Winner

    74 of 164 settled

    1
    45.1%
  9. 9
    Match Winner core

    132 of 301 settled

    1
    43.9%
  10. 10
    Second Half Winner

    117 of 300 settled

    –
    39.0%
  11. 11
    HT/FT Double

    63 of 273 settled

    –
    23.1%
All markets 2900 settled 47.4%

Core markets by month

Months with fewer than 30 settled picks are left out

75% 50% 25%
Nov Sep

A pick counts once its market can be settled from the final result. Voided markets and any we could not settle are left out of the rate rather than counted as a miss, and only predictions published before kick-off are included.