Agent Lab — a live forecast exhibit · Applied AI
← Schedule
Group F · Matchday 10 · 2026-06-20

Tunisia vs Japan

Guadalupe, Mexico · neutral venue

Full time TUN 0–4 JPN
The agent called Japan (53%) — Better than the model call right

The forecast below is what the agent published before kickoff.

Most likely scoreline
0–1 15.6%
22.0%
TUN win
27.8%
Draw
50.2%
JPN win
Top scorelines
  • 0–1 15.6%
  • 0–0 14.3%
  • 1–1 10.3%
  • 0–2 10.2%
Total goals
0 to 10 goals · most likely → tallest

The agent’s call

3-agent ensemble · deepseek-chat
TUN win
20.0%
Draw
26.7%
JPN win
53.3%

Japan's superior recent form (+0.89 pts/match vs expectation), much higher squad value (4.1x), and deeper attacking talent (Ueda, Kamada, Nakamura) all point to a clear advantage. Tunisia's turmoil (coach fired after heavy loss) and underperformance (-0.58 pts/match) further tilt the scale. The model's prior (48% away win) is already Japan-favored, but the intel justifies a moderate upward adjustment to 53% away win, with draw and home win reduced accordingly. Japan's attacking firepower (multiple players with strong recent goal contributions) suggests they can score 1-2 goals. Tunisia's defense (Talbi, Abdi) is modest but not porous, so a clean sheet is plausible. The most likely scorelines reflect a low-scoring Japan win (0-1, 0-2) and a draw (1-1, 0-0) as secondary outcomes.

What the agent weighed
What the agent chose to investigate
The agent decided what to check on tavily — 9 sources kept:
  • “Tunisia vs Japan 2026-06-21 lineup injuries” match · 3 hits
    Late fitness or lineup news could significantly shift the match forecast.
  • “Tunisia squad 2026 June injuries suspensions” TUN · 3 hits
    Key player absences would alter Tunisia's strength.
  • “Japan squad 2026 June injuries suspensions” JPN · 3 hits
    Key player absences would alter Japan's strength.
Agent scorelines
generated 2026-06-20
Last 5 — TUN
-0.55 pts/match vs model expectation · opp. avg #39
  • HAI #96 +1.0
  • CAN #41 +0.0
  • AUT #23 -0.8
  • BEL #10 -0.4
  • SWE #24 -1.0
GF / GA last 10: 12 / 17
Last 5 — JPN
+0.89 pts/match vs model expectation · opp. avg #37
  • BOL #73 +0.6
  • SCO #39 +1.5
  • ENG #4 +2.4
  • ICE #60 +0.7
  • NED #9 -0.0
GF / GA last 10: 15 / 8
Head-to-head (last 5)
  • 2022-06-14 Japan 0–3 Tunisia
  • 2015-03-27 Japan 2–0 Tunisia
  • 2003-10-08 Tunisia 0–1 Japan
  • 2002-06-14 Japan 2–0 Tunisia
  • 1996-10-13 Japan 1–0 Tunisia
▸ Full score grid (7×7)
P(home goals × away goals), %
0
1
2
3
4
5
6
0
14.3
15.6
10.2
5.2
2.3
0.9
0.3
1
9.5
10.3
6.8
3.5
1.5
0.6
0.2
2
3.8
4.1
2.7
1.4
0.6
0.2
0.1
3
1.2
1.3
0.8
0.4
0.2
0.1
0.0
4
0.3
0.3
0.2
0.1
0.0
0.0
0.0
5
0.1
0.1
0.1
0.0
0.0
0.0
0.0
6
0.0
0.0
0.0
0.0
0.0
0.0
0.0

Rows = TUN goals; columns = JPN goals. Above-diagonal cells are home wins; diagonal = draws; below = away wins. Truncated at 6 — the joint mass past that is < 1 %.

Expected goals (Poisson rate, λ)
0.76
TUN
1.4
JPN

Independent-Poisson rates. W/D/L and the score grid integrate the full 13×13 joint under these two parameters.

News context — injuries, suspensions, projected XI — is not part of the model, which sees only past results.