Agent Lab — a live forecast exhibit · Applied AI
← Schedule
Group I · Matchday 12 · 2026-06-22

France vs Iraq

Philadelphia, United States · neutral venue

Full time FRA 3–0 IRQ
The agent called France (81%) — Better than the model call right

The forecast below is what the agent published before kickoff.

Most likely scoreline
1–0 14.8%
76.3%
FRA win
16.3%
Draw
6.5%
IRQ win
Top scorelines
  • 1–0 14.8%
  • 2–0 14.6%
  • 3–0 11.3%
  • 0–0 9.0%
Total goals
0 to 10 goals · most likely → tallest

The agent’s call

3-agent ensemble · deepseek-chat
FRA win
81.0%
Draw
14.0%
IRQ win
5.0%

The model already heavily favours France (72% win), and the supplied intel reinforces that edge. France's recent form is strong against decent opposition, while Iraq lost 4-1 to Norway in their opener, highlighting defensive frailty. The enormous squad value gap (France 98th percentile, Iraq 2nd) and elite attacking talent (Mbappé, Olise, Dembélé) suggest the model's 72% may be conservative. I raise home win probability to 80%, with draw at 15% and away win at 5%, reflecting Iraq's limited threat and France's likely dominance. France's expected goals (2.18) and Iraq's weak defence (conceded 4 to Norway) point to a multi-goal French win. Clean-sheet odds are high given Iraq's attack (0.45 xG model). The most likely scorelines are 2-0 and 3-0, with 1-0 also plausible if France are wasteful. Higher-scoring outcomes (4-0) are possible but less probable due to France's potential complacency.

What the agent weighed
What the agent chose to investigate
The agent decided what to check on tavily — 9 sources kept:
  • “France injury news June 2026” FRA · 3 hits
    Key player injuries could significantly impact France's strength.
  • “Iraq squad news June 2026” IRQ · 3 hits
    Late fitness or lineup changes for Iraq could affect match outcome.
  • “France vs Iraq weather Philadelphia June 22 2026” match · 3 hits
    Extreme weather conditions can alter playing style and forecast.
Agent scorelines
generated 2026-06-22
Last 5 — FRA
+0.32 pts/match vs model expectation · opp. avg #25
  • BRA #3 +1.8
  • COL #8 +1.5
  • CIV #26 -2.3
  • NOR #65 +0.4
  • SEN #21 +1.1
GF / GA last 10: 26 / 10
Last 5 — IRQ
-0.04 pts/match vs model expectation · opp. avg #61
  • BOL #73 +1.8
  • AND #167 +0.8
  • ESP #1 +0.9
  • VEN #47 -1.0
  • NOR #19 -0.5
GF / GA last 10: 11 / 13
Head-to-head (last 5)

No prior meetings in the dataset.

▸ Full score grid (7×7)
P(home goals × away goals), %
0
1
2
3
4
5
6
0
9.0
3.5
0.8
0.2
0.0
0.0
0.0
1
14.8
5.8
1.4
0.2
0.0
0.0
0.0
2
14.6
5.7
1.3
0.2
0.0
0.0
0.0
3
11.3
4.4
1.0
0.2
0.0
0.0
0.0
4
7.5
2.9
0.7
0.1
0.0
0.0
0.0
5
4.4
1.7
0.4
0.1
0.0
0.0
0.0
6
2.5
1.0
0.2
0.0
0.0
0.0
0.0

Rows = FRA goals; columns = IRQ 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)
2.47
FRA
0.42
IRQ

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.