Agent Lab — a live forecast exhibit · Applied AI
← Schedule
Group L · Matchday 7 · 2026-06-17

Ghana vs Panama

Toronto, Canada · neutral venue

Full time GHA 1–0 PAN
The agent called Panama (37%) — Worse than the model call wrong

The forecast below is what the agent published before kickoff.

Most likely scoreline
0–0 15.1%
41.7%
GHA win
29.5%
Draw
28.8%
PAN win
Top scorelines
  • 0–0 15.1%
  • 1–0 14.4%
  • 0–1 11.5%
  • 1–1 11.0%
Total goals
0 to 10 goals · most likely → tallest

The agent’s call

3-agent ensemble · deepseek-chat
GHA win
33.3%
Draw
30.0%
PAN win
36.7%

The prior model slightly favours Ghana (40% win) but the key factor is Thomas Partey's confirmed absence due to visa denial, which removes Ghana's most influential midfielder. Ghana's recent form is poor (LLLLD) while Panama's is solid (DWLWD), and though Ghana's squad value is higher, the Partey loss and form gap tilt the match toward a more balanced outcome. I adjust the prior down for Ghana and up for Panama, making the match essentially a toss-up with a slight lean to Panama given the intangibles. With Partey missing, Ghana's midfield control and creativity drop, reducing their expected goals from the model's 1.21. Panama's defence is organized (Murillo, Davis) and they can counter. A low-scoring draw or a narrow win for either side is most likely; the 0-0, 1-0, 0-1, and 1-1 scorelines dominate, with the model's original distribution slightly shifted toward Panama's favour.

What the agent weighed
What the agent chose to investigate
Agent scorelines
generated 2026-06-17
Last 5 — GHA
-0.42 pts/match vs model expectation · opp. avg #27
  • KOR #40 -1.0
  • AUT #23 -0.8
  • GER #7 -0.3
  • MEX #20 -0.7
  • WAL #44 +0.0
GF / GA last 10: 11 / 14
Last 5 — PAN
+0.20 pts/match vs model expectation · opp. avg #68
  • RSA #61 -0.0
  • RSA #61 +2.0
  • BRA #3 -0.1
  • DOM #149 +0.5
  • BIH #64 -0.5
GF / GA last 10: 18 / 16
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
15.1
11.5
5.3
1.9
0.6
0.2
0.0
1
14.4
11.0
5.1
1.8
0.6
0.2
0.0
2
8.3
6.3
2.9
1.0
0.3
0.1
0.0
3
3.7
2.8
1.3
0.5
0.1
0.0
0.0
4
1.4
1.1
0.5
0.2
0.1
0.0
0.0
5
0.5
0.4
0.2
0.1
0.0
0.0
0.0
6
0.2
0.1
0.1
0.0
0.0
0.0
0.0

Rows = GHA goals; columns = PAN 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)
1.18
GHA
0.9
PAN

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.