Agent Lab — a live forecast exhibit · Applied AI
← Schedule
Group H · Matchday 16 · 2026-06-26

Cape Verde vs Saudi Arabia

Houston, United States · neutral venue

Full time CPV 0–0 KSA
The agent called Cape Verde (37%) — Better than the model call wrong

The forecast below is what the agent published before kickoff.

Most likely scoreline
0–0 21.2%
33.6%
CPV win
34.4%
Draw
31.9%
KSA win
Top scorelines
  • 0–0 21.2%
  • 1–0 15.5%
  • 0–1 15.0%
  • 1–1 10.9%
Total goals
0 to 10 goals · most likely → tallest

The agent’s call

3-agent ensemble · deepseek-chat
CPV win
36.7%
Draw
35.0%
KSA win
28.3%

The model and squad value both slightly favour Cape Verde due to superior recent form and a more valuable squad, but the absence of S. Lopes Cabral and Saudi Arabia's stronger opponent strength in recent matches (rank #44 vs #57) keep the match close. The prior's near-even split is reasonable, but I nudge slightly toward Cape Verde given their over-performance and squad edge, while keeping draw probability high due to low-scoring nature. Both teams have low expected goals (0.85-0.87), suggesting a tight, low-scoring affair. The most likely scorelines are 0-0, 1-0, 0-1, and 1-1, reflecting the defensive balance and lack of clear attacking dominance.

What the agent weighed
What the agent chose to investigate
The agent decided what to check on tavily — 6 sources kept:
  • “Cape Verde vs Saudi Arabia 2026-06-27 lineup injuries” match · 3 hits
    Check for late fitness updates or lineup changes beyond S. Lopes Cabral.
  • “Saudi Arabia squad 2026 June injuries suspensions” KSA · 3 hits
    Key Saudi player availability could shift forecast.
Agent scorelines
generated 2026-06-26
Last 5 — CPV
+0.61 pts/match vs model expectation · opp. avg #56
  • FIN #68 -0.3
  • SER #30 +2.1
  • BER #169 +0.1
  • ESP #1 +0.7
  • URU #12 +0.4
GF / GA last 10: 18 / 11
Last 5 — KSA
+0.06 pts/match vs model expectation · opp. avg #43
  • ECU #17 -0.7
  • PUE #166 +0.5
  • SEN #21 +0.1
  • URU #12 +0.4
  • ESP #1 -0.3
GF / GA last 10: 8 / 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
21.2
15.0
6.3
2.1
0.6
0.2
0.0
1
15.5
10.9
4.6
1.5
0.4
0.1
0.0
2
6.8
4.8
2.0
0.7
0.2
0.0
0.0
3
2.3
1.6
0.7
0.2
0.1
0.0
0.0
4
0.7
0.5
0.2
0.1
0.0
0.0
0.0
5
0.2
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 = CPV goals; columns = KSA 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)
0.86
CPV
0.82
KSA

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.