Agent Lab — a live forecast exhibit · Applied AI
← Schedule
Group L · Matchday 17 · 2026-06-27

Croatia vs Ghana

Philadelphia, United States · neutral venue

Full time CRO 2–1 GHA
The agent called Croatia (56%) — Better than the model call right

The forecast below is what the agent published before kickoff.

Most likely scoreline
1–0 15.7%
54.6%
CRO win
26.3%
Draw
19.0%
GHA win
Top scorelines
  • 1–0 15.7%
  • 0–0 13.4%
  • 2–0 11.1%
  • 1–1 9.8%
Total goals
0 to 10 goals · most likely → tallest

The agent’s call

3-agent ensemble · deepseek-chat
CRO win
56.0%
Draw
25.7%
GHA win
18.3%

The statistical model gives Croatia a clear edge (55% win, 1.54 xG vs 0.71), and the squad market value corroborates a talent gap (~2x more valuable). Recent form shows Ghana over-performing expectations slightly, but against weaker opposition, while Croatia's form is neutral. No injuries, no extreme conditions, and no head-to-head history to alter the baseline. The prior stands: Croatia are the stronger side and should win most often, though Ghana's resilience keeps the draw plausible. Croatia's expected goals of 1.54 and Ghana's 0.71 point to a low-scoring match. The most likely scorelines are 1-0 (16%), 0-0 (14%), and 2-0 (11%), reflecting Croatia's defensive solidity and Ghana's limited attacking threat. A 1-1 draw is also possible (10%) if Ghana's over-performance trend continues.

What the agent weighed
What the agent chose to investigate
The agent decided what to check on tavily — 8 sources kept:
  • “Croatia vs Ghana 2026-06-27 team news injuries suspensions” match · 3 hits
    Late fitness or lineup changes could significantly shift the match forecast.
  • “Croatia squad 2026 World Cup injuries” CRO · 3 hits
    Key player absences like Modric or Gvardiol would alter Croatia's strength.
  • “Ghana squad 2026 World Cup injuries” GHA · 2 hits
    Key player absences like Kudus or Partey would alter Ghana's strength.
Agent scorelines
generated 2026-06-27
Last 5 — CRO
-0.04 pts/match vs model expectation · opp. avg #28
  • BRA #3 -0.8
  • BEL #10 -1.4
  • SLO #50 +0.9
  • ENG #4 -0.9
  • PAN #72 +0.8
GF / GA last 10: 17 / 14
Last 5 — GHA
+0.44 pts/match vs model expectation · opp. avg #29
  • GER #7 -0.3
  • MEX #20 -0.7
  • WAL #44 +0.0
  • PAN #72 +1.4
  • ENG #4 +0.5
GF / GA last 10: 10 / 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
13.4
8.4
3.1
0.9
0.2
0.1
0.0
1
15.7
9.8
3.7
1.1
0.3
0.1
0.0
2
11.1
7.0
2.6
0.8
0.2
0.0
0.0
3
6.1
3.8
1.4
0.4
0.1
0.0
0.0
4
2.9
1.8
0.7
0.2
0.0
0.0
0.0
5
1.2
0.8
0.3
0.1
0.0
0.0
0.0
6
0.5
0.3
0.1
0.0
0.0
0.0
0.0

Rows = CRO goals; columns = GHA 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)
1.54
CRO
0.72
GHA

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.