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

Portugal vs DR Congo

Houston, United States · neutral venue

Full time POR 1–1 COD
The agent called Portugal (77%) — Worse than the model call wrong

The forecast below is what the agent published before kickoff.

Most likely scoreline
1–0 16.6%
63.6%
POR win
23.4%
Draw
12.9%
COD win
Top scorelines
  • 1–0 16.6%
  • 2–0 13.2%
  • 0–0 12.6%
  • 1–1 8.4%
Total goals
0 to 10 goals · most likely → tallest

The agent’s call

3-agent ensemble · deepseek-chat
POR win
77.3%
Draw
15.3%
COD win
7.3%

The model already favours Portugal (66% win, 1.91 xG), and the supplied intel strongly reinforces that. Portugal's recent form (+0.52 pts/match vs expectation against stronger opponents) and squad value (94th percentile, 7.1x DR Congo's) indicate a talent gap that the model may not fully capture. No injuries, neutral conditions, and DR Congo's weaker key players (lower club level and recent output) further tilt the match toward Portugal. I increase the home win probability to 75% and reduce draw/away accordingly. Portugal's expected goals (1.91) and DR Congo's low xG (0.55) suggest a low-scoring match for the underdog. The most likely scorelines are 2-0 (15%) and 1-0 (14%), with 3-0 (12%) also plausible given Portugal's firepower. Draws (0-0 or 1-1) are less likely due to the talent gap, and DR Congo winning is improbable.

What the agent weighed
What the agent chose to investigate
The agent decided what to check on tavily — 9 sources kept:
  • “Portugal vs DR Congo 2026-06-17 lineup injuries” match · 3 hits
    Late fitness or lineup changes could shift the forecast.
  • “Portugal squad 2026 World Cup injuries” POR · 3 hits
    Key player availability for Portugal is unknown.
  • “DR Congo squad 2026 World Cup injuries” COD · 3 hits
    Key player availability for DR Congo is unknown.
Agent scorelines
generated 2026-06-17
Last 5 — POR
+0.51 pts/match vs model expectation · opp. avg #44
  • ARM #99 +0.1
  • MEX #20 -0.7
  • USA #37 +1.0
  • CHI #28 +0.7
  • NIG #38 +0.6
GF / GA last 10: 26 / 9
Last 5 — COD
-0.24 pts/match vs model expectation · opp. avg #67
  • ALG #33 -1.1
  • BER #169 +0.2
  • JAM #91 +1.1
  • DEN #14 +0.0
  • CHI #28 -1.3
GF / GA last 10: 11 / 5
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
12.6
6.4
1.9
0.5
0.1
0.0
0.0
1
16.6
8.4
2.6
0.6
0.1
0.0
0.0
2
13.2
6.7
2.0
0.5
0.1
0.0
0.0
3
8.2
4.1
1.3
0.3
0.1
0.0
0.0
4
4.3
2.2
0.7
0.2
0.0
0.0
0.0
5
2.1
1.1
0.3
0.1
0.0
0.0
0.0
6
0.9
0.5
0.1
0.0
0.0
0.0
0.0

Rows = POR goals; columns = COD 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)
1.8
POR
0.56
COD

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.