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

DR Congo vs Uzbekistan

Atlanta, United States · neutral venue

Full time COD 3–1 UZB
The agent called DR Congo (41%) — Better than the model call right

The forecast below is what the agent published before kickoff.

Most likely scoreline
0–0 18.4%
39.8%
COD win
32.1%
Draw
28.1%
UZB win
Top scorelines
  • 0–0 18.4%
  • 1–0 15.9%
  • 0–1 12.7%
  • 1–1 11.0%
Total goals
0 to 10 goals · most likely → tallest

The agent’s call

3-agent ensemble · deepseek-chat
COD win
40.7%
Draw
33.0%
UZB win
26.3%

The model already gives DR Congo a slight edge (37% win vs 30%), and the supplied intel reinforces that lean. Uzbekistan's recent form is worse (DLLLL, -0.47 pts/match) against stronger opponents, while DR Congo's form is merely slightly below expectation. DR Congo's squad is ~2.3x more valuable and its key attackers play at higher club levels with better recent output (e.g., Kakuta 8G/3A, Bakambu at Real Betis) compared to Uzbekistan's Shomurodov (recent 2G/1A in 23 apps) and Shukurov (0G/2A). No injuries or conditions alter the picture. I nudge DR Congo's win probability up to 40% and reduce away win to 27%, keeping draw at 33% as both sides are underperforming. Expected goals (0.95 vs 0.80) suggest a low-scoring match. DR Congo's slightly stronger attack and Uzbekistan's poor form push the most likely scorelines toward 0-0, 1-0, and 0-1, with 1-1 also plausible. I keep the shape similar to the model but slightly increase DR Congo's scoring probability.

What the agent weighed
What the agent chose to investigate
The agent decided what to check on tavily — 9 sources kept:
  • “DR Congo vs Uzbekistan 2026 friendly lineup news” match · 3 hits
    Late fitness or lineup changes could shift expected strength.
  • “DR Congo squad injuries suspensions June 2026” COD · 3 hits
    Key player absences would significantly alter team strength.
  • “Uzbekistan squad injuries suspensions June 2026” UZB · 3 hits
    Key player absences would significantly alter team strength.
Agent scorelines
generated 2026-06-27
Last 5 — COD
-0.28 pts/match vs model expectation · opp. avg #29
  • JAM #91 +1.1
  • DEN #14 +0.0
  • CHI #28 -1.3
  • POR #6 +0.5
  • COL #8 -0.7
GF / GA last 10: 10 / 6
Last 5 — UZB
-0.44 pts/match vs model expectation · opp. avg #22
  • VEN #47 -0.2
  • CAN #41 -0.8
  • NED #9 -0.3
  • COL #8 -0.5
  • POR #6 -0.4
GF / GA last 10: 10 / 17
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
18.4
12.7
5.3
1.7
0.5
0.1
0.0
1
15.9
11.0
4.6
1.5
0.4
0.1
0.0
2
8.2
5.7
2.4
0.8
0.2
0.1
0.0
3
3.3
2.3
1.0
0.3
0.1
0.0
0.0
4
1.2
0.8
0.3
0.1
0.0
0.0
0.0
5
0.4
0.2
0.1
0.0
0.0
0.0
0.0
6
0.1
0.1
0.0
0.0
0.0
0.0
0.0

Rows = COD goals; columns = UZB 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)
1.05
COD
0.8
UZB

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.