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

Uzbekistan vs Colombia

Mexico City, Mexico · neutral venue

Full time UZB 1–3 COL
The agent called Colombia (62%) — Worse than the model call right

The forecast below is what the agent published before kickoff.

Most likely scoreline
0–1 17.3%
11.5%
UZB win
23.1%
Draw
65.3%
COL win
Top scorelines
  • 0–1 17.3%
  • 0–2 13.9%
  • 0–0 13.0%
  • 0–3 8.7%
Total goals
0 to 10 goals · most likely → tallest

The agent’s call

3-agent ensemble · deepseek-chat
UZB win
13.0%
Draw
25.0%
COL win
62.0%

The model's prior (59% away win) is already strong, and the supplied intel reinforces it. Colombia's recent form is solid against tougher opponents, their squad is far more valuable, and their key players are at higher-caliber clubs. Uzbekistan's recent form is poor and their squad depth is limited. I increase Colombia's win probability to 64% to reflect the talent gap, with the draw at 24% and Uzbekistan's win at 12%. Colombia's expected goals (1.54) and Uzbekistan's (0.55) from the model suggest a low-scoring match favoring Colombia. The most likely scoreline is 0-1 (18%), followed by 0-2 (15%) and 0-0 (12%). Colombia's defensive solidity and Uzbekistan's attacking limitations point to a clean sheet for Colombia in most scenarios.

What the agent weighed
What the agent chose to investigate
The agent decided what to check on tavily — 6 sources kept:
  • “Uzbekistan vs Colombia 2026-06-18 lineup injuries suspensions” match · 3 hits
    Late fitness or lineup news could change the forecast significantly.
  • “Mexico City altitude effect on football match” match · 3 hits
    Altitude in Mexico City may affect player performance and match outcome.
Agent scorelines
generated 2026-06-17
Last 5 — UZB
-0.25 pts/match vs model expectation · opp. avg #55
  • CHI #98 -0.8
  • GAB #82 +1.1
  • VEN #47 -0.2
  • CAN #41 -0.8
  • NED #9 -0.3
GF / GA last 10: 14 / 9
Last 5 — COL
-0.12 pts/match vs model expectation · opp. avg #38
  • AUS #29 +1.1
  • CRO #15 -1.6
  • FRA #5 -1.2
  • COS #66 +0.5
  • JOR #75 +0.4
GF / GA last 10: 25 / 10
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
13.0
17.3
13.9
8.7
4.7
2.2
1.0
1
6.0
8.0
6.4
4.0
2.2
1.0
0.5
2
1.7
2.2
1.8
1.1
0.6
0.3
0.1
3
0.4
0.5
0.4
0.2
0.1
0.1
0.0
4
0.1
0.1
0.1
0.0
0.0
0.0
0.0
5
0.0
0.0
0.0
0.0
0.0
0.0
0.0
6
0.0
0.0
0.0
0.0
0.0
0.0
0.0

Rows = UZB goals; columns = COL 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)
0.51
UZB
1.83
COL

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.