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

New Zealand vs Belgium

Vancouver, Canada · neutral venue

Full time NZL 1–5 BEL
The agent called Belgium (80%) — Better than the model call right

The forecast below is what the agent published before kickoff.

Most likely scoreline
0–2 13.3%
6.7%
NZL win
14.8%
Draw
77.2%
BEL win
Top scorelines
  • 0–2 13.3%
  • 0–1 12.7%
  • 0–3 10.9%
  • 0–4 7.6%
Total goals
0 to 10 goals · most likely → tallest

The agent’s call

3-agent ensemble · deepseek-chat
NZL win
6.0%
Draw
14.0%
BEL win
80.0%

The model's prior (74% away win) is already strong, and the supplied intel reinforces it. Belgium's squad value is 14.4x New Zealand's, and their key players (De Bruyne, Tielemans, Trossard, Lukaku) are world-class, while New Zealand's best players come from lower-tier leagues. The only slight tempering factor is Doku's absence, but Belgium still has ample attacking talent. Conditions are neutral. I increase away win probability slightly to 82% to reflect the squad quality gap and key player advantage. Belgium is expected to dominate possession and chances, with New Zealand likely to defend deep. The model's top scorelines (0-1, 0-2, 0-3) align with Belgium's expected goals of 2.40 and New Zealand's 0.48. I keep the same shape but slightly boost the away win scorelines to reflect the talent gap.

What the agent weighed
What the agent checked on the web
Searched tavily for NZL · BEL — 6 sources kept
Agent scorelines
generated 2026-06-26
Last 5 — NZL
-0.14 pts/match vs model expectation · opp. avg #39
  • CHI #28 +2.3
  • HAI #96 -1.1
  • ENG #4 -0.1
  • IRN #31 +0.4
  • EGY #34 -0.7
GF / GA last 10: 9 / 19
Last 5 — BEL
-0.27 pts/match vs model expectation · opp. avg #31
  • MEX #20 -0.8
  • CRO #15 +1.6
  • TUN #53 +0.6
  • EGY #34 -0.9
  • IRN #31 -0.9
GF / GA last 10: 26 / 7
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
7.3
12.7
13.3
10.9
7.6
4.8
2.8
1
3.2
5.7
5.9
4.9
3.4
2.2
1.3
2
0.9
1.5
1.6
1.3
0.9
0.6
0.3
3
0.2
0.3
0.3
0.3
0.2
0.1
0.1
4
0.0
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 = NZL goals; columns = BEL 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)
0.49
NZL
2.7
BEL

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.