Agent Lab — a live forecast exhibit · Applied AI
← Schedule
Group D · Matchday 9 · 2026-06-19

United States vs Australia

Seattle, United States

Full time USA 2–0 AUS
The agent called United States (47%) — Better than the model call right

The forecast below is what the agent published before kickoff.

Most likely scoreline
1–0 12.4%
41.6%
USA win
27.2%
Draw
31.2%
AUS win
Top scorelines
  • 1–0 12.4%
  • 0–0 12.1%
  • 1–1 10.8%
  • 0–1 10.5%
Total goals
0 to 10 goals · most likely → tallest

The agent’s call

3-agent ensemble · deepseek-chat
USA win
47.3%
Draw
27.0%
AUS win
25.7%

The model's prior (41% USA win) understates the gap in squad quality and recent form quality. USA's squad is vastly more valuable and features players at top European clubs, while Australia's key players largely come from lower-tier leagues. USA also has a strong head-to-head record and faced tougher opponents in recent form. I adjust USA's win probability upward to 48%, reducing away win to 25%, keeping draw near model at 27%. USA's expected goals rise to ~1.5 given the talent gap, while Australia's drop to ~0.9. The most likely scorelines reflect a low-scoring but USA-favored match: 1-0 (14%), 1-1 (12%), 2-0 (11%), 0-0 (10%). The clean sheet for USA is plausible given Australia's weaker attack.

What the agent weighed
What the agent chose to investigate
The agent decided what to check on tavily — 9 sources kept:
  • “USA vs Australia 2026-06-19 lineups injuries” match · 3 hits
    Late fitness or lineup changes could significantly alter match outcome.
  • “USA national team squad news June 2026” USA · 3 hits
    Key player availability or suspensions may shift forecast.
  • “Australia national team squad news June 2026” AUS · 3 hits
    Key player availability or suspensions may shift forecast.
Agent scorelines
generated 2026-06-19
Last 5 — USA
+0.39 pts/match vs model expectation · opp. avg #16
  • BEL #10 -1.0
  • POR #6 -0.8
  • SEN #21 +1.6
  • GER #7 -0.9
  • PAR #35 +1.4
GF / GA last 10: 22 / 16
Last 5 — AUS
+0.50 pts/match vs model expectation · opp. avg #46
  • CAM #51 +1.2
  • CUW #119 +0.4
  • MEX #20 -1.2
  • SUI #16 +0.2
  • TUR #25 +1.7
GF / GA last 10: 14 / 10
Head-to-head (last 5)
  • 2025-10-14 United States 2–1 Australia
  • 2010-06-05 Australia 1–3 United States
  • 1998-11-06 United States 0–0 Australia
  • 1992-06-13 United States 0–1 Australia
▸ Full score grid (7×7)
P(home goals × away goals), %
0
1
2
3
4
5
6
0
12.1
10.5
5.5
2.3
0.8
0.2
0.1
1
12.4
10.8
5.7
2.3
0.8
0.3
0.1
2
7.7
6.7
3.5
1.4
0.5
0.2
0.0
3
3.7
3.2
1.7
0.7
0.2
0.1
0.0
4
1.5
1.3
0.7
0.3
0.1
0.0
0.0
5
0.6
0.5
0.3
0.1
0.0
0.0
0.0
6
0.2
0.2
0.1
0.0
0.0
0.0
0.0

Rows = USA goals; columns = AUS 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)
1.3
USA
1.06
AUS

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.