Agent Lab — a live forecast exhibit · Applied AI
← Schedule
Group D · Matchday 15 · 2026-06-25

Paraguay vs Australia

Santa Clara, United States · neutral venue

Full time PAR 0–0 AUS
The agent called Paraguay (35%) — Worse than the model call wrong

The forecast below is what the agent published before kickoff.

Most likely scoreline
0–0 20.8%
32.1%
PAR win
34.1%
Draw
33.8%
AUS win
Top scorelines
  • 0–0 20.8%
  • 0–1 15.4%
  • 1–0 14.9%
  • 1–1 11.0%
Total goals
0 to 10 goals · most likely → tallest

The agent’s call

3-agent ensemble · deepseek-chat
PAR win
35.0%
Draw
32.0%
AUS win
33.0%

The model and squad value both point to a near-even match, but key intel shifts the balance slightly. Paraguay's superior squad value (3x) is offset by the absence of key midfielder Almirón and a poor historical head-to-head (0W 2D 3L). Australia's recent form is in line with expectations against stronger opponents, while Paraguay over-performed against weaker sides. The absence of Almirón and Australia's unbeaten H2H record push the prior slightly toward Australia, but the squad talent gap keeps Paraguay competitive. I move the away win probability up a few points from the model's 34% to 32%, and home win up from 33% to 35%, reflecting the conflicting signals. Both teams have low expected goals (0.88 vs 0.91), suggesting a low-scoring affair. The most likely scorelines are 0-0, 1-0, 0-1, and 1-1, consistent with the model's top four. The absence of Almirón may reduce Paraguay's attacking threat, but Australia's defense is not elite. I keep the scoreline distribution close to the model, with a slight tilt toward Paraguay scoring first due to squad value.

What the agent weighed
What the agent chose to investigate
The agent decided what to check on tavily — 6 sources kept:
  • “Paraguay vs Australia 2026-06-26 lineup news” match · 3 hits
    Late fitness or lineup changes could shift probabilities.
  • “Australia squad injury suspension 2026-06-26” AUS · 3 hits
    Key player availability for Australia is unknown.
Agent scorelines
generated 2026-06-25
Last 5 — PAR
+0.32 pts/match vs model expectation · opp. avg #52
  • GRE #36 +1.9
  • MAR #11 -0.9
  • NIC #151 +0.3
  • USA #37 -1.1
  • TUR #25 +1.7
GF / GA last 10: 14 / 13
Last 5 — AUS
-0.05 pts/match vs model expectation · opp. avg #43
  • CUW #119 +0.4
  • MEX #20 -1.2
  • SUI #16 +0.2
  • TUR #25 +1.7
  • USA #37 -1.2
GF / GA last 10: 11 / 11
Head-to-head (last 5)
  • 2010-10-09 Australia 1–0 Paraguay
  • 2006-10-07 Australia 1–1 Paraguay
  • 2000-06-15 Australia 2–1 Paraguay
  • 2000-06-12 Australia 0–0 Paraguay
  • 2000-06-09 Australia 0–0 Paraguay
▸ Full score grid (7×7)
P(home goals × away goals), %
0
1
2
3
4
5
6
0
20.8
15.4
6.8
2.3
0.7
0.2
0.0
1
14.9
11.0
4.9
1.7
0.5
0.1
0.0
2
6.4
4.7
2.1
0.7
0.2
0.1
0.0
3
2.1
1.6
0.7
0.2
0.1
0.0
0.0
4
0.6
0.4
0.2
0.1
0.0
0.0
0.0
5
0.2
0.1
0.1
0.0
0.0
0.0
0.0
6
0.0
0.0
0.0
0.0
0.0
0.0
0.0

Rows = PAR 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)
0.83
PAR
0.87
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.