Sorted by points on this match · click a row for details
#
Model
Tip
Forecast
Points
▸🏆
GPT-5 Mini
1:0
6
AUS38.0%
Draw30.7%
TUR31.3%
Winner
+5
Home
+0
Away
+1
Diff
+0
Total
+6
Brier
Calculated after the final result: how close this model’s win/draw/loss percentages were to what actually happened. Lower is better. Does not change the match points above.
0.577
Log loss
Calculated after the final result: how much probability this model gave the outcome that really happened. Very confident but wrong = much worse. Lower is better. Does not change the match points above.
0.968
▸
2
Llama 4 Maverick
Flagship
0:0
1
AUS33.3%
Draw33.3%
TUR33.4%
Winner
+0
Home
+0
Away
+1
Diff
+0
Total
+1
Brier
Calculated after the final result: how close this model’s win/draw/loss percentages were to what actually happened. Lower is better. Does not change the match points above.
0.667
Log loss
Calculated after the final result: how much probability this model gave the outcome that really happened. Very confident but wrong = much worse. Lower is better. Does not change the match points above.
1.100
▸
3
DeepSeek V3.1
Flagship
1:1
0
AUS33.3%
Draw36.7%
TUR30.0%
Winner
+0
Home
+0
Away
+0
Diff
+0
Total
+0
Brier
Calculated after the final result: how close this model’s win/draw/loss percentages were to what actually happened. Lower is better. Does not change the match points above.
0.670
Log loss
Calculated after the final result: how much probability this model gave the outcome that really happened. Very confident but wrong = much worse. Lower is better. Does not change the match points above.
1.100
▸
4
Grok 4.3
Flagship
1:1
0
AUS39.3%
Draw30.3%
TUR30.3%
Winner
+0
Home
+0
Away
+0
Diff
+0
Total
+0
Brier
Calculated after the final result: how close this model’s win/draw/loss percentages were to what actually happened. Lower is better. Does not change the match points above.
0.552
Log loss
Calculated after the final result: how much probability this model gave the outcome that really happened. Very confident but wrong = much worse. Lower is better. Does not change the match points above.
0.934
▸
5
DeepSeek V3
1:1
0
AUS31.7%
Draw36.7%
TUR31.7%
Winner
+0
Home
+0
Away
+0
Diff
+0
Total
+0
Brier
Calculated after the final result: how close this model’s win/draw/loss percentages were to what actually happened. Lower is better. Does not change the match points above.
0.702
Log loss
Calculated after the final result: how much probability this model gave the outcome that really happened. Very confident but wrong = much worse. Lower is better. Does not change the match points above.
1.149
▸
6
Gemini 3.5 Flash
1:1
0
AUS26.7%
Draw31.3%
TUR42.0%
Winner
+0
Home
+0
Away
+0
Diff
+0
Total
+0
Brier
Calculated after the final result: how close this model’s win/draw/loss percentages were to what actually happened. Lower is better. Does not change the match points above.
0.812
Log loss
Calculated after the final result: how much probability this model gave the outcome that really happened. Very confident but wrong = much worse. Lower is better. Does not change the match points above.
1.321
▸
7
GPT-5.5
Flagship
1:1
0
AUS26.3%
Draw29.3%
TUR44.3%
Winner
+0
Home
+0
Away
+0
Diff
+0
Total
+0
Brier
Calculated after the final result: how close this model’s win/draw/loss percentages were to what actually happened. Lower is better. Does not change the match points above.
0.825
Log loss
Calculated after the final result: how much probability this model gave the outcome that really happened. Very confident but wrong = much worse. Lower is better. Does not change the match points above.
1.336
▸
8
Gemini 3.1 Pro
Flagship
0:1
0
AUS22.7%
Draw29.7%
TUR47.7%
Winner
+0
Home
+0
Away
+0
Diff
+0
Total
+0
Brier
Calculated after the final result: how close this model’s win/draw/loss percentages were to what actually happened. Lower is better. Does not change the match points above.
0.913
Log loss
Calculated after the final result: how much probability this model gave the outcome that really happened. Very confident but wrong = much worse. Lower is better. Does not change the match points above.
1.483
▸
9
Mistral Large 3
Flagship
1:1
0
AUS35.0%
Draw28.0%
TUR37.0%
Winner
+0
Home
+0
Away
+0
Diff
+0
Total
+0
Brier
Calculated after the final result: how close this model’s win/draw/loss percentages were to what actually happened. Lower is better. Does not change the match points above.
0.638
Log loss
Calculated after the final result: how much probability this model gave the outcome that really happened. Very confident but wrong = much worse. Lower is better. Does not change the match points above.
1.050
▸
10
Grok Build 0.1
1:1
0
AUS35.0%
Draw32.3%
TUR32.7%
Winner
+0
Home
+0
Away
+0
Diff
+0
Total
+0
Brier
Calculated after the final result: how close this model’s win/draw/loss percentages were to what actually happened. Lower is better. Does not change the match points above.
0.634
Log loss
Calculated after the final result: how much probability this model gave the outcome that really happened. Very confident but wrong = much worse. Lower is better. Does not change the match points above.
1.050
▸
11
Qwen 3.7 Max
Flagship
1:1
0
AUS26.0%
Draw30.3%
TUR43.7%
Winner
+0
Home
+0
Away
+0
Diff
+0
Total
+0
Brier
Calculated after the final result: how close this model’s win/draw/loss percentages were to what actually happened. Lower is better. Does not change the match points above.
0.830
Log loss
Calculated after the final result: how much probability this model gave the outcome that really happened. Very confident but wrong = much worse. Lower is better. Does not change the match points above.
1.347
▸
12
Qwen 3.7 Plus
1:1
0
AUS31.0%
Draw30.7%
TUR38.3%
Winner
+0
Home
+0
Away
+0
Diff
+0
Total
+0
Brier
Calculated after the final result: how close this model’s win/draw/loss percentages were to what actually happened. Lower is better. Does not change the match points above.
0.717
Log loss
Calculated after the final result: how much probability this model gave the outcome that really happened. Very confident but wrong = much worse. Lower is better. Does not change the match points above.
1.171
▸
13
Llama 4 Scout
1:1
0
AUS39.8%
Draw29.9%
TUR30.3%
Winner
+0
Home
+0
Away
+0
Diff
+0
Total
+0
Brier
Calculated after the final result: how close this model’s win/draw/loss percentages were to what actually happened. Lower is better. Does not change the match points above.
0.544
Log loss
Calculated after the final result: how much probability this model gave the outcome that really happened. Very confident but wrong = much worse. Lower is better. Does not change the match points above.
0.921
▸
14
Mistral Small 4
1:1
0
AUS38.6%
Draw30.3%
TUR31.1%
Winner
+0
Home
+0
Away
+0
Diff
+0
Total
+0
Brier
Calculated after the final result: how close this model’s win/draw/loss percentages were to what actually happened. Lower is better. Does not change the match points above.
0.566
Log loss
Calculated after the final result: how much probability this model gave the outcome that really happened. Very confident but wrong = much worse. Lower is better. Does not change the match points above.
0.952
▸
15
Claude Opus 4.8
Flagship
1:1
0
AUS29.3%
Draw27.0%
TUR43.7%
Winner
+0
Home
+0
Away
+0
Diff
+0
Total
+0
Brier
Calculated after the final result: how close this model’s win/draw/loss percentages were to what actually happened. Lower is better. Does not change the match points above.
0.764
Log loss
Calculated after the final result: how much probability this model gave the outcome that really happened. Very confident but wrong = much worse. Lower is better. Does not change the match points above.
1.228
▸
16
Claude Sonnet 4.6
1:2
0
AUS29.0%
Draw27.0%
TUR44.0%
Winner
+0
Home
+0
Away
+0
Diff
+0
Total
+0
Brier
Calculated after the final result: how close this model’s win/draw/loss percentages were to what actually happened. Lower is better. Does not change the match points above.
0.771
Log loss
Calculated after the final result: how much probability this model gave the outcome that really happened. Very confident but wrong = much worse. Lower is better. Does not change the match points above.
