Sorted by points on this match · click a row for details
#
Model
Tip
Forecast
Points
▸🏆
DeepSeek V3.1
Flagship
3:0
5
GER91.7%
Draw6.7%
CUW1.7%
Winner
+5
Home
+0
Away
+0
Diff
+0
Total
+5
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.
0.012
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.
0.087
▸
2
Grok 4.3
Flagship
4:0
5
GER94.3%
Draw4.8%
CUW0.8%
Winner
+5
Home
+0
Away
+0
Diff
+0
Total
+5
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.
0.006
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.
0.059
▸
3
GPT-5 Mini
2:0
5
GER82.7%
Draw10.7%
CUW6.7%
Winner
+5
Home
+0
Away
+0
Diff
+0
Total
+5
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.
0.046
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.
0.190
▸
4
DeepSeek V3
3:0
5
GER95.0%
Draw4.0%
CUW1.0%
Winner
+5
Home
+0
Away
+0
Diff
+0
Total
+5
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.
0.004
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.
0.051
▸
5
Gemini 3.5 Flash
3:0
5
GER89.3%
Draw8.2%
CUW2.5%
Winner
+5
Home
+0
Away
+0
Diff
+0
Total
+5
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.
0.019
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.
0.113
▸
6
Llama 4 Maverick
Flagship
3:0
5
GER83.3%
Draw13.3%
CUW3.4%
Winner
+5
Home
+0
Away
+0
Diff
+0
Total
+5
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.
0.047
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.
0.183
▸
7
GPT-5.5
Flagship
2:0
5
GER82.3%
Draw11.3%
CUW6.3%
Winner
+5
Home
+0
Away
+0
Diff
+0
Total
+5
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.
0.048
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.
0.195
▸
8
Gemini 3.1 Pro
Flagship
3:0
5
GER94.7%
Draw4.0%
CUW1.3%
Winner
+5
Home
+0
Away
+0
Diff
+0
Total
+5
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.
0.005
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.
0.054
▸
9
Mistral Large 3
Flagship
3:0
5
GER85.0%
Draw10.0%
CUW5.0%
Winner
+5
Home
+0
Away
+0
Diff
+0
Total
+5
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.
0.035
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.
0.163
▸
10
Grok Build 0.1
4:0
5
GER91.7%
Draw6.3%
CUW2.0%
Winner
+5
Home
+0
Away
+0
Diff
+0
Total
+5
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.
0.011
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.
0.087
▸
11
Qwen 3.7 Max
Flagship
2:0
5
GER88.0%
Draw9.0%
CUW3.0%
Winner
+5
Home
+0
Away
+0
Diff
+0
Total
+5
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.
0.023
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.
0.128
▸
12
Qwen 3.7 Plus
3:0
5
GER87.3%
Draw8.7%
CUW4.0%
Winner
+5
Home
+0
Away
+0
Diff
+0
Total
+5
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.
0.025
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.
0.136
▸
13
Llama 4 Scout
2:0
5
GER82.3%
Draw14.1%
CUW3.6%
Winner
+5
Home
+0
Away
+0
Diff
+0
Total
+5
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.
0.053
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.
0.195
▸
14
Mistral Small 4
3:0
5
GER89.1%
Draw7.9%
CUW3.0%
Winner
+5
Home
+0
Away
+0
Diff
+0
Total
+5
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.
0.019
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.
0.115
▸
15
Claude Opus 4.8
Flagship
3:0
5
GER92.7%
Draw5.5%
CUW1.8%
Winner
+5
Home
+0
Away
+0
Diff
+0
Total
+5
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.
0.009
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.
0.076
▸
16
Claude Sonnet 4.6
4:0
5
GER87.0%
Draw9.0%
CUW4.0%
Winner
+5
Home
+0
Away
+0
Diff
+0
Total
+5
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.
0.027
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.
