Bundesliga Round 21 AI Model Performance: Top Models & Upsets
Mistral 7B v0.3 led Bundesliga predictions with 3.75 avg points/match, followed by Llama 3.2 3B Turbo (3.38) and DeepSeek V3.1 (3.14). Models achieved 43.08% correct tendency. FC St. Pauli's 2-1 win over VfB Stuttgart was the biggest consensus miss, with only 8.0% accuracy.
Mistral 7B v0.3 led Bundesliga predictions with 3.75 avg points/match, followed by Llama 3.2 3B Turbo (3.38) and DeepSeek V3.1 (3.14). Models achieved 43.08% correct tendency. FC St. Pauli's 2-1 win over VfB Stuttgart was the biggest consensus miss, with only 8.0% accuracy.
Bundesliga Regular Season - 21 featured 8 matches, including fixtures like Bayern München vs 1899 Hoffenheim and Borussia Mönchengladbach vs Bayer Leverkusen. AI prediction accuracy is critical for assessing model reliability in competitive matchups. This audit analyzes statistical performance across all predictions.
Top 10 Models
| # | Model | Matches | Total Points | Avg Pts/Match | Tendency % | Exact % |
|---|---|---|---|---|---|---|
| 1 | Mistral 7B v0.3 (Mistral) | 8 | 30 | 3.75 | 50.0% | 50.0% |
| 2 | Llama 3.2 3B Turbo (Meta) | 8 | 27 | 3.38 | 62.5% | 25.0% |
| 3 | DeepSeek V3.1 | 7 | 22 | 3.14 | 57.1% | 42.9% |
| 4 | Nemotron Nano 9B v2 (NVIDIA) | 7 | 20 | 2.86 | 71.4% | 28.6% |
| 5 | Qwen3 235B Instruct (Alibaba) | 5 | 14 | 2.80 | 60.0% | 40.0% |
| 6 | DeepSeek R1 (Reasoning) | 8 | 19 | 2.38 | 50.0% | 25.0% |
| 7 | Qwen 2.5 7B Turbo (Alibaba) | 8 | 18 | 2.25 | 62.5% | 12.5% |
| 8 | Llama 3 8B Lite (Meta) | 8 | 18 | 2.25 | 62.5% | 12.5% |
| 9 | Marin 8B Instruct (Marin Community) | 8 | 18 | 2.25 | 62.5% | 12.5% |
| 10 | MiniMax M2 (Synthetic) | 8 | 18 | 2.25 | 50.0% | 25.0% |
Match-by-Match Audit
- Bayern München vs 1899 Hoffenheim (5-1): 63.0% correct tendency, 0.0% exact score hits. Consensus: H (63.0%), correct.
- 1. FC Köln vs RB Leipzig (1-2): 69.2% correct tendency, 57.7% exact score hits. Consensus: A (69.2%), correct.
- Borussia Mönchengladbach vs Bayer Leverkusen (1-1): 23.1% correct tendency, 23.1% exact score hits. Consensus: A (69.2%), incorrect.
- FC St. Pauli vs VfB Stuttgart (2-1): 8.0% correct tendency, 8.0% exact score hits. Consensus: A (76.0%), incorrect.
- FSV Mainz 05 vs FC Augsburg (2-0): 8.3% correct tendency, 0.0% exact score hits. Consensus: A (50.0%), incorrect.
- VfL Wolfsburg vs Borussia Dortmund (1-2): 89.3% correct tendency, 35.7% exact score hits. Consensus: A (89.3%), correct.
- SC Freiburg vs Werder Bremen (1-0): 61.5% correct tendency, 3.8% exact score hits. Consensus: H (61.5%), correct.
- 1. FC Heidenheim vs Hamburger SV (0-2): 22.2% correct tendency, 0.0% exact score hits. Consensus: D (70.4%), incorrect.
Biggest Consensus Misses
- FC St. Pauli vs VfB Stuttgart (2-1) | Consensus: A (76.0%) | Counts H/D/A: 2/4/19
-
- FC Heidenheim vs Hamburger SV (0-2) | Consensus: D (70.4%) | Counts H/D/A: 2/19/6
- Borussia Mönchengladbach vs Bayer Leverkusen (1-1) | Consensus: A (69.2%) | Counts H/D/A: 2/6/18
- FSV Mainz 05 vs FC Augsburg (2-0) | Consensus: A (50.0%) | Counts H/D/A: 2/10/12
Methodology
kroam.xyz uses a quota-based scoring system that rewards both accuracy and boldness:
Tendency Points (2-6 points): Models earn points for correctly predicting the match outcome (home win, draw, or away win). The points awarded depend on prediction rarity—if most models predicted a home win but the away team won, models who correctly predicted the away win earn more points (up to 6). Common predictions earn fewer points (minimum 2).
Goal Difference Bonus (+1 point): If the model predicts the correct goal difference (e.g., predicted 2-1 and result was 3-2, both +1 difference), they earn a bonus point.
Exact Score Bonus (+3 points): Predicting the exact final score earns 3 additional points.
Maximum: 10 points per prediction (6 tendency + 1 goal diff + 3 exact).
This system ensures that models taking calculated risks on unlikely outcomes are rewarded when correct, while also recognizing precision in exact score predictions. Learn more about our methodology.
Frequently Asked Questions
Q: Which AI model performed best in Bundesliga Regular Season - 21? A: Mistral 7B v0.3 had the highest average points per match (3.75) among models with at least 3 matches.
Q: How accurate were AI predictions for Bundesliga this round? A: The average correct tendency was 43.08%, and the exact score hit rate was 16.04% across 209 predictions.
Q: What was the biggest upset in Bundesliga Regular Season - 21? A: FC St. Pauli's 2-1 win over VfB Stuttgart, where only 8.0% of models predicted the correct tendency.
Q: How does kroam.xyz score AI football predictions? A: Using a quota-based system awarding up to 10 points per prediction: 2-6 for correct tendency, +1 for goal difference, and +3 for exact score.
Generation cost: $0.0060
Tokens: 4,692 input + 1,888 output
Frequently Asked Questions
What is this article about?
Which AI model performed best in Bundesliga Regular Season - 21?**?
Q: Which AI model performed best in Bundesliga Regular Season - 21?
Q: How accurate were AI predictions for Bundesliga this round?
You might also like
Bundesliga AI Model Performance Audit - Regular Season 24
MiniMax M2.5 led Bundesliga predictions with 2.89 points per match, followed by GPT-OSS 20B (2.33) and Llama 4 Scout (2.11). Models achieved 34.24% correct tendency overall, with the 1899 Hoffenheim vs FC St. Pauli upset (0-1) catching 94.7% consensus predictions wrong.
Mar 2, 2026
Bundesliga Round 23 AI Model Performance Audit
Llama 3.3 70B Instruct led Bundesliga predictions with 3.13 points per match, followed by MiniMax M2.1 (2.50) and GLM-5 (2.25). Models achieved 32.75% correct tendency overall, though the 1. FC Heidenheim vs VfB Stuttgart 3-3 draw caught most models off guard.
Feb 23, 2026
UEFA Conference League Round of 32 AI Prediction Audit
GPT-OSS 20B led UEFA Conference League predictions with 2.88 points per match, followed by Trinity Large Preview (2.63) and GLM-5 (2.25). Models achieved 38.16% correct tendency overall, with Fiorentina vs Jagiellonia (2-4) as the biggest upset.