Bundesliga Round 22 AI Model Accuracy: DeepSeek R1 Leads
DeepSeek R1-0528 topped Bundesliga predictions with 3.00 points per match, followed by Llama 3.2 3B (2.71) and Trinity Large Preview (2.00). Models achieved 57.17% correct tendency overall. Hamburger SV's 3-2 win over Union Berlin was the biggest upset.
DeepSeek R1-0528 topped Bundesliga predictions with 3.00 points per match, followed by Llama 3.2 3B (2.71) and Trinity Large Preview (2.00). Models achieved 57.17% correct tendency overall. Hamburger SV's 3-2 win over Union Berlin was the biggest upset.
Bundesliga Regular Season - 22 featured 9 matches involving all 18 allowed teams. Prediction accuracy remains critical as models handle varied fixture difficulties. This audit examines statistical performance across all matches.
Top 10 Models
| # | Model | Matches | Total Points | Avg Pts/Match | Tendency % | Exact % |
|---|---|---|---|---|---|---|
| 1 | DeepSeek R1-0528 (OpenRouter) | 3 | 9 | 3.00 | 66.7% | 0.0% |
| 2 | Llama 3.2 3B (OpenRouter) | 7 | 19 | 2.71 | 85.7% | 0.0% |
| 3 | Trinity Large Preview (OpenRouter) | 8 | 16 | 2.00 | 75.0% | 0.0% |
| 4 | RNJ-1 Instruct (OpenRouter) | 7 | 14 | 2.00 | 71.4% | 14.3% |
| 5 | GPT-OSS 20B (OpenRouter) | 8 | 15 | 1.88 | 75.0% | 0.0% |
| 6 | Gemma 3 12B (OpenRouter) | 8 | 15 | 1.88 | 62.5% | 12.5% |
| 7 | Qwen3 30B A3B (OpenRouter) | 7 | 13 | 1.86 | 71.4% | 0.0% |
| 8 | Llama 3.1 8B (OpenRouter) | 7 | 13 | 1.86 | 71.4% | 0.0% |
| 9 | DeepSeek R1 (OpenRouter) | 7 | 12 | 1.71 | 57.1% | 14.3% |
| 10 | Qwen3 235B Thinking (OpenRouter) | 8 | 13 | 1.63 | 25.0% | 12.5% |
Match-by-Match Audit
- RB Leipzig vs VfL Wolfsburg (2-2): 8.3% correct tendency, 0.0% exact hits. Consensus: H (70.8%) incorrect.
- FC Augsburg vs 1. FC Heidenheim (1-0): 91.7% correct tendency, 0.0% exact hits. Consensus: H (91.7%) correct.
- VfB Stuttgart vs 1. FC Köln (3-1): 75.0% correct tendency, 8.3% exact hits. Consensus: H (75.0%) correct.
- Bayer Leverkusen vs FC St. Pauli (4-0): 87.5% correct tendency, 0.0% exact hits. Consensus: H (87.5%) correct.
- Eintracht Frankfurt vs Borussia Mönchengladbach (3-0): 20.0% correct tendency, 0.0% exact hits. Consensus: D (64.0%) incorrect.
- Hamburger SV vs Union Berlin (3-2): 4.2% correct tendency, 0.0% exact hits. Consensus: D (79.2%) incorrect.
- 1899 Hoffenheim vs SC Freiburg (3-0): 68.0% correct tendency, 0.0% exact hits. Consensus: H (68.0%) correct.
- Werder Bremen vs Bayern München (0-3): 92.0% correct tendency, 24.0% exact hits. Consensus: A (92.0%) correct.
- Borussia Dortmund vs FSV Mainz 05 (4-0): 67.9% correct tendency, 0.0% exact hits. Consensus: H (67.9%) correct.
Biggest Consensus Misses
- Hamburger SV vs Union Berlin (3-2): Consensus: D (79.2%) | Counts H/D/A: 1/19/4
- RB Leipzig vs VfL Wolfsburg (2-2): Consensus: H (70.8%) | Counts H/D/A: 17/2/5
- Eintracht Frankfurt vs Borussia Mönchengladbach (3-0): Consensus: D (64.0%) | Counts H/D/A: 5/16/4
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 - 22? A: DeepSeek R1-0528 (OpenRouter) achieved the highest average points per match (3.00) among models with at least 3 matches.
Q: How accurate were AI predictions for Bundesliga this round? A: Models achieved 57.17% correct tendency and 3.59% exact score hit rate across 223 predictions.
Q: What was the biggest upset in Bundesliga Regular Season - 22? A: Hamburger SV's 3-2 win over Union Berlin, where only 4.2% of models predicted the correct tendency.
Q: How does kroam.xyz score AI football predictions? A: Using a quota-based system awarding 2-6 tendency points, +1 goal difference bonus, and +3 exact score bonus (max 10 points per prediction).
Generation cost: $0.0021
Tokens: 4,928 input + 1,765 output
Frequently Asked Questions
What is this article about?
Which AI model performed best in Bundesliga Regular Season - 22?**?
Q: Which AI model performed best in Bundesliga Regular Season - 22?
Q: How accurate were AI predictions for Bundesliga this round?
You might also like
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 Europa League Round of 32 AI Model Performance Audit
GLM-5 (OpenRouter) led UEFA Europa League predictions this week with 3.25 points per match, followed by Llama 4 Scout (OpenRouter) at 2.88 and Mistral Small 3.2 24B (OpenRouter) at 2.25. Models achieved 52.63% correct tendency overall, though Ludogorets vs Ferencvarosi TC (2-1) caught most models off guard.
Feb 23, 2026
UEFA Conference League Round of 32 AI Prediction Audit
Trinity Large Preview led with 3.13 points per match, followed by Phi-4 (2.38) and Kimi K2.5 (2.13). Models achieved 33.19% correct tendency overall, with FC Noah's 1-0 win over AZ Alkmaar being the biggest surprise.