Eredivisie Round 23 AI Model Accuracy: Devstral Small Leads
Devstral Small (OpenRouter) led Eredivisie predictions this week with 3.43 points per match, followed by Llama 3.1 8B (OpenRouter) at 3.40 and Llama 3.3 70B Instruct (OpenRouter) at 3.25. Models achieved 36.86% correct tendency overall, though FC Volendam's 2-1 win over PSV Eindhoven caught most models off guard.
Devstral Small (OpenRouter) led Eredivisie predictions this week with 3.43 points per match, followed by Llama 3.1 8B (OpenRouter) at 3.40 and Llama 3.3 70B Instruct (OpenRouter) at 3.25. Models achieved 36.86% correct tendency overall, though FC Volendam's 2-1 win over PSV Eindhoven caught most models off guard. Eredivisie Regular Season - 23 featured 8 matches, including high-stakes fixtures where AI prediction accuracy is crucial for assessing model reliability in competitive scenarios. This analysis provides a statistical audit of model performance based on provided data.
Top 10 Models
| # | Model | Matches | Total Points | Avg Pts/Match | Tendency % | Exact % |
|---|---|---|---|---|---|---|
| 1 | Devstral Small (OpenRouter) | 7 | 24 | 3.43 | 71.4% | 28.6% |
| 2 | Llama 3.1 8B (OpenRouter) | 5 | 17 | 3.40 | 80.0% | 20.0% |
| 3 | Llama 3.3 70B Instruct (OpenRouter) | 4 | 13 | 3.25 | 50.0% | 25.0% |
| 4 | Phi-4 (OpenRouter) | 7 | 22 | 3.14 | 57.1% | 28.6% |
| 5 | Llama 4 Scout (OpenRouter) | 8 | 24 | 3.00 | 62.5% | 25.0% |
| 6 | Mistral Small 3 24B (OpenRouter) | 5 | 15 | 3.00 | 40.0% | 40.0% |
| 7 | GPT-OSS 20B (OpenRouter) | 8 | 22 | 2.75 | 50.0% | 25.0% |
| 8 | Qwen 2.5 7B (OpenRouter) | 5 | 13 | 2.60 | 60.0% | 20.0% |
| 9 | RNJ-1 Instruct (OpenRouter) | 5 | 13 | 2.60 | 60.0% | 20.0% |
| 10 | Llama 3.2 3B (OpenRouter) | 5 | 12 | 2.40 | 40.0% | 20.0% |
Match-by-Match Audit
- Telstar vs Twente (1-1): 24 models, 12.5% correct tendency, 8.3% exact score hits. Predicted outcomes: 1 home win, 3 draws, 20 away wins. Consensus: away win (83.3%), incorrect.
- Heerenveen vs PEC Zwolle (4-2): 23 models, 34.8% correct tendency, 0.0% exact score hits. Predicted outcomes: 8 home wins, 12 draws, 3 away wins. Consensus: draw (52.2%), incorrect.
- Feyenoord vs GO Ahead Eagles (1-0): 21 models, 66.7% correct tendency, 0.0% exact score hits. Predicted outcomes: 14 home wins, 5 draws, 2 away wins. Consensus: home win (66.7%), correct.
- Groningen vs Utrecht (1-2): 25 models, 28.0% correct tendency, 20.0% exact score hits. Predicted outcomes: 4 home wins, 14 draws, 7 away wins. Consensus: draw (56.0%), incorrect.
- Ajax vs Fortuna Sittard (4-1): 26 models, 65.4% correct tendency, 0.0% exact score hits. Predicted outcomes: 17 home wins, 9 draws, 0 away wins. Consensus: home win (65.4%), correct.
- Excelsior vs AZ Alkmaar (1-2): 23 models, 52.2% correct tendency, 47.8% exact score hits. Predicted outcomes: 1 home win, 10 draws, 12 away wins. Consensus: away win (52.2%), correct.
- Heracles vs NAC Breda (0-1): 25 models, 32.0% correct tendency, 4.0% exact score hits. Predicted outcomes: 2 home wins, 15 draws, 8 away wins. Consensus: draw (60.0%), incorrect.
- FC Volendam vs PSV Eindhoven (2-1): 30 models, 3.3% correct tendency, 3.3% exact score hits. Predicted outcomes: 1 home win, 2 draws, 27 away wins. Consensus: away win (90.0%), incorrect.
Biggest Consensus Misses
- FC Volendam vs PSV Eindhoven (2-1): Consensus: away win (90.0%), counts: 1 home win, 2 draws, 27 away wins
- Telstar vs Twente (1-1): Consensus: away win (83.3%), counts: 1 home win, 3 draws, 20 away wins
- Heracles vs NAC Breda (0-1): Consensus: draw (60.0%), counts: 2 home wins, 15 draws, 8 away wins
- Groningen vs Utrecht (1-2): Consensus: draw (56.0%), counts: 4 home wins, 14 draws, 7 away wins
- Heerenveen vs PEC Zwolle (4-2): Consensus: draw (52.2%), counts: 8 home wins, 12 draws, 3 away wins
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 Eredivisie Regular Season - 23? A: Devstral Small (OpenRouter) performed best with an average of 3.43 points per match across 7 matches.
Q: How accurate were AI predictions for Eredivisie this round? A: Models achieved 36.86% correct tendency and 10.44% exact score hit rate across 8 matches.
Q: What was the biggest upset in Eredivisie Regular Season - 23? A: FC Volendam's 2-1 win over PSV Eindhoven, where only 3.3% of models predicted the correct tendency.
Q: How does kroam.xyz score AI football predictions? A: Using a quota-based system awarding 2-6 points for correct tendency, +1 for correct goal difference, and +3 for exact score, with a maximum of 10 points per prediction.
Generation cost: $0.0022
Tokens: 4,612 input + 2,050 output
Frequently Asked Questions
What is this article about?
Which AI model performed best in Eredivisie Regular Season - 23?**?
Q: Which AI model performed best in Eredivisie Regular Season - 23?
Q: How accurate were AI predictions for Eredivisie this round?
You might also like
Eredivisie Week 24 AI Model Audit: Llama 4 Scout Leads
Llama 4 Scout (OpenRouter) performed best with 4.00 avg points/match, followed by Trinity Large Preview (3.22) and Llama 3.3 70B Instruct (3.00). Overall accuracy was 52.05% correct tendency. The biggest upset was Twente's 2-1 win over Groningen, which fooled 84.2% of models predicting a draw.
Feb 23, 2026
Eredivisie AI Predictions Audit - Regular Season 22
Gemma 3n E4B led with 3.11 avg points/match, followed by Mistral 7B v0.3 and Kimi K2 Thinking (3.00). Models achieved 50.62% correct tendency overall. Twente vs Heerenveen (5-0) was a major upset, with only 12% correct tendency.
Feb 9, 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.