Turkish Super Lig Week 22 AI Model Prediction Audit
Kimi K2.5 led Turkish Super Lig predictions this week with 4.00 points per match, followed by Llama 3.2 3B (3.20) and Trinity Large Preview (2.83). Models achieved 33.53% correct tendency overall, though Göztepe's 0-0 draw with Kayserispor caught most models off guard.
Kimi K2.5 led Turkish Super Lig predictions this week with 4.00 points per match, followed by Llama 3.2 3B (3.20) and Trinity Large Preview (2.83). Models achieved 33.53% correct tendency overall, though Göztepe's 0-0 draw with Kayserispor caught most models off guard.
Turkish Super Lig Regular Season - 22 featured 8 matches including high-profile fixtures like Trabzonspor vs Fenerbahçe and Başakşehir vs Beşiktaş. AI prediction accuracy remains critical for evaluating model consistency amid competitive fixtures. This audit analyzes statistical performance across all predictions.
Top 10 Models
| # | Model | Matches | Total Points | Avg Pts/Match | Tendency % | Exact % |
|---|---|---|---|---|---|---|
| 1 | Kimi K2.5 (OpenRouter) | 5 | 20 | 4.00 | 80.0% | 0.0% |
| 2 | Llama 3.2 3B (OpenRouter) | 5 | 16 | 3.20 | 60.0% | 20.0% |
| 3 | Trinity Large Preview (OpenRouter) | 6 | 17 | 2.83 | 50.0% | 16.7% |
| 4 | Qwen 2.5 7B (OpenRouter) | 5 | 13 | 2.60 | 60.0% | 20.0% |
| 5 | Phi-4 (OpenRouter) | 6 | 11 | 1.83 | 33.3% | 16.7% |
| 6 | DeepSeek R1-0528 (OpenRouter) | 5 | 9 | 1.80 | 40.0% | 0.0% |
| 7 | Ministral 3 14B (OpenRouter) | 5 | 9 | 1.80 | 40.0% | 20.0% |
| 8 | Cogito 671B (OpenRouter) | 7 | 12 | 1.71 | 42.9% | 0.0% |
| 9 | RNJ-1 Instruct (OpenRouter) | 5 | 8 | 1.60 | 40.0% | 0.0% |
| 10 | DeepSeek V3.1 (OpenRouter) | 5 | 8 | 1.60 | 60.0% | 0.0% |
Match-by-Match Audit
Başakşehir vs Beşiktaş (2-3)
- Total models: 22
- Correct tendency: 18.2% (4/22)
- Exact score hits: 0.0% (0/22)
- Consensus: D (72.7%) | correct: no
Göztepe vs Kayserispor (0-0)
- Total models: 24
- Correct tendency: 4.2% (1/24)
- Exact score hits: 0.0% (0/24)
- Consensus: H (87.5%) | correct: no
Kocaelispor vs Gazişehir Gaziantep (3-0)
- Total models: 21
- Correct tendency: 4.8% (1/21)
- Exact score hits: 0.0% (0/21)
- Consensus: D (76.2%) | correct: no
Trabzonspor vs Fenerbahçe (2-3)
- Total models: 24
- Correct tendency: 58.3% (14/24)
- Exact score hits: 8.3% (2/24)
- Consensus: A (58.3%) | correct: yes
Alanyaspor vs Konyaspor (2-1)
- Total models: 24
- Correct tendency: 20.8% (5/24)
- Exact score hits: 16.7% (4/24)
- Consensus: D (66.7%) | correct: no
Genclerbirligi vs Rizespor (2-2)
- Total models: 23
- Correct tendency: 60.9% (14/23)
- Exact score hits: 0.0% (0/23)
- Consensus: D (60.9%) | correct: yes
Antalyaspor vs Samsunspor (3-1)
- Total models: 27
- Correct tendency: 11.1% (3/27)
- Exact score hits: 0.0% (0/27)
- Consensus: A (55.6%) | correct: no
Galatasaray vs Eyüpspor (5-1)
- Total models: 30
- Correct tendency: 90.0% (27/30)
- Exact score hits: 0.0% (0/30)
- Consensus: H (90.0%) | correct: yes
Biggest Consensus Misses
- Göztepe vs Kayserispor (0-0) | Consensus: H (87.5%) | Counts H/D/A: 21/1/2
- Kocaelispor vs Gazişehir Gaziantep (3-0) | Consensus: D (76.2%) | Counts H/D/A: 1/16/4
- Başakşehir vs Beşiktaş (2-3) | Consensus: D (72.7%) | Counts H/D/A: 2/16/4
- Alanyaspor vs Konyaspor (2-1) | Consensus: D (66.7%) | Counts H/D/A: 5/16/3
- Antalyaspor vs Samsunspor (3-1) | Consensus: A (55.6%) | Counts H/D/A: 3/9/15
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 Turkish Super Lig Regular Season - 22? A: Kimi K2.5 (OpenRouter) performed best with 4.00 average points per match across 5 matches.
Q: How accurate were AI predictions for Turkish Super Lig this round? A: Models achieved 33.53% correct tendency and 3.12% exact score hit rate across 195 total predictions.
Q: What was the biggest upset in Turkish Super Lig Regular Season - 22? A: Göztepe vs Kayserispor (0-0) was the biggest upset, with 87.5% consensus incorrectly predicting a home win.
Q: How does kroam.xyz score AI football predictions? A: The system awards 2-6 points for correct tendency, +1 point for correct goal difference, and +3 points for exact score predictions, with a maximum of 10 points per match.
Generation cost: $0.0023
Tokens: 4,670 input + 2,075 output
Frequently Asked Questions
What is this article about?
Which AI model performed best in Turkish Super Lig Regular Season - 22?**?
Q: Which AI model performed best in Turkish Super Lig Regular Season - 22?
Q: How accurate were AI predictions for Turkish Super Lig this round?
You might also like
Turkish Super Lig Week 23 AI Predictions Audit & Accuracy Stats
Kimi K2.5 led AI predictions with 2.63 avg points/match, followed by GLM-4.7 and Qwen3 30B A3B at 2.25. Models achieved 26.32% correct tendency. Biggest upset: Konyaspor's 2-0 win over Galatasaray fooled 84.2% consensus.
Feb 23, 2026
Turkish Super Lig Week 21 AI Prediction Audit & Stats
Llama 3.2 3B Turbo led with 3.67 avg points/match, followed by Llama 3.1 8B Turbo (2.50) and Ministral 3 14B (2.17). Models achieved 44.88% correct tendency overall, with the Konyaspor vs Göztepe 0-0 draw as the biggest surprise.
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.