AFL
Performance

How our tipping heuristics are performing.

Most Recent Past Season 2025

Final results for the complete season (not year-to-date)

Boosted Tip

Final 2025
Final Profit $0.00
Rounds Played0
Tips Graded0 (0 correct)
Odds Coverage0%

Weighted Tip

Final 2025
Final Profit $-535.00
Accuracy39.4%
Rounds Played29
Tips Graded216 (85 correct)
Odds Coverage0%

Current Season 2026

29 / 30 rounds completed

How profit is calculated: each tip carries a simulated $10 stake. Games with bookmaker odds (The Odds API) settle at the real decimal price for the tipped team; games without an odds snapshot settle at a representative $1.90; drawn games refund the stake. Odds coverage shows the share of tips settled at real prices.

Boosted Tip

Current Season
Year-to-Date Profit $0.00
Accuracy—
Rounds Played0
Avg Profit/Round $0.00
Odds Coverage0%

Weighted Tip

Current Season
Year-to-Date Profit $757.00
Projected Annual Profit $757.00
Projections are based on early season performance and may change as the season progresses.
Accuracy70.2%
Rounds Played30
Avg Profit/Round $25.23
Odds Coverage0%
Individual Model Accuracy
Form★
67.4%
Accuracy
$643
Profit
Elo Rating
67.0%
Accuracy
$624
Profit
Matchup
64.7%
Accuracy
$529
Profit
Value
61.0%
Accuracy
$377
Profit
Weather Impact
60.1%
Accuracy
$339
Profit
Home Advantage
59.2%
Accuracy
$301
Profit
Player Form
58.7%
Accuracy
$282
Profit
Injury Impact
58.3%
Accuracy
$263
Profit

Weighted Tip Model

v69
Trained: Tue, 10 Nov, 08:00 amTraining rows: 650Intercept: -20.36R²: 0.2507MAE: 27.5

The Weighted Tip model combines all 8 ML model predictions using a linear regression trained on 650 historical games. Each model contributes a margin weight (influence on score margin) and a confidence weight (influence on confidence). The current model (v69) has R² = 0.251 with intercept -20.36. Most influential models: Player Form, Elo Rating, Matchup. The model is retrained weekly with updated coefficients.

Model Equation
predicted_margin =-20.36−2.281·Player Formm+0.769·Elo Ratingm+0.348·Matchupm+0.211·Formm−0.034·Weather Impactm−0.023·Valuem+0.007·Injury Impactm−0.006·Home Advantagem

Each model contributes a margin weight (× its predicted margin toward home) and a confidence weight. Models with larger absolute weights have more influence on the final tip.

Margin Weights

Active Boosted Model (XGBoost)

v55
Trained: Tue, 10 Nov, 08:00 amTraining rows: 650R²: 0.8760MAE: 11.0SHAP base value: 5.81

The Boosted Tip model is a gradient-boosted tree ensemble (XGBoost) trained on 650 historical games to predict the home-team-signed margin from the same model predictions the Weighted Tip model uses. The current model (v55) reaches R² = 0.876 with MAE = 11.0 points and is retrained weekly. Feature importances below are mean |SHAP| values — how much each model's margin and confidence signal moves the prediction, with Elo Rating, Matchup, Home Advantage the most influential.

SHAP Feature Importance