How Alphr's NRL Predictions Work

The 4-Model XGBoost Ensemble

Alphr's NRL prediction system is not a single model — it is a four-model ensemble where each component is optimised for a different task. The outputs are combined into the head-to-head, margin, and over/under tips published on the NRL hub each round.

1. NRL Win Predictor — a binary classifier returning the home-team win probability. This drives every head-to-head tip.
2. NRL Margin Predictor — a regression model estimating the winning margin in points. Feeds the margin-band picks (1–12, 13+, Draw).
3. NRL Dual-Score Predictor — two regression models forecasting home and away scores independently. Used for total-points (over/under) tips.
4. Edge Filter — only a tip with ≥7% positive edge over the bookmaker's implied probability is published. This single rule explains why Alphr's tips outperform a naive "tip every match" approach.

~160 NRL Features per Match

Each match is described by roughly 160 engineered features grouped into seven families:

• Team ELO ratings — custom NRL ELO with margin-adjusted K and season regression.
• Positional player ELO — separate ratings for forwards, backs, and spine players (halfback, five-eighth, hooker, fullback).
• Rolling form — last 5/10 game wins, points for/against, average margin.
• Venue & travel — venue home-win rate, interstate travel penalties, rest days.
• Weather — temperature, rain, wind speed and conditions at venue.
• Referee bias — historical referee tendencies on penalty counts and home-team favourability.
• Market signals — live Sportsbet head-to-head, line and totals odds at lock-in time.

Training, Validation & Anti-Leakage

Alphr uses a strict temporal train / validation / test split: training on more than a decade of historical seasons, validation on a held-out season (for hyperparameter tuning and early stopping), blind out-of-sample testing on the most recent completed season. Every feature uses anti-leakage window functions — only data available before kick-off is allowed. This means the historical accuracy you see is a real estimate of forward-looking skill, not a backtested overfit.

Why the Edge Filter Matters

Predicting an NRL winner is easier than beating the bookmaker. Sportsbet's opening line is already a sharp prediction. To make money from NRL tips you need an edge — your model probability must exceed the implied probability built into the odds. Alphr requires ≥7% positive edge before publishing a tip. See the NRL best bets page for the highest-edge selections each round.

2026 & 2025 Performance Snapshot

In the 2025 NRL season Alphr's head-to-head model hit a 61.5% strike rate (131 / 213), the margin-band model hit 60.4% with a +14.8% ROI on flat $100 stakes, and Round 19 was a perfect 7-from-7 round. Live 2026 NRL results update automatically each weekend.