AFL Engine · Live

2026 Season
+17.6% Edge16 seasons backtested★Proven★above a 50/50 baseline
Introductory offer
36 FREE match predictions, every month.
Free for a limited time: earlybird access
4 rounds aheadLocked pre-bounceEmailed to you freeCredit Card NOT RequiredCancel anytime
Audited Baseline Accuracy
✓ 67.58% ✓
3,134 games · 16 seasons · live now
Live Rolling Coming soon Updated to the bounce
  • Late team changes & key outs
  • Weather & venue conditions
  • Re-calculated as the news breaks
$8/mo upgrade. The sharpest read, right up to first bounce.
Not pundit picks. Not bookmaker odds. Historical data and calibrated probability, with a published, reproducible track record.

Don’t miss out on this limited time offer!!!

What you get

Calibrated probability. honestly measured, openly proven.

01
Every game, called early

36 games free every month: the next 4 rounds, with winner, probability and margin for each.

02
A number you can check

67.58% across 3,134 games. You see the full track record, not a highlight reel, so you know exactly what you are getting.

03
The edge as it sharpens

Upgrade to live rolling and your picks update with late team changes and conditions, right up to first bounce.

The future of enhanced prediction technology

All predictions, models, data and content are the property of Intelligence Edge Pty. Ltd. The audited accuracy figure and all prediction outputs are generated by our own statistical engine from publicly available match data; every figure stated is true and independently reproducible. Past performance does not guarantee future results. Predictions are provided for information only and are not betting or financial advice. Intelligence Edge is an independent analytics provider, not affiliated with, endorsed by, or associated with the Australian Football League (AFL), its clubs, or any official body. “AFL” refers to the competition only, to describe the matches our engine analyses. All team and competition names remain the property of their respective owners. © 2026 Intelligence Edge Pty. Ltd. All rights reserved.

Platform Architecture

Intelligence Engine
Multi-factor architecture
built for scale

Intelligence Edge is not a single model. It is a multi-layer prediction platform combining eight independent component families — each contributing a distinct analytical dimension to the final output.

Per-fixture outputs include predicted winner, calibrated win probability, projected scores and margin, and a HIGH / MEDIUM / LOW confidence band.

Live data: Squiggle API · Footy Tables · FootyWire · OpenWeatherMap

Sequential ELO
Era-aware team strength updated every game
Schedule Quality
SRS margin and opponent strength adjustments
Venue & Travel
Per-venue home advantage and travel penalties
H2H History
Quality-weighted head-to-head records
Form & Momentum
Short-window net rating and hot-streak detection
Roster Context
Injury-adjusted roster weight and finals pressure
Live Weather
Rain and wind score-compression via live forecast
Bayesian Calibration
Platt-style probability calibration with confidence bands
✦67.58% Verified Walk-Forward Accuracy ✦Zero Look-Ahead Bias — Confirmed ✦SHA-256 Hash Published — Verify Yourself ✦3,261 Games Backtested · 2010–2025 ✦16 Consecutive Seasons Above 60% ✦Live API · Sub-Second Latency ✦B2B Licensing Available Now ✦Full Audit Package Under NDA ✦Intelligence Engine · Live ✦6/6 Independent Audit Checks Pass ✦+17.6pp Above Coin-Flip Baseline ✦Beats Bookmaker Accuracy by 11pp ✦21 Data Integrity Checks · All Pass ✦p < 0.001 · Statistically Significant ✦Bayesian Platt Calibration · Confidence Bands ✦Best Season 2016 · 73.43% ✦Immutable Predictions · Locked Before Kick-Off ✦White-Label · API · Data Feed · Escrow ✦67.58% Verified Walk-Forward Accuracy ✦Zero Look-Ahead Bias — Confirmed ✦SHA-256 Hash Published — Verify Yourself ✦3,261 Games Backtested · 2010–2025 ✦16 Consecutive Seasons Above 60% ✦Live API · Sub-Second Latency ✦B2B Licensing Available Now ✦Full Audit Package Under NDA ✦Intelligence Engine · Live ✦6/6 Independent Audit Checks Pass ✦+17.6pp Above Coin-Flip Baseline ✦Beats Bookmaker Accuracy by 11pp ✦21 Data Integrity Checks · All Pass ✦p < 0.001 · Statistically Significant ✦Bayesian Platt Calibration · Confidence Bands ✦Best Season 2016 · 73.43% ✦Immutable Predictions · Locked Before Kick-Off ✦White-Label · API · Data Feed · Escrow
Independently Verified · Walk-Forward Audited

The number you can verify yourself

67.58%
Walk-Forward Accuracy

Every prediction locked before kick-off using only historically available data. No hindsight. No retroactive changes. True out-of-sample results across 16 complete seasons.

6/6
Audit Checks Pass

Six independent structural checks confirm zero look-ahead bias. Predictions in chronological order. Timestamps before game dates. No future-season data leaked. All pass.

SHA-256
Hash Published

A published SHA-256 hash lets any party run the full backtest on their own infrastructure and verify the identical result independently. The number never changes.

+11pp
Beats the Bookmaker

Bookmaker-implied accuracy sits at 54–56%. Intelligence Edge exceeds it by 11+ percentage points — sustained across every single season without exception.

Floor
67.58% Is Our Minimum

Two engine components disabled for the public audit. The live production engine performs higher. We only publish the number we can prove.

16/16
Seasons Above 60%

Above 60% every single year from 2010 to 2025 — including COVID, rule changes and expansion teams. Worst season 2019: 61.84%, still +11.8pp above baseline.

Security · IP Protection · Independent Audit

Six audit checks. All pass.

— PASS
Predictions appear in strict chronological order
— PASS
Every prediction timestamp precedes its game date
— PASS
Data-availability cutoff precedes every game date
— PASS
No game predicted more than once
— PASS
No future-season information leaked into any prediction
— PASS
Deterministic prediction stream — SHA-256 hash published

Data integrity: 21 PASS / 0 WARNING / 0 FAIL

Where We Are Heading

One engine.
Every sport.

The Intelligence Engine architecture was designed from day one to be sport-agnostic. The same multi-factor methodology applies across any sport with sufficient historical data.

The NBA engine is 88% complete and will be released to the same verification standard as AFL. EPL follows at 35%. Every engine we release will carry a published SHA-256 hash and pass the same six-stage structural audit.

The goal is a multi-sport B2B prediction platform — licensed by operators, broadcasters and analysts who demand verified data, not marketing claims.

SportStatusProgressVerification StandardAvailable
FootyLive67.58% verifiedWalk-forward · SHA-256 published · 6/6 audit checksNow
NBAIn Development
88%
Same verification standard as footy engine2026
EPLIn Development
35%
Same verification standard as footy engine2026
NRLResearch Stage
0%
Same verification standard as footy engineTBC

INTELLIGENCE EDGE
Pty Ltd · Registered Australia 2026
intelligenceedge.com.au

B2B · White-Label · API

License the engine that called 67.58% of the AFL right.

Same engine. Your brand. AFL live now. NBA, EPL, and NRL platform expansion underway.

Request a partnership conversation
67.58%Audited accuracyWalk-forward, full season
3,134GamesIndependently verifiable
16Seasons2010 – 2025

Public audit live at /verify · Updated weekly

Built for integration

Four ways to deploy. Pick the one that fits your stack.

API Access

Build on top

Raw predictions delivered via REST API. JSON endpoints for matches, predictions, ratings, and results. Use your own UI, your own brand, your own analytics layer. Bearer JWT auth.

  • GET /predictions/round/{n}
  • GET /teams/elo/current
  • GET /predictions/historical

Data Feed

Stream into your platform

Structured JSON prediction feed delivered directly into your platform. Built for operators, broadcasters, and aggregators. Sub-second latency, per-client rate limits, scoped endpoints.

White-Label App

Deploy as-is

Your brand, our engine. iOS, Android, Web. The six screens you've seen above, themed to your colours. Full audit trail included. We run the backend; you own the user relationship.

  • Custom branding (logo, colours, typography)
  • Push notifications + reminders
  • Auth + payment integration

Source Escrow

Long-term enterprise

Engine held in third-party escrow for business-continuity guarantees. Air-gapped deployment options available for enterprise data-residency requirements. Long-term enterprise licensing.

All routes include SLA, dedicated support, NDA package on request, methodology documentation, and audit cooperation.

Available under NDA

Concrete deliverables, not promises

Full technical documentation — exact model weights and calibration fit
Complete per-prediction CSV — every game ID, timestamp, confidence, result
Live API pilot access with per-client key and 30-day evaluation window
Backtest re-run on your own infrastructure to verify SHA-256 independently

Production-grade across every surface

Full mobile platform, ready to white-label. Built on FastAPI + AWS Fargate, integrated with our prediction engine via REST API.

Sign in screen
Sign in · Engineered to be verified
Home hub
Home · Live engine + navigation
Today's predictions
Today's Edge · Live predictions
Future fixtures
Future · Round-by-round forecasts
Results audit
Results · Audit trail (engine vs reality)
Engine profile
Profile · Engine spec (ELO + 14 factors)

In-app accuracy badges show 2026-season-to-date performance. Full audited engine accuracy: 67.58% across 3,134 walk-forward predictions (2010–2025) — see /verify.

Built for serious operators

AFL Clubs

Performance & analytics teams. Pre-game opponent intelligence, ELO tracking, injury-adjusted ratings, walk-forward audit history.

Sports Media

Embed audited predictions in match coverage. Drive engagement on your platform with our engine. Confidence-banded picks, head-to-head splits, methodology footnotes.

Betting Operators

Independent risk benchmarks. Pre-game edge signals, audit-grade predictions, cross-reference our public verify page for credibility.

Other Sports Orgs

NBA, EPL, NRL, Cricket. Multisport platform expansion underway. Same engine architecture. Same audit philosophy.

Process

01

Discovery call

30 min, NDA optional

02

Sample data + integration spec

We share methodology, you share use case

03

Commercial terms + onboarding

Typical: 2–4 weeks

04

Go live with audit cooperation

Ongoing support, transparent reporting

"Send one email. NDA within 24 hours. Full audit package within 48."

Ready to talk?

Request a partnership conversation

Or email Chris directly: chris.argent@intelligenceedge.com.au

Intelligence Edge logo

Intelligence Engine.
Independently verified. Available for licensing.

Current model · AFL Engine · Live

67.58%Verified Accuracy
+17.6ppAbove Baseline
16/16Seasons Above 60%
6/6Audit Checks Pass
Ready to Verify

We don't ask you to
take our word for it.

Full audit package available under mutual NDA. Run the backtest on your own infrastructure. Verify the SHA-256 hash independently. Call the live API yourself.

Request a Demo Request NDA
CompanyIntelligence Edge Pty Ltd
DemoUnder NDA · 48hrs
Our Story

Built on proof,
not promises.

Intelligence Edge was founded with one objective — to build the most rigorously verified sport prediction engine in Australia. Not a tipping service. Not a model estimate. A true and correct record of predictions made before kick-off, with zero hindsight, independently auditable by anyone.

The Intelligence Engine was developed over several years, trained on 3,261 games across 16 AFL seasons. Every decision was made with one question in mind: can we prove it?

The answer is 67.58%. Walk-forward verified. SHA-256 published. Six audit checks. All pass.

We only publish the number we can prove.

Six Commitments

Statements we stand behind

Run the backtest yourself. The SHA-256 hash never changes.

I published the SHA-256 hash before anyone asked me to. If the number was going to mean anything, it had to be verifiable by someone other than me. Run it yourself — the result will be identical.

67.58% is our floor. Two components disabled for the audit. The live engine runs higher.

I made the decision to disable two modules for the public audit. They're trained on historical data that overlaps the test window — leaving them active would have been dishonest. The floor is real. The ceiling is higher.

We only publish the number we can prove.

There were higher numbers I could have published. I chose not to. Every figure on this site comes with a methodology you can inspect and a hash you can verify. If I can't prove it, it doesn't go on the page.

Not a script. A deployed, versioned, fully audited prediction platform.

I've seen a lot of 'AI tipsters' that are just a spreadsheet with a logo. This is a production API running on AWS Fargate, version-controlled on GitHub, with a six-stage audit trail. The difference matters.

Six independent checks. Zero future data. All pass.

The six checks weren't designed to pass — they were designed to catch me out. Chronological order, timestamp integrity, no data leakage, no duplicates. I ran them hoping they'd find nothing. They found nothing.

Above 60% every season for 16 years — including COVID, rule changes and expansion teams.

2020 was the hardest year to model — hub games, no crowds, condensed schedule. 2019 brought rule changes. GWS and Gold Coast joined mid-dataset. The engine stayed above 60% through all of it. That consistency is what I'm most proud of.

VERIFIEDPREDICTIONS

Every prediction the engine makes, locked before bounce and posted publicly on X before each round — then the full result and margin error logged after the game. The public posts are the timestamp. The receipts speak for themselves.

—

Predictions played

—

Winners correct

—

2026 accuracy

—

Avg margin error

Public record from Round 8 (start of timestamped X posting). Separate from the audited 16-season baseline of 67.58%.

Round Match Venue Pre-bounce proof Pick Pred margin Actual margin Δ Result

Want the full audit?

SHA-256 hash and reproduction package available under NDA for licensing partners and serious enquiries. Ten years of walk-forward predictions, methodology, and integrity-check results.

Request audit package

Methodology

HOW THIS TABLE IS POPULATED

1 · LOCK PRE-BOUNCE

Every prediction is generated and written to the production database before the round begins. Once written, predictions are immutable — the database enforces idempotency at the writer level.

2 · PUBLISH TIMESTAMPED

The round's PDF is generated from the database state, with macOS Quartz auto-stamping the file's creation date. The PDF is then posted publicly to X with a third-party server timestamp.

3 · MATCH AGAINST RESULT

After full time, results are entered against each prediction. The engine never sees the result; the result is only joined to the prediction record after the fact for public reporting.

4 · AUDIT LOGGED

Each row is timestamped, versioned, and cross-referenced against the public X post and the daily S3 audit snapshot. Anyone with the NDA package can verify the entire chain end-to-end.