Dashboard·Overview
Current Round

Update Data

This workbook is the single source of truth for the entire site — Player Search, Disposal Predictor, Confidence-Based Multi, Positive EV Finder, Best Bets, Confidence Scores, Kelly Stake, fixtures, and every chart all read from whatever is uploaded here. Nothing is hardcoded: upload a new workbook each week and every page updates automatically, with no code changes required.

If you enter the password here, this upload is published to everyone who visits the site — not just this device. Leave it blank to save this upload only on this browser, the way it worked before.

Sync from Google Sheets Beta

Pulls the current version of your Google Sheet directly — no need to download and re-upload a file each time you update it. The sheet needs to stay shared as "Anyone with the link can view" for this to work.

Syncs automatically every hour in every visitor's browser, plus once shortly after the page loads — so the whole site stays current with whatever's in the sheet without anyone needing to click anything. Editing the sheet mid-hour is safe most of the time, but if a sync happens to land while you're mid-edit, visitors could briefly see a half-updated state — worth keeping in mind if you're making a big structural change, rather than a routine data update.

NBA Data Beta

Separate upload for the NBA side of the site — its own workbook, its own data, entirely independent of the AFL upload above.

Sync NBA Data from Google Sheets Beta

Pulls the current version of your NBA Google Sheet directly — no need to re-upload a file each time you update it. The sheet needs to stay shared as "Anyone with the link can view" for this to work.

Syncs automatically every hour, plus once shortly after the page loads — so the NBA side of the site stays current with whatever's in the sheet without anyone needing to click anything.

Live AFL Disposal Odds Beta

Live AFL disposal player-prop odds from The Odds API (Sportsbet, Ladbrokes, TAB, PointsBet, Betr), merged into the odds already loaded — it doesn't replace your Excel upload, it fills in live prices on top of it. Multiple lines per player (e.g. 24.5, 29.5, 34.5) are all kept, not collapsed to one.

Refreshes automatically every Thursday at 9:00 AM AEST/AEDT in the background — a scheduled server-side job fetches once a week regardless of how many people are on the site, so every visitor reads the same cached prices instead of each browser hitting the odds API on its own.

Compare NBA Players Beta

Search two NBA players to compare season form, recent form, and consistency across all five tracked stats.

VS
Search two players above — the comparison updates automatically once both are selected.

NBA Confidence Scan Beta

EdgeMetrics scans your tracked NBA players and identifies the strongest confidence-based opportunities for a chosen stat. There's no priced NBA odds sheet yet, so this ranks by model confidence, not EV — enter your own odds per leg in My Multi once you're ready to see combined price and EV.

Step 1 · Select Matches
Step 2 · Stat & Minimum Confidence
Step 3 · Multi Size

Available Confidence Picks

My Multi

NBA Hot Stats Beta

Finds players who've cleared the same line in every one of their last games straight — a genuine streak, not a model guess.

Stat
Streak Length
Minimum Line

NBA Board

Manually add a player, stat, line, and odds — since there's no priced NBA odds sheet yet, you enter the odds yourself, same as a manual add on the AFL side.

NBA Team Dashboard Beta

Select a team to see their defensive profile (how many Points/Rebounds/Assists/3PM they concede relative to the league, straight from your DefenceRank sheet) and their tracked roster.

Select a team above.

NBA Performance Tracker Beta

Test how EdgeMetrics performs on the NBA side using simulated bets. Tracked completely separately from the AFL ledger. Uses manually entered odds, since there's no priced NBA odds sheet yet.

Bankroll Settings

Stake per bet is calculated automatically from your starting bankroll and unit size, and is used as the default stake for Units Won / ROI below.

Performance Summary

Bet History

Add More Analytics

Model Performance — Overall

Breakdown by Market

Breakdown by Confidence

This tracker uses simulated bets only.
Past simulated performance does not guarantee future results.

Evaluate EdgeMetrics NBA Prediction Accuracy Beta

Track how accurately EdgeMetrics predicts NBA player stats over time

Walk-forward accuracy — every prediction shown here was made using only data available before that game, same no-lookahead methodology as the AFL side. Choose a stat to backtest.

Round Summary

Accuracy Trends

Mean Absolute Error (MAE) by round, last 20 rounds — tracks average prediction error. Lower is better. Points only — MAE is less meaningful on the narrower Rebounds/Assists/3PM ranges.

Correct prediction rate (actual met or exceeded the prediction) and within-±2 rate, last 20 rounds.

Player Line History

Search any NBA player to see every EdgeMetrics projection made for them this season for the selected stat, lined up against what they actually did.

Search for a player above to see their projection history.

Prediction History

Every EdgeMetrics prediction for the selected round and stat, checked against the actual result.

Accuracy by Confidence Level

Buckets every prediction in the selected scope by the model's confidence at the time, and checks how often the actual result met or exceeded the prediction.

NBA Dashboard Beta

Upcoming Fixtures
Fixtures not loaded yet
Top Confidence Pick
Upload the NBA workbook to see this
Highest Confidence (All Stats)
Best pick across Points/Rebounds/Assists/3PM
Best Multi
Combined model confidence
Model Accuracy (Points)
Walk-forward hit rate, met or exceeded

Top Confidence Picks

All stats
PlayerStatConfidencePredicted

Upcoming Fixtures

Schedule
DateHomeAwayVenue
No NBA fixtures loaded yet — this panel is ready for when fixtures data is added.

Model Performance — Accuracy by Round (Points)

Season
Correct rate Within ±2 points

Watchlist ⭐

Favourite players you're tracking week to week. Add them here or star them from Player Search — this page shows their current projection and best available value bet at a glance, no need to search each one individually.

My Watchlist

Confidence-Based Multi

EdgeMetrics scans available player disposal markets and identifies the strongest low-risk opportunities based on your selected confidence level. Lines are ranked using model probability, market price and expected value.

Step 1 · Select Matches
Step 2 · Minimum Confidence & Minimum EV
Step 3 · Multi Size

Available Multi Selections

My Multi

Best Bets

Scans every player and every priced line in your Player Odds sheet against the prediction model, all at once, and ranks every bet that clears both your EV bar and a minimum win confidence — so it's not surfacing +EV bets that only pay off one time in three.

Hot stats

Finds players who've cleared the same disposal count in every one of their last games straight — e.g. 23+ disposals in each of the last 5, no exceptions. That's a genuine streak, not a model guess. Where your Player Odds sheet has a price at that exact line, it shows the EV of actually taking it.

Board — ranked by expected value

Multi / parlay builder

Tick two or more legs from the board above to combine them into a multi — combined odds, combined model probability, and whether the parlay as a whole is still +EV, not just each leg on its own.

Performance Tracker

Track EdgeMetrics performance over time

Test how EdgeMetrics performs using simulated bets. Add selections from the analysis boards, record outcomes after games are completed, and track historical model performance. No real money involved.

Bankroll Settings

Stake per bet is calculated automatically from your starting bankroll and unit size, and is used as the default stake for Units Won / ROI below.

Performance Summary

Bet History

Add More Analytics

Model Performance — Overall

Breakdown by Category

This is the AFL ledger only — NBA has its own separate Performance Tracker.

Breakdown by Confidence

This tracker uses simulated bets only.
Past simulated performance does not guarantee future results.

Evaluate EdgeMetrics Prediction Accuracy

Track how accurately EdgeMetrics predicts AFL player disposals over time

EdgeMetrics evaluates predictions against actual match results to measure model performance, reliability and improvement.

Round Summary

Accuracy Trends

Mean Absolute Error (MAE) by round — tracks average prediction error over the season. Lower is better.

Correct prediction rate (actual met or exceeded the prediction) and within-±3 rate, by round. The dashed line marks the professional benchmark.

Advanced Prediction Models: professional algorithms and automated software generally target a hit rate of 53% to 55% on standard-juice bets (-110 odds, or $1.91 decimal) — considered the high-end ceiling for long-term profitability. That's the "pro benchmark" line on the chart above, not a promise this model reaches it.

Player Line History

Search any player to see every EdgeMetrics projection made for them this season, lined up against what they actually did — a running scorecard of how the model has tracked that specific player.

Search for a player above to see their projection history.

Prediction History

Every EdgeMetrics prediction for the selected round, checked against the actual result. (Historical bookmaker lines aren't retained, so this tracks raw prediction accuracy rather than betting outcomes — see Performance Tracker for that.)

Accuracy by Confidence Level

Does a tighter, more consistent player profile actually predict better? Buckets every prediction in the selected scope by the model's confidence at the time, and checks how often the actual result met or exceeded the prediction.

Watchlist
Pending Bets
Today's Matches
Best Value Today
Today's Matches
Fixtures loaded this round
Top Value (EV)
Scan Best Bets or add odds to see this
Highest Confidence
Confidence for a specific disposal line
Best Multi
Combined model probability
Model Accuracy
Season hit rate, met or exceeded line

Top Player Edges

This week
PlayerEdgeEVConfidenceQualityLine

Today's Matches

Today
DateHomeAwayVenue

Model Performance — Accuracy by Round

2026 season
Correct rate Within ±3 disposals Pro benchmark (55%)

Compare Players

Search two players to compare them head-to-head using the same prediction engine, confidence scoring, and EV calculations used everywhere else in EdgeMetrics — season form, matchup edge, consistency, and which one represents the stronger betting opportunity right now.

VS
Search two players above — the comparison updates automatically once both are selected.

Team Dashboard

Select a team to see their upcoming fixture, tracked roster, and which positions they concede the most — or fewest — disposals to, relative to the league average.

Select a team above to see their positional matchup profile.

How It Works

An explanation of how EdgeMetrics builds its projections, scores confidence, and calculates edge — so you know what's behind every number on the site.

Model vs Market

Player Projection Engine

Our model combines multiple performance signals to estimate expected disposals:

  • Season performance
  • Recent form
  • Last 5 game trend
  • Opponent history
  • Venue history

Then compares the projection against bookmaker markets to calculate:

  • Predicted disposals
  • Probability estimates
  • Expected Value (EV)
  • Kelly staking guidance
+18.4%
Estimated EV
Tom Green Confidence: High
31
Model projection
27.5
Sportsbook line
+3.5
Model edge
1.55
Example odds
Full methodology
How the prediction works: Predicted = Season Avg×0.25 + Last 10 Avg×0.20 + Last 5 Avg×0.30 + Vs-opponent avg×0.125 + Vs-venue avg×0.125, then adjusted for opponent strength if that's on, then −1 as a deliberate conservative shift, then rounded to the nearest whole disposal. If there's no history yet vs that opponent and/or at that venue, those weights are dropped and the rest are scaled back up proportionally — it never silently treats "no data" as zero. Spread (for the probability model) is the standard deviation of the player's last 12 games.

On the odds: alternative disposal lines (20.5, 25.5, 30.5, etc.) are each quoted on their own, not as a two-sided Over/Under market, so there's no matching price to de-vig against. Edge is model probability minus the bookmaker's raw implied probability (1/odds) — it still includes their margin, so real edge is a little smaller than shown. EV is per $1 staked; Kelly % is the full-Kelly stake many bettors scale down (¼–½ Kelly) to manage variance.

Hit rate this season is separate from the modelled probability — it's the plain, empirical count of how many of the player's actual games this season cleared that line, no distribution assumptions involved. When it and "model prob" disagree by a lot, it usually means recent form has shifted quickly, or a small sample (like a single past meeting with this opponent) is pulling the model in a direction the season-long record doesn't support.

The label under the EV number (AVOID / PASS / WEAK BET / GOOD BET / STRONG BET / ELITE BET) is a straight read of the EV percentage: below 0% is Avoid, up to 5% Pass, up to 10% Weak, up to 20% Good, up to 30% Strong, above that Elite. A positive EV number on its own isn't the same as a good bet — it's one estimate built on a limited sample.

Understanding Value

What does EV mean?

Expected Value (EV) estimates whether the available market price is favourable compared with the model's predicted probability.

A positive EV result means the model identifies a potential difference between its projection and the market price.

Important: Positive EV does not guarantee profit or successful outcomes. It represents a statistical estimate based on historical data and modelling assumptions.

How is it calculated?

EV per $1 staked = (Model Probability × (Decimal Odds − 1)) − (1 − Model Probability)

In plain terms: it's the average amount the model estimates you'd win or lose per dollar staked, if its probability estimate is accurate and the same bet were repeated many times over.

Worked example: the model gives a player a 60% chance of going over their line, and the market is offering $2.00 (decimal odds).

  • (0.60 × (2.00 − 1)) − (1 − 0.60)
  • = (0.60 × 1.00) − 0.40
  • = 0.60 − 0.40
  • = +0.20, or +20% EV

That means for every $1 staked, the model estimates an average return of 20 cents — assuming its probability estimate holds up across many similar bets. Any single bet still only wins or loses; EV describes the average across repeated bets, not the outcome of any one of them.

Why EdgeMetrics?

Built for data-driven sports analysis

  • No gut feeling
  • No hype-based tips
  • Transparent methodology
  • Historical performance tracking
  • Multi-sport expansion

Responsible Gambling

EdgeMetrics provides statistical analysis only. Predictions are estimates and do not guarantee betting outcomes. Gambling involves risk. Only bet what you can afford to lose.

Gambling Help Online  ·  BetStop  ·  Support Services

Support Services

If gambling is affecting you or someone you know, free and confidential help is available in Australia.

Gambling Help Online

24/7 counselling, online chat and self-help tools.
gamblinghelponline.org.au · 1800 858 858

BetStop

Australia's National Self-Exclusion Register — block yourself from all licensed wagering services in one place.
betstop.gov.au · 1800 858 858

Lifeline

24/7 crisis support and suicide prevention.
lifeline.org.au · 13 11 14

Privacy Policy

Effective Date: 15 July 2026

1. Introduction

EdgeMetrics™ ("we", "us", "our") respects your privacy and is committed to protecting your personal information. This Privacy Policy explains how we collect, use, store and disclose information when you access or use the EdgeMetrics sports analytics platform ("the App"). By using EdgeMetrics, you agree to the practices described in this Privacy Policy.

2. Information We Collect

The information we collect depends on how you use EdgeMetrics. We may collect:

Account Information. If you create an account, we may collect:

  • Name
  • Email address
  • Login information
  • Account preferences

Usage Information. We may collect information about how you interact with the App, including:

  • Pages viewed
  • Features used
  • Search history within the App
  • Saved players or preferences
  • Device information
  • Browser type
  • IP address
  • Usage statistics

Technical Information. We may collect technical information to:

  • Maintain security
  • Improve performance
  • Identify errors
  • Improve user experience

3. How We Use Your Information

We may use collected information to:

  • Provide access to EdgeMetrics services
  • Maintain and improve the platform
  • Personalise user experiences
  • Save user preferences
  • Monitor system performance
  • Respond to enquiries
  • Detect and prevent misuse of the App
  • Comply with legal obligations

4. Sports Data and Analytics

EdgeMetrics provides statistical analysis, player projections and predictive modelling based on sports data. The App may process information including:

  • Player statistics
  • Match data
  • Historical performance data
  • Sports-related information

This information is used to generate analytical insights and predictions.

5. Cookies and Analytics

EdgeMetrics may use cookies and similar technologies to improve functionality and understand user behaviour. These technologies may collect information such as:

  • Website activity
  • Session duration
  • Device information
  • General usage patterns

You may disable cookies through your browser settings, however some features may not function correctly.

6. Third-Party Services

EdgeMetrics may use third-party services to support platform functionality, including:

  • Hosting providers
  • Analytics providers
  • Data providers
  • Authentication services

These providers may process information according to their own privacy policies. EdgeMetrics does not sell personal information to third parties.

7. Sharing of Information

We may disclose information where necessary:

  • To provide requested services
  • To trusted service providers assisting with operation of the App
  • To comply with legal obligations
  • To protect the rights, safety and security of EdgeMetrics and its users

8. Data Security

We take reasonable steps to protect personal information from:

  • Unauthorised access
  • Loss
  • Misuse
  • Disclosure

However, no online service can guarantee complete security.

9. Data Retention

We retain personal information only for as long as necessary to:

  • Provide services
  • Maintain records
  • Meet legal requirements
  • Resolve disputes

When information is no longer required, reasonable steps will be taken to securely delete or anonymise it.

10. Your Rights

Depending on applicable privacy laws, you may request:

  • Access to personal information we hold about you
  • Correction of inaccurate information
  • Deletion of your account or personal information
  • Information about how your data is handled

Requests can be made using the contact details below.

11. Children and Age Restrictions

EdgeMetrics is intended for users aged 18 years and older. We do not knowingly collect personal information from individuals under 18 years of age.

12. Responsible Gambling

EdgeMetrics provides statistical analysis and sports insights only. Predictions, projections and statistical models do not guarantee betting outcomes or financial returns. If gambling is causing harm, support is available through the Support Services section of the App, including BetStop and other responsible gambling organisations.

13. External Links

EdgeMetrics may provide links to external websites, including:

  • Responsible gambling services
  • Sports information providers
  • Third-party resources

We are not responsible for the privacy practices, security or content of external websites.

14. Changes to This Privacy Policy

We may update this Privacy Policy from time to time. Any changes will be published within the App or website.

Terms of Use

Effective Date: 15 July 2026

1. Acceptance of Terms

By accessing or using this App, you agree to these Terms of Use. If you do not agree with these terms, you should discontinue use of the App.

2. Eligibility

This App is intended for users aged 18 years or older. By using the App, you confirm that you are at least 18 years of age.

3. Purpose of the App

The App provides AFL statistical analysis, player projections and betting-related information based on historical data and predictive modelling. All predictions are estimates only.

4. No Guarantee

The App does not guarantee:

  • Prediction accuracy
  • Betting success
  • Profit
  • Positive returns

Past performance does not guarantee future results. Users are solely responsible for any betting decisions they make.

5. Responsible Gambling

Users should gamble responsibly. If gambling is causing harm, assistance is available through BetStop and the support services listed within the App.

6. Limitation of Liability

To the maximum extent permitted by law, we are not liable for:

  • Financial losses
  • Betting losses
  • Lost profits
  • Incorrect predictions
  • Data inaccuracies
  • Service interruptions

Your use of the App is entirely at your own risk.

7. Intellectual Property

All content, including prediction models, software, graphics, logos and written material, remains our intellectual property unless otherwise stated. Users may not copy, distribute or reproduce any part of the App without permission.

8. Third-Party Information

The App may display information obtained from third-party providers, including sporting statistics and bookmaker odds. We do not warrant the accuracy, completeness or availability of third-party information.

9. Availability

We may modify, suspend or discontinue any part of the App at any time without notice. We do not guarantee uninterrupted access.

10. User Conduct

Users agree not to:

  • Attempt to interfere with the operation of the App
  • Use automated methods to access the App without permission
  • Misuse or exploit the prediction models
  • Engage in unlawful activity while using the App

11. Amendments

We may amend these Terms of Use at any time. Continued use of the App constitutes acceptance of any updated terms.

12. Governing Law

These Terms are governed by the laws of Australia. Any disputes will be subject to the jurisdiction of the relevant Australian courts.