What is a value bet in AFL?
A value bet is a price that pays more than the true chance of the result justifies. It has nothing to do with backing winners — favourites, underdogs, anyone can be value. It's the gap between two numbers: what a result is really worth, and what the bookmaker is paying for it.
The one equation that matters
Every set of decimal odds implies a probability: implied % = 100 ÷ odds. A team at $2.00 is priced as a 50% chance; $4.00 implies 25%; $1.25 implies 80%.
A bet has value when:
A worked AFL example
Say the Bulldogs are $2.10 to beat the Swans. The implied probability is 100 ÷ 2.10 ≈ 47.6%. Suppose a statistical model — one that weighs form, home ground, injuries via team ratings — makes the Dogs a 56% chance.
| Bulldogs | |
|---|---|
| Bookmaker price | $2.10 |
| Market implied probability | 47.6% |
| Model probability | 56% |
| Edge | +8.4 points |
If the model is right, $10 on this spot 100 times returns about $10 × 2.10 × 56 = $1,176 for $1,000 staked — roughly an 18% profit in the long run. Any single Saturday? The Dogs still lose 44 times out of 100. Value is a repeated-game concept, not a promise about this week.
Why the edge must beat the bookmaker's margin
Bookmakers don't price markets to 100% — they price to roughly 104–106% so that they profit whichever side wins (the "overround" or vig). That means a small model edge of one or two points usually isn't value at all; it's inside the margin's noise. Serious value hunting first removes the vig (rescales both sides so they sum to 100%) and then compares model vs market. That's exactly what FootyAI does before it calls anything a value pick — and why most games have none.
Where value comes from in footy
- Public sentiment — big clubs and flag favourites get over-bet; unfashionable sides drift past their true price.
- Slow reactions — markets can lag late team news, a venue change, or a form turnaround the ratings have already caught.
- Blowout bias — markets tend to be more extreme than models on lopsided games, which is why rank outsiders sometimes carry a mathematical (if unappealing) edge.
- Weather and venue quirks — wet-weather footy compresses margins; some sides travel badly. Models that account for this find spots the headline price misses.
What value betting is not
- Not a lock. A 56% chance loses nearly half the time. Edges pay off over seasons, not rounds.
- Not about picking winners. Backing a $1.20 favourite that should be $1.10 is value; backing a $1.20 favourite that should be $1.35 is a losing bet that usually wins.
- Not a guarantee anyone profits. Even genuine edges are small, and variance is brutal. Bet for entertainment, within limits you set in advance — see our responsible betting guide.
See the edge before you tip
FootyAI compares a statistical model's probabilities against de-vigged bookmaker consensus for every AFL game — and shows model %, market % and the edge on each pick.
🏉 Get FootyAI