Bookmakers do not choose prices arbitrarily. They combine model probabilities, margin policy, and risk management objectives to produce opening and live-adjusted odds. Understanding the process turns the betting market from a mysterious black box into a predictable pipeline you can target. This guide breaks down the steps a typical Australian operator follows from opening price to settlement, and the structural differences between sharp and square books that determine where edge lives. Read it alongside positive EV betting and devigging for the full pricing picture.
The bookmaker pricing process
Most books follow three stages:
- Generate opening fair probabilities from internal models, market references (typically Pinnacle and the Asian markets), and recent form data.
- Apply margin to convert fair probabilities into priced odds. Margin is distributed across outcomes according to the book's policy (sometimes evenly, sometimes loading favourites or longshots).
- Adjust in response to money flow and new information. This includes net liability, sharp account activity, public action, injuries, weather, and competitor line moves.
Most Australian books do not originate their own opening prices. They copy Pinnacle and the Asian markets with 30-120 second lag. Only the largest operators (TAB, Sportsbet, bet365) have meaningful in-house pricing capacity, and even they default to Pinnacle for markets outside their core competence.
Opening lines and where they come from
Globally, Pinnacle and Circa Sports (Las Vegas) are recognised as the originators of sharp opening prices. Their lines are built from internal models and refined by accepting large bets from sharp accounts in the opening minutes. The act of taking those bets is the market discovery process — Pinnacle does not penalise sharp action; they use it to set the market price.
Asian books (SBOBet, IBC, ISN) drive most soccer and basketball pricing globally. They offer the highest limits and the lowest vig in those sports. Australian books reference Asian prices as their second sharp benchmark.
After opening, every Australian operator runs their own response: some books match Pinnacle's moves within seconds, others lag minutes to hours. Books with weak in-house modelling capacity lag the longest, which is where exploitable mispricing windows live.
Model inputs and components
Typical inputs to a bookmaker pricing model:
- Power ratings. Each team or player gets a rating updated after each match. Rating differences map directly to spread/H2H probabilities.
- Recent form weighting. Last 5-10 matches weighted higher than season average.
- Player availability. Injuries, suspensions, late lineup changes. Star-player out can move a spread 3-5 points in NBA.
- Home/away effects. Quantified separately by sport (NBA ~3 points, NFL ~2.5 points, EPL ~0.4 goals).
- Weather. Wind, rain, temperature for outdoor sports.
- Pace and style. Possessions/minute, scoring rate, defensive efficiency.
- Competitor-market references. Pinnacle line, Asian market line, peer Australian book lines.
- Customer mix. Some books skew lines slightly against their typical customer base (e.g. shading public favourites).
Modelling quality varies dramatically by sport. NFL and NBA are heavily modelled. Lower-tier cricket and rugby competitions are minimally modelled, often using simple form-based shortcuts.
How margin is constructed
Margin is the excess implied probability over 100%. On a two-way market with both sides at $1.91, implied probabilities sum to 1.047 — a 4.7% margin. The book holds that as expected profit per stake.
Distribution of margin across outcomes is not always even:
- Favourite-loaded margin. More margin on the favourite price than the underdog. Common in head-to-head markets on heavy favourites where the book wants to discourage favourite action.
- Long-shot-loaded margin. More margin on longshots. Common in outright markets with large fields (Brownlow, premiership) where public action skews toward favourites and books need protection on long-shot upset scenarios.
- Even margin. Sharp books like Pinnacle distribute margin evenly, making both sides equally fair-relative.
This matters because long-shot-heavy margin structures systematically shift fair value opportunities toward specific sides of a market. If you devig a market and find the underdog underpriced relative to your model, that may simply be the book's margin skew — not a true edge.
What moves lines
Lines move from three drivers:
- Net money flow. The book's liability across outcomes. If one side attracts disproportionate stakes, the price shortens to discourage further action.
- Sharp account activity. Bets from accounts the book has flagged as informed. Books move further on a $1k bet from a known sharp than on a $10k bet from a known recreational.
- Genuine information updates. Injuries, lineup news, weather changes, referee assignments.
The skill is reading which driver caused a given move. A 2-point spread move at 9am Monday is probably sharp action. A 2-point move at 6pm on game day after a star is ruled out is information. A 1-point move with no news is usually money flow correction.
Sharp books vs square books
Sharp books (Pinnacle, Circa, certain Asian operators) build their business model on taking large bets from informed customers. They use sharp action as price discovery and do not restrict winning accounts. Their margins are thinner and their lines are recognised market benchmarks.
Square books (most Australian operators, US retail books) build their business on recreational customers. They restrict winning accounts to protect retail margins, run higher vig, and follow Pinnacle's lead on pricing. They are happy to be wrong on lines as long as they make money on volume.
The implication for bettors: sharp books are your benchmark, square books are your execution venue. You compare prices across square books to capture dispersion, but you measure your own edge by comparing your bets to the sharp closing line.
Why books price differently
- Model assumptions. Each book's model emphasises different inputs.
- Customer mix. A book with mostly recreational customers shades lines toward public favourites.
- Exposure / liability. Books with one-sided action shorten the popular side.
- Business strategy. Promotional books take loss leaders on featured markets to attract sign-ups.
- Update lag. Some books update within seconds of a Pinnacle move; others lag minutes.
Different books can post materially different prices on the same event at the same time. On marquee events you might see 5-15 cent differences on H2H prices and half-point differences on spreads. That dispersion is the bettor's structural edge.
Risk management and liability balancing
Books are not target-balanced on every bet — they accept calculated risk. They move lines when one side's stake load creates uncomfortable variance, but for routine bets they simply book the action.
Sharp money triggers larger moves than equivalent square money. Books have profiled customer types and weight bets accordingly. A $5k bet from a known sharp moves more line than $50k from a known public bettor.
Books also manage tail risk via stake caps. They limit large bets on certain markets and accounts to cap downside. This is why winning accounts get restricted — the book is not managing your bet; it is managing the variance you contribute to their overall book.
Closing line efficiency
By kickoff, lines have absorbed all public information and sharp action. The closing line is widely treated as the market's best estimate of true probability. Beating the closing line consistently is the single most reliable indicator of long-term edge.
Beating closing line by 2-3% over hundreds of bets is sufficient to demonstrate genuine edge. The mechanism: even if your individual bets lose more than expected in the short run, consistently beating the close means your probability estimates are sharper than the market's final consensus.
See closing line value explainedfor the full methodology.
Implications for bettors
- Compare aggressively across books to capture best available price.
- Track CLV as a process scorecard against market closing efficiency.
- Diversify execution across accounts to reduce restriction risk.
- Prioritise markets where bookmaker model depth is lower (props, lower-tier competitions, derivatives).
- Use sharp books (Pinnacle/Asian) as price benchmarks even if you can't bet there.
- Time bets around the inefficient windows: opening lines on lower-tier markets and late-injury announcements before square books update.
- Read margin structure before assuming devigged probabilities are fair — long-shot loading distorts side-by-side comparison.