Betfair Starting Price is the closest thing Australian racing has to a true market price. It is a single number, calculated at the moment a race jumps, representing where the exchange settled after every piece of information and every dollar of money had its say — with no bookmaker margin built into it at all.
That makes BSP two things simultaneously: a way to bet, and a measuring stick. Both are worth understanding, and the second is arguably more valuable than the first.
How BSP is calculated
At the moment the market is turned in-play or suspended for the start, Betfair runs a reconciliation:
- Collect all BSP orders. Every back order placed "at SP" and every lay order placed at SP goes into the pool.
- Add unmatched limit orders. Resting orders on the exchange book that are still unmatched at the off are included, since they represent real willingness to trade.
- Find the balancing price. The algorithm searches for the single price at which the total money wanting to back equals the total money wanting to lay.
- Match everyone at that price. Every BSP backer on that runner gets the same number. Every BSP layer gets the same number.
Two protections are built in. Backers specifying a minimum price will not be matched below it. And Betfair applies a "best execution" principle — if a limit order on the book offers a better price than the reconciled BSP, you get the better one.
A worked example
Suppose a runner has $8,000 of BSP back money and $6,000 of BSP lay money, plus limit orders resting on the book. The reconciliation searches upward and downward:
| Candidate price | Back money wanting in | Lay money available | Balanced? |
|---|---|---|---|
| $5.00 | $8,000 | $4,200 | No — too much back money |
| $5.50 | $8,000 | $6,600 | Closer |
| $6.00 | $8,000 | $8,050 | Yes — reconciles here |
BSP settles at $6.00, and every BSP backer on that runner receives $6.00 regardless of when they placed the order — three days out or thirty seconds before the jump.
BSP versus the alternatives
| Price type | Margin / takeout | Known when you bet? | Restricts winners? |
|---|---|---|---|
| Fixed odds (corporate book) | 14–20% built into odds | Yes | Yes, aggressively |
| Tote / SP dividend | 14–25% takeout from pool | No | No |
| Betfair fixed exchange price | ~0% + 6.5% commission on net wins | Yes | No |
| Betfair Starting Price | ~0% + commission | No | No |
The critical structural point: a tote takeout and a bookmaker margin are extracted from the price; exchange commission is charged separately on net winnings. That is why a 6.5% commission does not put the exchange on a level footing with a 17% bookmaker margin — commission applies only to profit, and only on winning positions, whereas margin applies to every dollar of turnover.
What that means in practice
Consider a runner you believe is a genuine $6.00 chance:
| Where you bet | Price offered | Effective return on $100 |
|---|---|---|
| Corporate book at 17% margin | $5.00 | Losing proposition against true price |
| Best available fixed across market | $5.50 | Still short of fair |
| BSP | $6.10 | $510 net of 6.5% commission if it wins |
The gap widens as prices lengthen, which is the practical heart of the BSP case.
Where BSP beats fixed odds, and where it does not
| Situation | BSP vs best fixed | Why |
|---|---|---|
| Metro Saturday, mid-range runner | BSP usually better | Deep liquidity, no margin |
| Metro Saturday, longshot | BSP much better | Books apply heaviest margin to longshots |
| Short-priced favourite | Often fixed better | Promotions and top-fluc offers subsidise favourites |
| Country / provincial | Mixed | Thin exchange liquidity widens reconciliation risk |
| Greyhounds, metro | BSP usually better | Fixed greyhound margins are 18-25% |
| Greyhounds, country | Risky | Very thin liquidity |
| Melbourne Cup and majors | BSP better outside the favourite | Peak exchange liquidity; books over-shorten the top |
| You have been limited | BSP, obviously | The exchange does not restrict winners |
You can see this comparison live — the Krok Odds exchange view places Betfair pricing alongside every corporate book on the same runner, which makes the "is BSP better here" question a glance rather than a calculation.
The bigger use: BSP as a benchmark
Betting at BSP is useful. Measuring against it is transformative, and it is the reason serious punters care about it.
Because BSP carries no margin and reflects the market's final aggregated view, it is the sharpest closing reference available in Australian racing. Comparing the price you took against BSP gives you closing line value:
CLV % = (price you took / BSP − 1) × 100
Take $8.00 on a runner that returns a BSP of $6.50 and you have +23% CLV on that bet. Take $5.00 on a runner that returns $6.50 and you have −23%.
Why this matters more than your profit-and-loss in the short term: CLV converges far faster than results do. Across a few hundred racing bets, your average CLV tells you whether your selection process is finding prices the market later agrees were too generous. Your profit-and-loss over the same sample is dominated by variance.
Krok Odds settles tracked racing bets against captured Betfair starting prices automatically, so the bet tracker produces a CLV figure per bet without manual reconciliation.
Practical mechanics
- Always set a minimum price. Placing an order "at SP" with no floor means accepting whatever the reconciliation produces, including a collapse caused by a late plunge. Specifying a minimum costs you nothing when the price holds.
- Check liquidity before relying on BSP. In a metro Saturday market the reconciliation is robust. In a Tuesday country greyhound race with a few hundred dollars matched, it is not.
- Account for commission in your maths. 6.5% on net winnings is not nothing. A $6.00 BSP winner returns $5.675 of profit per dollar, not $5.00 — but you should still model it explicitly rather than ignoring it.
- Know that BSP is available on some corporate products too. Several Australian books offer "best of best" or BSP-matching promotions on selected meetings. Read the caps.
- You cannot get BSP after the jump. Orders must be in before the market suspends. There is no late entry.
Common misunderstandings
"BSP is just the last traded price." No. Last traded price is whatever the final matched bet happened to be. BSP is a reconciliation across all SP orders and resting limit orders, which is a materially different and more robust number.
"BSP is the same as the tote SP." No, and the gap is large. Tote SP has 14-25% takeout removed from the pool before dividends. BSP has no margin extracted from the price at all.
"Betting at SP means you get a bad price." True at a corporate bookmaker, where SP is usually the worst price on offer. Not true on the exchange, where BSP is frequently the best number available on the runner.
"BSP always beats fixed odds." No. Short-priced favourites with a promotion attached, and thin country markets, are both genuine exceptions.
When to use what
- Short favourite with a top-fluc or best-tote promotion: take the fixed price.
- Mid-range runner in a liquid market: compare best fixed against expected BSP; BSP usually wins.
- Longshot in a liquid market: BSP, almost always.
- Thin country or provincial market: fixed odds, or a limit order rather than SP.
- You have been restricted at the corporates: BSP is the whole answer.
- Measuring whether your process works: BSP as the benchmark, regardless of where you bet.
For a deeper treatment of how BSP interacts with Australian racing markets specifically, see our racing-focused BSP piece, and the commission and premium charge explainer for the cost side of the exchange.
Compare BSP against every bookmaker
Krok Odds shows Betfair Exchange prices alongside 65+ AU bookmakers on every runner, and settles tracked racing bets against captured BSP for closing line value.
Open the exchange view →
David contributes long-form retrospectives, case studies, and operational pieces on AU advantage betting markets. His writing focuses on the realities of running a sustainable AU advantage operation — what works, what fails, and the operational details most blogs gloss over.