Two business models
A mainstream bookmaker (Winamax, Betclic, Unibet, bet365…) makes its living on the margin and on advertising. It attracts recreational bettors, limits or closes winners' accounts, and doesn't need perfect odds: a balanced book is enough.
A sharp bookmaker does the opposite: low margin (around 2%), high stake limits, and it accepts winners. To survive professional bettors, its odds have to be right. It refines them constantly with the money coming in: when an informed bettor stakes big on an outcome, the odds move. Its odds end up containing the information of everyone who bets there — which is why the rest of the market copies them.
Pinnacle has been the reference sharp bookmaker for twenty years. Betfair Exchange is not a bookmaker: it is an exchange, where bettors trade positions with each other, and where the price is the balance point between supply and demand, with volumes running into the millions on big matches.
Why their odds are the best probability available
Because they are the result of aggregated information, not of an opinion. Thousands of bettors, including professionals armed with models, put their money where they think the odds are wrong; the price converges on the money-weighted consensus. No single model, no tipster, has more information than that consensus.
This is not a belief: it can be measured. Over thousands of settled outcomes, you compare the stated probability (devigged sharp odds) with the actual result. The Brier score, the predicted-versus-actual curve: the Results page publishes these measurements continuously for Babeth's probabilities, sport by sport.
The limits of each reference
Pinnacle covers the major sports and leagues well, less so lower-tier tennis, darts, snooker, esports or motorsport. Its margin, low as it is, still has to be removed; and since 2025, its API is no longer open to the public.
Betfair Exchange covers almost everything, but the quality of its price depends on liquidity: on a big match, the price is excellent; on a Challenger tennis match three hours before the start, with €200 matched, it is indicative. So you have to weight by volume, and not take an empty market seriously.
Neither is perfect; combined, they correct each other. That is why Babeth uses Pinnacle as the main reference (60%) and Betfair Exchange as the secondary reference (40%) when both price the market and the exchange market is liquid enough (at least €1,000 matched, a back/lay spread of 6% at most, match not yet in play), and Betfair alone when Pinnacle doesn't price it.
How a bettor uses them
Three uses, from the simplest to the most profitable:
- Check a price. Before placing a bet, compare the odds offered with the references' fair odds. If they are below, you are paying too much — even if the bet "looks" good. The odds checker does it in one click on the matches of the next four days.
- Measure your close. After kick-off, the sharp closing line tells you whether you got a good price: that is closing line value.
- Look for gaps. When a mainstream bookmaker pays more than the sharp fair odds, it is a value bet. That is Babeth's job: every minute, she compares the reference prices with about forty bookmakers (the British bookmakers are re-read every one to two minutes).
Can you bet at the sharps?
Yes, and it is even recommended for big stakes: they don't limit. But you rarely find value there — their odds are the value. A typical winning bettor combines both: they hunt for gaps at mainstream bookmakers as long as their accounts allow it, and place their big volumes on Pinnacle or the exchange, where the closing line is theirs. On every matched bet, Proba Bet also shows the best Betfair Exchange price with a direct link to the market.
Key takeaways
Sharp bookmakers aren't better because they are smarter, but because their business model forces them to be right. Their odds, margin removed, are the best probability available — and therefore the yardstick for judging all the others.