TIPSTA is a racing service built around a proprietary form model. Every runner in every Australian and New Zealand thoroughbred race is assessed and given a rated price. That price is our opinion of what a horse should be. The market is what it is actually paying. The gap between those two numbers is the only edge that matters.
How selections are made
Every runner is assessed using a combination of form analysis and a trained predictive model. Form factors — class, ability, fitness, race shape, track suitability, and connections — are used to produce a rated price for each runner in the field.
That rated price is then compared against the live market. The gap between the two is what drives selection decisions — not gut feel, not recency bias.
TIPSTA PRICE is the fair dollar we publish. It reflects our opinion of where that runner sits in the field. It is set when the card is published and does not move because the tote moved.
The methodology is grade-aware. Country races are weighted differently to metro races. Short-priced favourites are treated differently to mid-field runners. The model is not a single rule applied across every race.
HOW THE NUMBERS WORK
0–39 Low · 40–59 Moderate · 60–69 Good · 70–79 High · 80–89 Very High · 90–94 Elite · 95–100 Exceptional
Best Bet, Next Best, Best Value
Three published selections per race. They are not a ranked 1–2–3 from the same list. All three chips fill. Prefer three different horses.
Best Bet is the strongest betting proposition. Probability, conviction, and the available price all sit in that call. It is not automatically the favourite, not automatically the highest Confidence score, and not the top RATING on the page.
Next Best is the next horse we want included. Same quality standard. A different horse when the field allows.
Best Value is the runner that interests us at the available price, usually a $6 to $101 chance with an edge on our book. It is not a roughie label, not Best Long Shot, and not the Best Bet. We do not sort by biggest edge. We do not invent EDGE. If nothing sits in the band, we still name a Value runner from the same book.
How to use the card
- Read TIPSTA PRICE down the field. That is our book — the shortest price is who we have in front.
- Compare TIPSTA RATING and MODEL. Same horse, two reads. A big gap between them is where the race gets interesting.
- Check $WIN last. If the market is paying more than our TIPSTA PRICE, that is the overlay. If it is paying less, we are already shorter than the tote.
- Read the published selections — Best Bet, Next Best, Best Value — as propositions that sit on top of the full field assessment. They are not a shortcut past reading the race.
A horse can be Best Bet without being an overlay. Best Value can sit at a mid price. The selections and the numbers are separate reads. Use both.
What we will not claim
- We will not claim we know the winner.
- We will not print a large edge on a runner we give little real chance.
- We will not copy the market and call it a model.
- We will not invent an EDGE just to make Best Value look clever. The third chip still fills from the same book.
- We will not update a published TIPSTA PRICE because the money came in.
- We will not claim a Best Bet or Next Best is a guaranteed return — win rate is the measure, not a promise.
The card is an opinion with a price. Use it the way you would use a sharp form analyst's view — as one informed read, not as a tip sheet that owes you a result.
Understanding the Confidence Rating
Every selection carries a Confidence Rating from 0–100. That number is how strongly TIPSTA stands behind this horse winning this race. It is not a win probability, not EDGE, and not a stamp that appears because the chip says Best Bet. Price is context. Race difficulty matters. Two horses at the same TIPSTA PRICE can carry different scores. We do not force a 90 just because something is Best Bet or Lock of the Day.
A score of 76 or above means all primary signals are aligned — the model and the form read agree, the data is clean, and the selection sits clearly ahead of the field on our metrics. Below 50, the pick comes with meaningful caveats — it may still be the best available selection, but uncertainty is factored in.
