The market is very good, so what is left?
We spent months testing things racing repeats. Most of them turned out to be true and already paid for. This is the reckoning, and then the honest answer about where we are still looking.
Where we got to
Five beliefs went in. Three did not survive.
Everybody watches the favourite, so we counted. Across 60,671 races with one clear favourite, it won 35% and did not win 65%, which is roughly what the prices implied it would do. Short prices are where the value hides, so we split favourites by price band and looked for one that was systematically better than its number. We could not find it.
Follow the money died on a single comparison. Steamers against drifters at the same final price, and the advantage vanished. Good form finds good horses turned out to be true and mostly already absorbed into the price by the off. The draw decides it kept its dignity in some conditions and returned a clean null on held out races as a standalone edge, so we killed it and kept the record of killing it.
Favourites are also more vulnerable in big fields, which sounds like a finding until you check whether the market has noticed. It has, and it charges accordingly.
The pattern underneath all of it
Look at those results together and they are not five separate stories. They are one story told five times.
Anything printed on a racecard in a form a thousand people can read ends up inside the price. Form figures. The draw number. The rating. The market move you can see on the board. None of it is hidden, none of it requires interpretation, and all of it has been read by everybody before you got there.
That is not the market being mysterious or clever. It is the market being a crowd. If the information is legible and free, the crowd has it, and the number reflects it. You do not beat a crowd using the page the crowd is reading.
So what is actually left
Not a secret factor. Not a rating nobody has thought of. If it were a number you could put in a column, somebody would have put it in a column already, and it would be in the price by lunchtime.
What is left is the part that does not reduce to a column at all. The interactions. Pace, position, course shape, and what a specific race is going to ask of a specific horse.
Think about how that works in practice. A horse who needs the lead is a different proposition depending on whether it is the only one in the field who needs it or one of four. A closer in a race with nobody willing to go forward is being asked to win a sprint from the back after a crawl. A track that turns sharply, a trip that is a furlong further than ideal, and a draw that puts the horse on the wrong side of the pace are three small things that can compound into one big one.
None of that is about how good the horse is. It is about what the race will do to it.
Why that is hard to price
Because you cannot look it up. You have to read the whole race as a single object, work out what each runner is likely to do based on what it usually does, and then work out the shape that follows from all of them doing it at once.
That is a judgement about a system with a dozen moving parts, made before the off, on information that is genuinely ambiguous. Two careful readers can look at the same race and disagree about where the pace is coming from. Nobody disagrees about what a horse's last three form figures are.
And here is the uncomfortable symmetry. Hard for the market to see means hard for us to see. It also means hard to measure, because the thing you are trying to measure resists being written down cleanly, and anything that resists being written down cleanly is where measurement error goes to hide.
What we are actually claiming
Less than you might expect from a company that has just spent a thousand words explaining where it is looking.
We are not claiming the race shape work produces an edge. We have not shown that. We have a hypothesis about where the unpriced information plausibly sits, and a method for testing it that has already killed three ideas we liked. If the settled record says this one is null as well, we will publish that with the same wording we used for the draw, and it will be an unpleasant morning.
What we can say is that the discipline is real. We record our own timestamped market snapshots before racing rather than reconstructing prices afterwards, because a price read after the race has already been contaminated by the result. Everything gets settled against starting prices, including the things we hoped would work.
The bottleneck was never a shortage of ideas. Ideas are cheap and racing is full of them. The bottleneck is evidence, and evidence arrives one card at a time, which means the honest answer to what is left is that we are still finding out.
Elite Pass has no validated betting edge, does not claim one, and recommends no selections. We publish the tests that fail alongside the ones that hold.
Related research
- What price control did to our best looking ideasI used to think the draw was worth money. Then we held the final price constant and watched idea after idea thin out. Hundreds of systems analysed, none validated.
- Everyone remembers the beaten favouriteShort-priced favourites getting turned over feel like events. The wins feel like the default. That asymmetry is memory, not maths.
- Good form finds good horses, and the market knowsRecent form is genuinely predictive. Hold the final price constant and most of what it was telling you has already been absorbed by the off.
Elite Pass publishes research, not guaranteed outcomes. Findings are measured against the market and remain subject to replication and sample size. The evidence room and the public ledger carry the full record.