Follow the money, and what happened when we did
Money is information. Informed money moves prices. So follow the money. Every step of that is true, and the conclusion still fell over.
The money is real
Sit with a market for the hour before a race and you can watch it breathe. Prices move. They move most in the last few minutes, and favourites move furthest and latest of all. Nobody who has watched a show board needs convincing of this.
We wanted to measure it properly, so we recorded our own timestamped market snapshots instead of reconstructing prices afterwards. That matters more than it sounds. A price you fetch after the race has already been touched by the result. If you are going to make a claim about when the money arrived, you have to have been there when it arrived.
So the money is real, and the movement is real. That was never the question. The question is narrower and more annoying. Does the move tell you anything the final price does not already tell you?
The comparison that kills it
Here is the test, and it is not a clever one. Take the horses that shortened. Take the horses that drifted. Compare them at the same final price.
The advantage disappears.
A horse that steamed into a given price wins about as often as a horse that drifted out to it. What looked like a signal about market moves turned out to be a fact about which price band the horse ended up in.
You can see why the raw version looks so convincing. If you line up steamers against drifters without controlling for where they finished, steamers win more often. Obviously they do. Shortening means arriving at a shorter price, and shorter horses win more often. That comparison is not measuring the move. It is measuring the price, with an extra step in front of it that makes the price look like insight.
The move told you where the horse was going to end up. The destination told you the rest.
Why the reasoning feels airtight
Because it nearly is. Read the chain again. Money carries information. Informed money moves prices. Therefore a price that has moved carries information. All three true.
The failure is not in the logic. It is in the clock. By the time you can see the move, the information is sitting inside the price you are now being asked to take. You are not following the smart money. You are arriving after it, and being charged for the privilege.
A market move is a receipt, not a tip. It tells you a transaction has already happened. Reading it as an instruction is like hearing a door slam and concluding somebody is about to walk in.
The versions people run instead
Everyone who has held this idea has a refinement. Only big moves count. Only moves in the last ten minutes. Only favourites. Only handicaps. Only when the move goes against the tissue.
Some of those are reasonable ideas and worth specifying properly. But they all have to clear the same hurdle, and the hurdle is not difficult to state. Once two horses are being offered at the same price, is the one that got there by shortening a better bet than the one that got there by drifting? If the answer is no, the refinement has not saved the idea. It has just made it harder to check.
We could not build a version that survived that control. If somebody has one, specified in advance and settled against starting prices rather than fitted afterwards, we would genuinely like to see it. That is not a rhetorical offer. It would be the most interesting thing anyone sent us this year.
What we are not saying
We are not saying market moves are meaningless. We are not saying nobody knows anything, and we are certainly not saying the money is stupid. The money is the opposite of stupid. That is precisely the problem.
What we are saying is that we could not show a steam or drift signal that adds information once the final price is held constant. That is a narrow claim, and we would rather make a narrow claim we can defend than a wide one that flatters us.
It also fits the pattern we keep running into. Across 60,671 races with one clear favourite, the favourite won 35% of the time and did not win 65%. The most watched, most heavily traded runner on the card gets beaten about two times in three, and its price says so in advance. The market is not there to be caught out by things it can already see. Market moves are something it can see.
This was one of the first five beliefs we put under pressure. Three of the five did not survive, and we publish the ones that failed for the same reason we publish the ones that held. Elite Pass has no validated betting edge and does not claim one.
Related research
- 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.
- A real effect and a priced effect are different thingsDraw bias is real and worthless as a standalone edge, both at once. This is how almost every good racing idea dies, and most people never notice it happening.
- The market is very good, so what is left?Everything easy to see is already in the price. What remains is the awkward part: pace, position, course shape, and what a race will actually ask.
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.