The Thursday Take: The Rising-Star Jockey Edge Is Real. You Can Only See It Afterwards.
This is The Thursday Take — our weekly column on RaceMetrics, where we take a racing talking point and put it to the test against the Proform database. No hunches, no hype. Just what the data says. This week: Billy Loughnane has been crowned champion Flat jockey, and the question every punter asks about a rider on the way up. Is there money in following them?
This week's talking point
On Champions Day at Ascot, Billy Loughnane was presented with the Flat jockeys' title. Three years ago most punters had never heard of him. That arc — unknown claimer to champion — is the one every jockey-follower dreams of catching early. So we asked two questions. Did the market underrate Billy on the way up? And if it did, is that a pattern you can actually use on the next one?
How to read the numbers. A/E (actual over expected) is winners divided by the winners the Betfair SPs said to expect. 1.00 means the prices were exactly right. 1.12 means 12% more winners than the prices implied; 0.93 means 7% fewer. Every ride counts, including fallers and pulled-up horses. All figures are British and Irish Flat races (turf and all-weather), by calendar year, to 19 October 2026.
Billy, year by year
| Year | Rides | Wins | Strike rate | A/E at Betfair SP |
|---|---|---|---|---|
| 2023 | 875 | 130 | 14.9% | 1.119 |
| 2024 | 1,173 | 162 | 13.8% | 0.996 |
| 2025 | 1,321 | 222 | 16.8% | 0.967 |
| 2026 (to 19 Oct) | 1,137 | 225 | 19.8% | 0.930 |
Read across: he wins more often every year, and the prices have more than caught up. In 2023 the market did not know what it had; by 2026 his name is on the price before the horse is. Two honest notes before anyone builds on this. First, a single rider's year is a small sample: each of these figures sits within about one and a half standard errors of 1.00, the range chance alone covers. It is a direction, not proof. Second, profit at Betfair SP tells a noisier story than A/E, because one or two big-priced winners swing it; A/E is the steadier measure, which is why we lead on it.
Is Billy unusual? 43 breakthrough seasons
To find out, we took every Flat rider whose first 50-winner season came between 2019 and 2025, and lined up their years around it.
| Year, against the breakthrough | Riders | Rides | Wins | A/E |
|---|---|---|---|---|
| Two years before | 31 | 7,718 | 704 | 1.013 |
| The year before | 43 | 12,499 | 1,207 | 1.014 |
| The breakthrough year | 43 | 22,173 | 2,727 | 1.077 |
| The year after | 43 | 22,542 | 2,498 | 1.004 |
| Two years after | 36 | 17,461 | 1,974 | 0.987 |
| Three years after | 28 | 12,901 | 1,569 | 0.994 |
Billy is the textbook case. The breakthrough year beats the prices by 8%, and by the next season the riders are priced right. Split the breakthrough year into quarters of each rider's rides and you can watch the market learn: 1.151 in the first quarter, 1.066 in the second, 1.047 in the third, 1.040 in the fourth. The value is front-loaded, and it is gone by the following January.
The catch: you only know it was a breakthrough afterwards
Here is the problem with that table. We picked the breakthrough year because it was the year the rider won 50 races. So of course it looks good: the winners that made it a breakthrough are the winners that make the A/E high. It is like picking the best month in a fund's history and calling it the fund's edge. Nobody could have backed it, because nobody knew in January which riders would get there.
So we asked the question properly: using only what you could have known at the time, can you pick out the rising rider before the market does?
| Rule you could set in advance | Rides | A/E | Return at Betfair SP |
|---|---|---|---|
| On 1 January: 30+ winners last year, and double the year before | 28,869 | 1.009 | −2.9% |
| On 1 January: established riders, 30+ winners last year | 268,362 | 0.998 | −2.4% |
| Each month: 15+ winners this year, double the same point last year, under 50 all last year | 37,809 | 0.981 | −7.4% |
| Each month: everyone else with 15+ winners so far this year | — | 0.994 | −3.8% |
Nothing. The riders who doubled their winners last year came in at 1.009, within chance of 1.00. And the in-season version — the rider visibly ahead of last year's pace, the one you would read about — is the worst group of all, at 0.981 and a 7.4% loss. By the time a rising rider is visible in the numbers, the market has already moved the price. The edge in the rising star is real. It just only exists in the rear-view mirror.
And the champion? Not overbet either
The other half of the folk wisdom is that the top riders are always overbet, so you should look elsewhere. That one does not survive either. In British Flat races since 2019, riders who finished in last year's top three by winners returned an A/E of 0.997 the following year. Riders ranked fourth to tenth: 0.999. Everyone else: 1.000. The top riders are not value, but they are not a lay. The market prices them right, the same as everyone else.
What this does not say. It does not say jockeys do not matter. A good rider wins more races; that is why the prices are shorter. It says the market already knows which riders are good, and adjusts quickly when a new one arrives. Billy's 0.930 this year is the same story from the other side: the price now carries his name.
Build it yourself
You can check any rider in the System Builder, Billy included, and see the same pattern for yourself.
- One rider by year: add a Jockey(s) rule (type the rider's name) and a Race Type rule for Turf and A/W. Open the Breakdowns tab and the Year table: strike rate, A/E and profit at Betfair SP, year by year.
- Every rider at once: with just the Race Type rule, open the Jockey breakdown and set a minimum number of runs. Or use the Grid tab with Jockey down the side and Year across the top, to watch any rider's A/E move as they make their name.
- The claimers: a Jockey Claim (lbs) rule of 1 to 10 gives you the apprentice rides, where the next Billy is riding today.
- The hot rider, right now: Jockey SR % (28d) and Jockey A/E (28d) show each rider's last four weeks. Use them as a lens, not a signal: on the evidence above, a rider's recent form is already in the price.
The take
Billy Loughnane is the best illustration you could ask for of how quickly the betting market learns. The year nobody knew him, his rides beat their prices by 12%. In the year he became champion, they fall 7% short. Across 43 riders the pattern repeats: the breakthrough season always looks like value, and the edge is gone within a year. But we could only find it by knowing the ending. Every rule you could have set at the time came out at the price or worse. A jockey's name is not an angle, and the champion is not a lay. If you want an edge, it has to come from something the market is not already looking at.
Last week we looked at why fewer runners now trade at half their price in-running. Next week: last year's winner comes back for the same race. Does the market overrate it?
The full article — with five common questions answered — is on the RaceMetrics blog: The rising-star jockey edge is real. You can only see it afterwards..