Backing the Second Favourite in Betfair Horse Racing: End of September 2026 Update
What happens if you simply back the second favourite in horse races on Betfair?
That is the question behind our Global Second Favourite Horse Collector, which went live on 25 September 2026.
The collector operates across global Betfair horse-racing markets. Shortly before the scheduled start, it identifies the horse sitting second in the market and backs it at the available exchange price.
There are no preferred countries, no selected race types and no carefully chosen odds range. The idea is deliberately broad. Collect the bets first, build a substantial dataset and see what the evidence eventually tells us.
The first snapshot contains 1,240 actual live bets.
A Warning Before the Results
This is not a horse-racing betting strategy and it should not be copied with meaningful stakes.
We haven't historically tested hundreds of combinations and then selected the profitable version. The collector was set up to answer a question and gather data. It may ultimately lose money — and its opening results provide a fairly good reminder of that possibility.
The objective here is research, not tips.
The First 1,240 Bets
The opening results are:
Total Bets: 1,240
Winners: 256
Losers: 984
Strike Rate: 20.7%
Average Odds: 5.07
Average Winning Odds: 4.55
Longest Losing Run: 23
Profit/Loss: −93.31 units after 2% commission
That's not a typo.
The collector lost more than 93 units during its opening 1,240 bets.
It also produced a substantial amount of data very quickly. The bets came from racing in Australia, the United States, Great Britain, France, Ireland, New Zealand and South Africa, with Australia and the US accounting for much of the early volume.
The races themselves were varied too, including everything from small fields to markets containing as many as 30 runners.
For something that had only been running for a matter of days, that's already a useful amount of live-market information.
1,240 Bets Is Enough to Be Interesting, Not Enough to Start Fixing Things
This is where research can become dangerous.
We could immediately start digging through the results looking for profitable countries, odds bands, field sizes or race types. With 1,240 bets, we'd almost certainly find something that looked good.
Then we could remove everything that performed badly, keep everything that performed well and announce that we'd improved the strategy.
We're not going to do that.
The collector was deliberately designed to be broad. At this stage its job is still to collect.
The −93.31 units matters. We're not hiding it or dismissing it. But we're also not going to react to it by changing the rules after a few days.
If the same experiment is still losing after 5,000 or 10,000 bets, that will be considerably more informative.
What Are We Actually Trying to Learn?
“Does backing the second favourite make money?” is only the starting question.
As the dataset grows, we'll be able to examine how these horses behave across different prices, countries, field sizes, race characteristics and levels of market liquidity.
Perhaps nothing useful emerges.
Perhaps the broad approach continues losing steadily.
Or perhaps the larger dataset eventually gives us a reason to investigate a particular characteristic more closely.
We don't know yet. That's why we're collecting the data.
And there's another number in this opening sample that demonstrates why looking only at strike rate can be misleading.
The second favourites won 20.7% of their races. One winner in roughly every five bets might sound respectable in isolation. But at average odds of 5.07, the relationship between the prices, winners and losers ultimately matters far more than the strike rate itself.
After commission, this opening sample finished 93.31 units down.
That's the evidence so far.
Next Stop: More Data
The collector will continue running without us trying to rescue it.
The next few thousand bets should make the dataset much more interesting. We'll be watching how the overall result develops, but also what happens beneath it: price ranges, countries, race characteristics, field sizes and other patterns that may become worth investigating as the sample grows.
For now, though, the conclusion is deliberately simple.
We asked what happens when you consistently back the second favourite across a broad range of Betfair horse-racing markets.
After the first 1,240 live bets, the answer is −93.31 units after commission.
That's not a betting system.
It's the beginning of a dataset.
If you're interested in the research process behind these experiments, you can explore the Trade Carefully Research Framework.
Trade carefully.
Comments