Everyone loves to tell the tale of the time they backed a 16/1 long shot winner at Kempton. The thing is, the storyteller probably won’t tell you about the 236 times they backed a long shot loser! If you have studied betting and economics for long enough, you have probably heard of the ‘favourite/long shot bias.’
Basically, it means that punters have a tendency to overvalue a long shot and undervalue a favourite. For instance, in a random horse race, the favourite could be 6/4, and the outsider might be 150/1 when in reality, the true odds of the favourite winning are closer to Evens, and the chances of the long shot winning are about 250/1. In this simple example, the long shot bet is a much worse one statistically speaking.
Recommended Betting Site for Today’s Discussion:
An incredibly detailed study by academics at Stanford and Wharton examined the results of over 6 million races in the United States between 1992 and 2001. Here is what they uncovered:
- Your ROI from backing all favourites would be -5.5%.
- It would be -18% when backing horses at odds between 3/1 and 15/1.
- You would have a desperate ROI of -61% if you backed all outsiders at 100/1+.
In other words, blindly backing long shots is the best way to lose every penny you own. However, it is a mistake to completely discount long shots because as always, there is a way to make a profit if you’re prepared to research thoroughly enough. For the purposes of this article, I will consider horses that start the race at 10/1+ as a ‘long shot.’
Is There Such a Thing as Course Bias For Outsiders?
Let’s begin by showing the folly of blindly backing outsiders. If you bet on every 100/1 SP shot since 2013 in UK races, you would have an ROI of -87% as only 0.11% win (so about 1 in every 900 or so). If you did the same for every 10/1 shot since the start of 2016 (over 91,000 bets), your ROI would be -40.7% as just 3.31% of these horses win.
It isn’t unreasonable to expect there to be a course bias as some courses have conditions more conducive to bizarre results. Here are a few selected courses with the win rate and ROI (on Betfair) for backing horses at 10/1+ since the beginning of 2013.
- Aintree: 3.84% win rate and +13.82% ROI!
- Ascot: 3.4% win rate and -21.06% ROI.
- Exeter: 3.1% win rate and -25.16% ROI.
- Market Rasen: 3.64% win rate and –12.01% ROI.
- Windsor: 3.32% win rate and -25.59%.
Although I have only taken the stats of 5 courses, you’ll notice that all of them have a win ratio between 3 and 4 percent. Aside from Aintree which appears to be an anomaly, the rest of the courses offer you a steady ROI loss of at least 12% and over 21% in 3 of the 5 examples.
How important is Form?
Clearly, you don’t want to toss your money away on a complete no-hoper with no form whatsoever. You would expect a horse with a decent previous run to do better than a long shot that performed poorly last time out.
In fact, while 4.66% of long shots that finished in the top 4 in their previous race went on to win the next one, your ROI would be -10.87% since 2013. In contrast, outsiders that finished fifth or worse last time out only won 4% of races, but your ROI is better at -5.35%.
Of course, you want more proof than a single race. 4.35% of long shots that finished in the Top 4 at least once in their last 3 races won although your ROI is -9.2%. In contrast, the win rate of horses that failed to finish in the Top 4 in any of their last 3 races was just 3.58%. Bizarrely, your ROI is better -3.49%. This indicates that the occasional fluke wins after showing no form whatsoever. The win ratio shows that horses in decent form perform better.
Threading the Needle
It’s about time that I played detective and performed some detailed research. An interesting avenue to go down is the performance of long shots in National Hunt races over longer distances. In all NH races since the beginning of 2015, backing all horses at odds of 10/1+ would result in an ROI loss of over 12%. If you focus on races of at least 2 miles 4 furlongs, the ROI loss falls to 4.72% on Betfair.
The next step is to focus on Class. After some further digging, I found that you will make a small profit of 1.21% if you look at Class 5 races only. Logically, races with fewer runners make the most sense, so I reviewed the performances of our specially chosen long shots in different sized races.
|No. of Runners||Win Rate||ROI on Betfair|
As you can plainly see, the ROI skyrockets when you whittle down the field. Races with just 1-6 runners provide a fantastic ROI of almost 46%. However, with just 14 winning bets since the beginning of 2015, punters might be inclined to look at races with 1-7 runners as there have been 29 winning bets in the same timeframe. If you want even more bets with a smaller ROI, there have been 52 winners in 1-8 runner races.
You’ll also notice that the ROI drops like a stone as soon as you look at races with 5 runners or less. This is good news as it makes it easy to focus your betting.
The Issue of Handicaps
Experts suggest that outsiders perform better in handicap events because there is a chance that the handicapper gets it wrong or the selection turns in a career best performance. It is certainly a common sense approach, but the statistics suggest this is another ‘assumption’ that is completely incorrect. I replicated the table above only this time; I looked at non-handicap races only.
|No. of Runners||Win Rate||ROI on Betfair|
Except for races with 1-6 and 1-5 runners, the win rate goes down in non-handicaps, but the ROI goes up in every case. With 1-6 runner races, you almost double your money. The problem is, there have only been 9 winners since the start of 2015! I guess it is the sole preserve of the very patient punter, but it is interesting to see that non-handicap races fare better. Clearly, there are a couple of extremely long priced winners.
There are a number of great ways to make long-shot bets work for you. However, no matter what system you use, the actual percentage of winners will always be rather low as every bet involves a horse with an SP of 10/1 or higher. Anyone who claims to have a system where long shots win 20% of the time is a liar.
To recap, here is a summary of the criteria mentioned above which led to an ROI of at least 24.88%:
- National Hunt races only.
- Races at least 2m 4f long.
- Horses with a Starting Price of at least 10/1.
- Class 5 races only.
- Races with a minimum of 6 runners and a maximum of 8 entrants.
- Non-handicap races are best, but there are very few bets.
And finally, we should note that Betfair tends to offer the best prices for long shot horses. Betfair does have to manage risk in the same way as other betting sites because you bet against other punters rather than against the house. Therefore, you will often find better payouts for long shot runners at Betfair.