Free Horse Racing System – Outsider of Three

By on June 24, 2015

I was inspired to try to create this horse racing system when I read a very interesting piece a while back by Matt Bisogno of GeeGeez where he examined five betting cliches.

One of which was the old adage, “back the outsider in a three horse race”.

I often find age-old advice is still very applicable to today’s racing.

So let’s see if it makes for a profitable system.

I’m using Proform Racing as my horse racing system builder.

(In my mind it’s the best out there but it is expensive at £60 per month.)

I’m going to use BSP as a profit guide so I’ll be testing back to 2008 as that is when BSP was introduced.

And I’ll be testing using both UK and Irish racing.


Here are the results when backing every outsider in a three horse race since 2008…


Free Horse Racing System - Overview Since 2008


Most of that data is self explanatory, but there may be a couple you aren’t familiar with.

A/E = Actual winners divided by expected winners. It is useful for determining if a stat is being over or under bet. A value over 1 is good.

Chi Scr = Archie Score. This is used to determine with what probability selections are based on skill or luck. Anything over 4 is considered a robust system.


Straight away you can see that it is indeed profitable to back every single outsider in a 3 horse race.

Betting to BSP you get a very nice ROI.

Amazingly it’s profitable to SP as well!

We have a profitable horse racing system!

But don’t get too excited yet, it’s an up-and-down ride.


Here’s the yearly breakdown…


Free Horse Racing System Outsider Of Three - Yearly Breakdown


But it’s certainly the basis of a profitable system.

The logic being most punters will most probably over-bet the favourite leading to the outsider’s odds being over inflated.

So I decided to check out if I could improve the profit levels adding in some logical rules.

I checked several different form filters such as, days since last run, sex of horse, age of horse, distance, going etc.

But I didn’t find anything of interest.

However I did notice that horses that weren’t wearing headgear were much more profitable to follow.


Here’s the bottom line when just backing those without headgear…


Free Horse Racing System Outsider Of Three No Headgear - Overview


If you were to add this rule you would remove 150 odd runners but increase the BSP profit by 60 points and the SP profit by around 80 points!


Here’s the yearly breakdown…


Free Horse Racing System Outsider Of Three No Headgear - Yearly Breakdownew


So still 3 losing years (just) but profitability has improved.

The problem is, I can’t really explain why adding this rule would result in higher profits.

And I always think adding rules to horse racing systems without understanding why is a recipe for disaster.

So I’m going to throw this one over to you?


Can you think of a reason why this rule should improve profits so much?


If you have any ideas please met me know in the comments section at the bottom of this post.

One final thing I looked at was SP odds.

Now because the logic of the system is based on outsider’s odds being inflated it should be profitable for all odds ranges.

Looking at the stats I can see that there were 82 runners priced over 25/1 with no winners.

Now, these horses probably still represent value bets.

They are still underbet and in these odds ranges it just takes one winner at a large BSP to bring it into profit.

But of course the higher the odds you bet at the longer the losing runs.

And do you really want to be backing no-hopers, waiting for the big priced winner?

I’m not sure I do.


If you were to omit all horses priced over 25/1 here would be the final breakdown…


Free Horse Racing System Outsider Of Three - No headgear and <=25/1 - Overview

Free Horse Racing System Outsider Of Three - No headgear and <=25/1 - Yearly Breakdown

Horse Racing System Rules


1. Back the outsider in a three horse race.

2. Only if the horse in not wearing any headgear and is priced 25/1 or under.


It’s probably not a horse racing system you would want to follow as-is due to the fluctuating nature of the profits.

Rather than the finished product I think this is more of a starting point to be honest and more analysis is needed.


If you have any other horse racing system ideas you’d like me to test or if you can think of ways to improve this one,  just let me know in the comment section below.


Thanks for reading,


  1. XRumerTest

    April 16, 2017 at 3:05 am

    Hello. And Bye.

  2. Mike

    July 3, 2015 at 2:13 am

    Do most of these 3 horse races start out as 4+ horse races with scratches to get down to 3? Just trying to think out the logistics of actually following a system like this, having to watch all the races to see if they became 3 horse races. Sounds like a bot would be needed. Thanks for sharing your analysis.

    • Paul Pendleton

      July 6, 2015 at 10:35 pm

      Hi Mike,

      Unfortunately I’m not able to see whether these are mostly races with non runners.

      I’m sure quite a few of them would be.

      So a bot could well be needed.


  3. Kenny Turnbull

    July 2, 2015 at 7:59 pm

    I’ve noticed that this system preforms slightly better (ROI wise) in non handicap races but produces a slightly lower profit due to reducing the selections.

    Has anyone got any theories as to why the outsider of the 3 might be under bet more often in non handicaps?

    • Paul Pendleton

      July 6, 2015 at 10:38 pm

      Hi Jeremy,

      My guess would be that most punters know that favourites win more often in non-handicaps. So they are less likely to back outsiders in these races hence they are more under-bet.


  4. ian josephs

    June 26, 2015 at 2:46 pm

    I like simple systems showing profits every year for 20 years and average R.O.I around 50% !(proof adrian massey site)
    For example,7lb claimers don’t do well jumping in large fields unless put on horses quickly with a penalty to avoid reassessment;
    Therefore “LAY any horse in a hurdle or chase (UK or Irish)with 20 runners or more,last outing 15 days ago or more,and ridden by a jockey claiming 6lbs or more.




    Win Strike Rate

    Win %Return at Estimated Betfair Odds

    All Selections 9669 266 2.8% 52.6%

    Results broken down by Year



    Win Strike Rate

    Win %Return at Estimated Betfair Odds

    1992 370 8 2.2% 34.1%
    1993 422 13 3.1% 35.6%
    1994 277 11 4.0% 85.7%
    1995 292 8 2.7% 52.4%
    1996 295 5 1.7% 30.0%
    1997 381 10 2.6% 48.6%
    1998 435 10 2.3% 44.5%
    1999 567 8 1.4% 15.9%
    2000 623 19 3.0% 52.9%
    2001 890 22 2.5% 52.6%
    2002 712 18 2.5% 50.9%
    2003 449 13 2.9% 72.8%
    2004 642 25 3.9% 84.8%
    2005 634 15 2.4% 42.7%
    2006 623 26 4.2% 63.0%
    2007 654 15 2.3% 40.9%
    2008 523 11 2.1% 42.8%
    2009 388 13 3.4% 76.9%
    2010 397 14 3.5% 77.5%
    2011 95 2 2.1% 79.8%

    • Paul Pendleton

      July 6, 2015 at 11:01 pm

      Hi Ian,

      I actually get massively different results when I program that system into my software.

      For example here are my results for a couple of the years…

      2009 287 bets 13 wins 4.53% -22.67 points profit to BSP
      2010 341 bets 15 wins 4.4% -351.7 points profit to BSP

      The Adrian Massey site is a decent site for playing around with system ideas but I have to say their estimated Betfair odds are pretty inaccurate.

      For example in 2010 the highest BSP of a winner using this system was 371.4!

      I’m guessing the Adrian Massey site under-estimated that one drastically as it’s SP was 33/1!


      P.S. Sorry for the late reply, your comment was labelled as spam because it contained a link to another website.

  5. frederick

    June 25, 2015 at 8:25 pm

    Regarding headgear, clive keelling gave advice on this fairly recently. Backing all horses wearing any kind of headgear lends itself to a profitable laying system. But like this it is a roller coaster ride and most people cant stand the bad times.

    • Paul Pendleton

      July 6, 2015 at 10:41 pm

      Thanks for the comment Frederick.

      I might look into this idea in a future blog post.


  6. Brian

    June 25, 2015 at 4:56 pm

    Try excluding debutants and horses with a SP of 20/1 or more last time out. I think this makes sense as you never know how debutants will run and horses that were more than 20/1 lto are probably fairly poor. By my reckoning this will give 107 wins from 553 runners and a profit to Betfair SP of 262 points.

    • Paul Pendleton

      June 25, 2015 at 8:03 pm

      Hi Brian,

      I think you’re right about debutants.

      I’ll do some further investigation and include it in the next blog post.


  7. ugpug

    June 25, 2015 at 1:42 pm

    I assume this is forFLAT, but I expect more cases
    over jumps, with less entry’s……?

    • Paul Pendleton

      June 25, 2015 at 8:03 pm

      Hi Ugpug,

      The figures are for all codes of racing not just the flat.



    June 25, 2015 at 11:52 am


    • Paul Pendleton

      June 25, 2015 at 8:00 pm

      Hi Kevin,

      Don’t leave us all hanging, what system do you use?



  9. ian

    June 25, 2015 at 11:15 am

    Headgear is put on to prevent horses interest straying. If headgear was tried out in previous race did it do any good? Perhaps one should read the previous race summary(Bet 365 is good and easy to find).
    If the horse was a stallion but is now a gelding.
    This is getting heavy!
    One question is dutching outsider two any good?

    • Paul Pendleton

      June 25, 2015 at 8:01 pm

      Hi Ian,

      It certainly is getting heavy lol, but thanks for the ideas.

      I’m not sure I can test all those factors with my software but I’ll have a look.

      If I can I’ll include them in my update post.


  10. paul

    June 25, 2015 at 11:04 am

    I was always told by granddad back if over 8/1

    I wonder though with 2 places available on exchanges is there a place system in this ??

    • ian

      June 25, 2015 at 11:18 am

      Only if the declaration list on Betfair is of required number of declared runners.

  11. David

    June 25, 2015 at 10:57 am

    I wonder if this also works with the outsider in a 4 horse race, 5 horse race, etc..

    • Paul Pendleton

      June 25, 2015 at 7:57 pm

      Hi David,

      That’s something I didn’t actually look into David.

      Thanks for the idea. I’ll do a follow up blog post and investigate this.


  12. Will

    June 25, 2015 at 9:59 am

    Hi Paul, Thanks for the analysis. I use HorseRaceBase and by concentrating on 2014 and 2015 only, and these filters:-
    Age 3YO
    Odds 2/1-12/1
    We get these results:-
    Bets Wins WinStrike SP_PL Places PlaceStrike
    37 18 48.65% 99.25 18 48.65%

    Not many selections admittedly but impressive returns.


    • Paul Pendleton

      June 25, 2015 at 8:05 pm

      Hi Will,

      A very nice win strike rate too! Almost 50%!

      Thanks for the comment,

  13. ian

    June 25, 2015 at 9:48 am

    i wonder if age makes any difference? younger horses are surely more likely to improve surprisingly than exposed ones. maybe if you break out 2yo and 3yo? but thanks paul

    • Paul Pendleton

      June 25, 2015 at 8:17 pm

      Hi Ian,

      Your logic makes sense but it’s not borne out by the stats.

      The most profit does comes from 3yo’s, but they also number the most selections.

      2yo’s make a very small profit.

      The 2nd and 2rd most profitable ages are 8 and 9.

      So there doesn’t appear to be an age angle with this system.

      Thanks for the comment though.


  14. John

    June 25, 2015 at 9:10 am

    The old system I knew: whenever there are 2 or more races with just 3 runners in each race, back them in cross doubles, i.e. 3×3 = 9 win bets if just 2 races, or 27 bets if there are 3 races, etc.

    I’ve seen this system produce big returns, but don’t have any figures for you. It’s also an easy system to forget to use, since there are so few days with 2 or more races featuring just 3 runners in each race.

  15. bob gibbs

    June 25, 2015 at 8:34 am

    very interesting paul will look at it more closely

  16. Tony Cox

    June 25, 2015 at 8:26 am

    Hi Paul, informative as always and some impressive figures. How does the 2nd favorite fare under the same conditions? Is there a bias towards favorites in higher class races? Does distance become a factor? ie the shorter the race, more favorites win?

    • Paul Pendleton

      June 25, 2015 at 8:26 pm

      Hi Tony,

      5f races have a poor record with this system. With just a 9% winning strike rate for the outsiders.

      But then 6f is very profitable for outsiders so I don’t know if this is just randomness.

      Overall there’s no real pattern to the profits when just looking at distance.

      Race class doesn’t seem to have any affect at all.

      2nd favourites show a loss with this system.


Leave a Reply

Your email address will not be published. Required fields are marked *