Outsiders in small fields
 

I was watching the races today and a few outsiders in small fields (ie 8 or less horses) ended up winning. I usually ignore races like these. But I remember hearing Richard Freedman mention that it happens frequently...

I was wondering what the stats are like for these type of races if anyone has them. At Port Macquarie, race 6 the 100 rating won - even if it was the outsider. At Mornington, race 8, the 94 rating won (second worse horse) - and it was the clear outsider.

In a sense, Freedman is correct, but not telling the whole story? The smaller the field, the strike rate of the outsider will always be higher. It's not that you've stumbled across an edge or anything, it's just that the less horses to beat, the higher the percentage of win rate. The price of the outsider/longshot usually reflects that though, unless it's an absolute 'donkey' with no hope at all.

The more important factor to consider is the price of the favourites in a small field. Sometimes punters Focus and plough too much money into the first three and ignore the outsiders....to their detriment, as some of these outsiders can end up a value bet. The best place to cash in on this, is the Harness/Dogs, but only if you're an expert on form. Lightly traded Dog and Harness races often see a big bet placed on a particular selection by connections, then the sheep stampede to get on this 'sure thing' with little else wagered on the rest, pushing their prices out.

Similarly big fields of novice two year olds. The few with any form at all, tend to attract the punters money, as what else will the AAP computers be able to rate? Then along comes a longshot smokie, with only barrier trial form, and wins.

"The less horses to beat, the higher the percentage of win rate" - I would think this also refers to favourites in small fields? If so, one can assume there is no real advantage in backing outsiders for these races?

If you're looking for races with small fields, then Exeter in the UK, is the place to be! Their race card overnight, comprised 2 x 7 horse, 2 x 4 horse and 3 x 2 horse races. And if you think there's no money for these races, forget it, as I think the smallest Betfair pool was 360,000 pound and over a million for one of the 2 horse races where you had a 1.05 fave and 18 2nd fave. Fortunately for some, the 1st fave got up (it was a Hurdle race, so anything can happen)

It's the same in any size field. The price of the outsider or the favourite roughly corresponds with the strike rate for that size field. With exception of course of various stars or donkeys.
Over time it evens out.
However you have periods where outsiders get up in small fields, and you remember it. Then you have periods where the faves get up. You don't remember that as well, because it's expected, if you get my drift.

So are we saying that the favourite still wins roughly 30% of the time for smaller fields. Or do the outsiders (ie 10-1 and greater) have a better strike rate for smaller fields ?

Sheesh....I'm sure that's what the last 3 posters said??....but so that it really sinks in:

5 horse race..1st fave 41%.......5th fave 6%
6 horse race..1st fave 38%.......6th fave 3%
7 horse race..1st fave 37%.......7th fave 2%
8 horse race..1st fave 35%.......8th fave 1%

14 horse race.1st fave 29%......14th fave 0%*

* No such thing as zero % strike rate in racing (rounded down from less than half of 1%)

dcpg,

For all races of all field sizes the 1st fave is around that magical 30% mark.

For each field size the 1st fave win % is different as per RCP's post. But what CP and RCP are getting at is that those %'s essentially do not change over time and that the available odds generally reflect these % chances of winning less the take out.

So yes the 5th fave in a 5 horse field does have a higher chance of winning - 6% - BUT the odds available on that horse are also much lower on average too i.e. $15 compared to the 14th fave of a 14 horse field where the odds would be maybe $300 which reflects its <1% chance of winning.

Hopefully that clears it up.

If you lay horses in Oz gallops races whose price is 1.5 - 1.95 at the jump and with 7 or more starters you will collect 48% of bets. If you lay off in running any that lead you will do even better.

I've been monitoring this since I saw it on 24/4 and since a 7.45pm winning race at Cranbourne. There have been 14 bets for 7 winning lay bets and a profit of $4.13 for $5 lay bets after commission (+$6.40 before commission) and no laying off after the race - just straight out lay bets. Does this correspond with your findings?

Sorry to disappoint you TO, but Speedy spoke of strike rate, but not POT!

Since the 16th March, for $1000 liability, UK, IRE & AUS:

IN nett..$170,265
OUT......$170,000

170 wins, 134 losses = 55.92% win strike rate
Implied strike rate = 57.97%

It's actually a perfect Premium Charge minimisation scheme, that's about all. Use it to offset PC charges on some other profit making method.

Sorry RP, I thought it may have been a slow burning profitable method but obviously not.

Ha ha haaa, you don't think Speedy's going to give away anything profitable do you, it'll cruel his edge. What it is though is a good starting point. Add a few filters, like class, barrier, track cond, blah, blah....and you might have yourself a profitable system. Good luck :)

That is correct, Ocho. It does win at LS but I also use a modified Fibonacci method which accelerates the profits.
I gain a significant edge by laying off on any that lead in the run.
Shame on you, RP. I have given away almost everything I know on this site. I'm sure I'll get my 16 virgins for that. I'll pass on the virgins if I have to strap a bomb to my chest. I always thought that they were overrated.

I don't do Ireland and I have different rules for the UK. Same price range tho.

Before you dive in TO, or anyone else, as soon as I saw the word Fibonacci staking (a fancy word for chasing losses by increasing stakes), I saw the word DANGER in my mind.

First thing to consider is the 'Run of Outs'....for selections with a 55% strike rate you can expect and IT WILL happen, a run of losses of 9 at some stage.
Using standard Fibonacci, the sequence of staking is 1, 1, 2, 3, 5, 8, 13, 21, 36... By the time you reach your 9th out, you've already lost 90 times your starting stake and about to plonk down 57 x stake on the eventual winner! Think your nerves can handle that.....I don't think so.

I took my original data and started working it out. The biggest draw down was 33 x stake. After 100 bets, all the Fibonacci staking had done was accelerate my losses to nett -9 x stake. Not saying it won't come good, but not hanging round to find out! If I'd left it at level stakes I'd have a $163 profit. I did a spreadsheet and was going to attach it, but thought, "nah, you'll find out!" Stick to non-staking filters IMHO :)

PS: As for what region of the world this data is taken from, makes no difference, maybe a bit of difference between Harness and Gallops, that's all.

Thanks for that SB (and RP). So it is profitable at level stakes but obviously only a slow burner. I have registered only 1 bet in the last 3 days (Bendigo race 3 today which was a losing lay bet with no bets at all on Thursday or Wednesday) is that right?

I don't suppose you can hint at the UK racing different rules you have (or email me at equalthree at yahoo dot com dot au)?

Any chasing never works for me and as RP says the stakes would get pretty high.

Why did you work out the losses at even money? Very slack of you, RP, as we are talking about an ODDS ON system. Senior's moment, perhaps?

Will email TO. The method was voted best system for each of the last two years in the UK. I found a free copy on the web but it sells for about $300 now having started out much more humbly.

Speedy raises a good point (not so much about the 'senior's moment' though, that was a bit low. Still got a few years to go before I get the pension Speedy). BUT does he base his system results on Starting Price or Last matched or Last Back offer??

Sure when you look at historical results you can develop all sorts of weird and wonderful systems. TRY it live or with a bot, and see how you go? My results are based on a bot recording the Back price at the official start time. Below are the results for the last 3 days, to answer TO's request. 4 lay losers in a row. Might be a different result if you sat there 24/7 and watched live vision to get as close to the start time as you can. OR, you could always get the bot to place an SP bet into the market based on a price trigger, but SP might end up outside of the $1.50 ~ $1.95 range?

The only way you can publish results accurately is to clarify how they were recorded, historically (published by Betfair), manually or botted (if so at what time to start?). My comments on staking systems still stand though, but Speedy was using a 'modified' version, again another detail missing?

Thanks SB.