Is AW racing getting predictable? ( A system for FAVS)

[man o bong]

I have long been a fan of All Weather racing since Lingfield kicked off in 2001 and Wolverhampton 4 years later. Checking my personal records confirms the reason why when i see I have been able to make a consistent profit, more than I thought in fact, on these tracks. The exception would be Kempton, but being a relatively new track I guess I have yet to find my level there. It has been nice to enjoy some action from Ireland from Dundalk in the latter part of 2007 and it is yet to be seen how I fare at Great Leigh’s when it eventually opens it doors as the first new racecourse since Taunton in 1927.

I have publicly advocated all weather racing as a good thing many times in the past, as I feel from a form reading perspective the job is far easier than trying to sort out the idiosyncrasies of turf courses and interpreting that into my form reading. The advantage I see from analysing horses performances is the level playing field we have in regards to the ever so slight differences in going, the similarities and differences between the polytrack and fibre sand and more importantly the fact that races are run over the same advertised distances rather than having to factor in guesswork in timings due to rails being moved at turf courses to help, save and preserve ground. All these things I feel take some of the work out of assessing each runner’s previous performances.

I got to thinking the other day whilst on a lengthy drive that I thought I had noticed a larger than normal amount of favourites going in on the all weather over the last couple of months. Wolverhampton especially, I was thinking had thrown up a good number in particular.

So I began to ask myself, “had the all weather started to become more predictable?”

There was only one way to find out – Hours of research, but I was determined to answer my own question and set about the task in hand.

I ran my software over all all weather races run between 1st November and 28th of December inclusive analysing just the favs, sole and joint, and their performance. Backing all favs to level stakes would have produced a 30%SR and a loss of 40.60pts, this is only lessened slightly if concentrating on clear favs by an up in strike rate to 31% and a loss of 35.68. I did notice however that by concentrating on races with a minimum of 10 runners a better return could be had, presumably by the fact the favourite prices would be larger, and also to limit the analysis to races between 5f and 10f.

With Southwell and Lingfield performing as I thought they might, giving heavier losses, I decided to take those two courses out of the equation, factor in the runners and race distances and analyse again.

Now I began to get excited. From 146 favourites at SP we had now found 48 winners for a strike rate that didn’t improve by a great deal, only to 32% but the level stakes investment had turned from a loss into a healthy looking profit of +20.03. A 20pt profit over two months is more than acceptable for me. However a two month period that had stuck out in my mind as being successful was not a big enough sample for me to base a system going forward. So time to bite the bullet and test the system for a longer period. Before doing so I noted that top class races were surprisingly unfavourable for the system as were banded class 7, so they were omitted as I looked to analyse all races for 2007 with the following criteria


SP FAVS
Courses: WOLVERHAMPTON – KEMPTON
Distances: 5f to 10f
Runners: 10 or more.
Class: 3,4,5 and 6

I was disappointed at first to discover that the strike rate had dropped by 3% down to 29% (208-704) and all the profit had disappeared as a single point investment in each runner would have returned a loss of -2.46 for all favourites, there was a slight increase in backing clear favs of 31% and a profit of +5.05.

So what had changed? Well presumably my initial thought that the past two months had seen more favourite than usual winning is borne out by these figures compared to the smaller two month sample I had run previously. But the returns were still in a region where I thought it worth a little look to see if there were any other places we could look to improve returns betting blindly on favs at these two courses. It didn’t take me long to fins it. Going through the specific race types I noticed that all specs had returned profit with the exception of two, namely Apprentice races in particular had let the system down with just two winners from 19 races (10%) and a heavy loss as they both returned short odds, the other was auction races which although a 32%SR was achieved once again the prices were so short that they took a huge chunk of the profits out.

There was only one thing for it, omit these types of races out of the criteria, and run a full test to the beginning of 2006, a date decided as a good starting point as racing was still new at Dunstall Park at this time but had been going for long enough to now be a part of trainers regular schedules. I looked at another couple of factors to tweak and decided on concentrating on 9-10f races in classes 4 and 6 only. There are not enough class 3 events to warrant long term analysis and class 5 [imo], is a testing ground for trainer to see if their charges are ready for stepping up or stepping them down to assess the level of regression in a horses form.

So we would now get a large enough sample to decide if this might be a system worth following going into 2008 based on the returns the tests would prove.

A quick run through for 2007 alone gave a nice manageable amount of bets for the year at 132 rather than the 704 previously so looked ready for the full test.

RESULTS

SYSTEM TESTED BETWEEN 1/1/06 – 28/12/07

CRITERIA

SP FAVS
Courses: WOLVERHAMPTON – KEMPTON
Distances: 9f to 10f
Runners: 10 or more.
Class: 4 and 6
NOT auction or Apprentice races

RESULTS

ALL 76-212 (35%) +48.71
ALL [LONE] 83 252 (32%) +45.25

2006 35-120 (29%) +9.47
2007 48-132 (36%) +35.79
2008 ?

The only difficulty with this system is the flip flopping of favs in the market, during the run up to the race, otherwise it is relatively simple to run.

The improvement of profits from 2006-2007 is nothing out of the ordinary, or at least that we can gauge as yet, with another years results it will become clearer if this is a trend of “predictability” for AW racing under the criteria used or of course the system could return a loss over the coming year. I am however confident that we should get another year of good results from this system.

One final thing that I think is worthy of note is that this system for both 2007 alone and for the two years combined has produced a profit in 10 of the 12 months of the year, so it does produce a nice trickle charge of interest throughout the year

I shall update the 2008 figures regularly, although for obvious reasons it will be impossible for me to post up the selections on a daily basis unless I am sat here 24/7 watching the markets, and as much as I love all the dear OLBG forumites and lurkers alike, I aint prepared to do it.

However I have set out the full criteria of the system for all to use for future reference should you so wish to do so.

Good luck and happy systematic punting.


MOB.
_________________
6 Months Horse Racing Blogs
+£13,376 (LSP +113.5pts)
6 Months Football Blogs
+£2519 (LSP +44.6pts)
Click here to CHECK IT OUT

[man o bong]

Quick 2008 update. We are actually losing thus far in 2008, but have had just the 2 runners so far.

0-2 then. The average price of these two losers is 4/1. The average price of winners in the system is 2.77/1 and losers 3.11/1. Make of that what you will.

But the real reason for the post is to highlight that e have a qualifying race tomorrow (15th Jan).

1.40 Kempton.

At this stage it is not clear who will be favourite.

RUWAIN is currently 4.3 on Betfair but close behind is GRANARY GIRL 4.5 and SAHF LONDONat 5.1. So anything could happen there.

Racing post tissue gives us the three same but in different order

GRANARY GIRL 3/1, SAHF LONDON 7/2 & RUWAIN 4/1.

No spotlightwrite up at time of writing but we know that this opinion can have an affect of 3% or so on the prices so will be interesting to see what they pick.

In any case it is still prudent to follow the fav, clear or joint, or maybe even CO here, for the system at SP. So possibly a tricky one to call, best just before the off.

We shall wait and see, and fingers crossed for the first winner for this system of the year.
_________________
6 Months Horse Racing Blogs
+£13,376 (LSP +113.5pts)
6 Months Football Blogs
+£2519 (LSP +44.6pts)
Click here to CHECK IT OUT

[The Shark]

Hi man o bong - I read your posts with much interest - thanks for the above.

I too am a "systemite" - I confess!

What I'm struggling to understand with this one is the basis behind the system.

With your Cheltenham National Hunt flat system - the way I see it is - it identifies potentially high class horses (have enough ability to be considered by their trainers worthy of running in such a competitive race) when dropped to very low class races (National hunt maidens or novice hurdles).

If you had a system that say identified top weights in 2 year old handicaps (nurseries) when xx lb clear of the 2nd rated for exa,mple - that too would be a system to identify class.

Does this system identify when the market gets it right

Or does it identify when the market gets it wrong (able to blindly back the favs and make a profit - therefore favourites starting at higher odds than they should - therefore the market isnt supporting them as much as they should)

My thought is the latter - but this contradicts your thoughts on it becoming easier to predict the AW.

The question I would ask is why does the market allow favs to go off too long at Kempton and Wolves in 9/10f races with 10 or more runners in class 4 to 6 and not (say) allow favs to go off too long in other types of races. Are these races harder to predict (not easier)?

If we were looking to prove the point that the AW is becoming easier to predict then wouldnt we be looking at areas where the market gets it right - and the winning favs go off too short?

[man o bong]

darn you pesky kids!!!!!! (scooby doo circa 70's, when the baddie gets foiled)

I see what your saying and had not really thought about it like that. In fact i think if the title was different, then the question may not have arisen.

In essence i guess both arguments are correct in as much as the favs are favs, but maybe are longer priced favs than they should be.

its a bit of a crystal maze question really, and one i'm gonna have to sit and think about for a minute before answering any further,

Just thought about it some more, and its a bit of both isnt it?

Anyways, its working for now, and a longer,bigger sample of results may even come up with something more decisive.

I'm working on it Shark. Thanx for your reply
_________________
6 Months Horse Racing Blogs
+£13,376 (LSP +113.5pts)
6 Months Football Blogs
+£2519 (LSP +44.6pts)
Click here to CHECK IT OUT

[man o bong]

Quickly add that SAHF LONDON went off fav and duly obliged at 5/2 today.
_________________
6 Months Horse Racing Blogs
+£13,376 (LSP +113.5pts)
6 Months Football Blogs
+£2519 (LSP +44.6pts)
Click here to CHECK IT OUT

[The Shark]

Another great winner for the system - nice one..

Im not knocking it - looks a great system - just trying to assess what we can learn from it.

Rather than backing systems blindly I prefer to try and learn from them - if we can learn why this is chruning out good price winners then maybe we can find some other golden egg laying geese as a result

[man o bong]

Tell ya what mate, if ya want a spreadsheet with all the results to analysie furtehr i can send ya a copy?

Its pretty comprehensive.
_________________
6 Months Horse Racing Blogs
+£13,376 (LSP +113.5pts)
6 Months Football Blogs
+£2519 (LSP +44.6pts)
Click here to CHECK IT OUT

[man o bong]

Results for 2008 thus far

1-3 (33%) +.50
_________________
6 Months Horse Racing Blogs
+£13,376 (LSP +113.5pts)
6 Months Football Blogs
+£2519 (LSP +44.6pts)
Click here to CHECK IT OUT

[man o bong]

Next qualifying race

17th Jan 08

7.20 Wolverhampton

SPITFIRE JANE looking like the fav for the race, and strong at this stage. Watch the markets tomorrow, and we'll see how we get on, hoping for a follow up to the last and get the roll of winners under way with our second in a week.
_________________
6 Months Horse Racing Blogs
+£13,376 (LSP +113.5pts)
6 Months Football Blogs
+£2519 (LSP +44.6pts)
Click here to CHECK IT OUT

[Nobby Luton Outlaws]

This is a red hot thread, cheers.

Edit: A SSN has tipped Spitfire Jane! Not a good sign! But I guess she is due a winner after picking loser day after day!

[man o bong]

Spitfire Jane now declareed as a non runner.

Onus is now onto TAPAS LAD or HAWA KHANA, look fairly close in the market right now, so will have to wait till race time to see who goes off fav.
_________________
6 Months Horse Racing Blogs
+£13,376 (LSP +113.5pts)
6 Months Football Blogs
+£2519 (LSP +44.6pts)
Click here to CHECK IT OUT

[The Shark]

man o bong - have you tried running through your little machine the fate of second favs in the same races?