System Research

[Benny]

Does anyone know of any Databases that I could use in researching a system that I have in mind.

Benny

[crash]

Sorry that no one has answered your query.
What you are looking for will cost you plenty and the alternative of putting your systems on here for those that do have the data to check the system out will also cost you [you loose exclusive use and have no control over what happens to it].

It is a hot topic as to whether systems follow past trends anyway, so you may as well run with your system for small amounts and see how you go. As long as all the rules have a strong handicapping ingredient [reason to be a rule in the first place] you can't do much damage.

Cheers.

<font size=-1>[ This Message was edited by: crash on 2003-12-21 13:48 ]</font>

[TheAvenger]

I have developed computer handicapping programs here in the US. Just pick a selective sample of races at selective tracks, which meet your needs, and bet them "on paper". I have done many theoretical studies in this manner. Good luck.

Glenn

[gumboot]

It depends on what data you are looking for.
I developed my system by going to the State library and copying all the form guides and results for main venue races on the East coast of Australia for a period of 3 years.
I then entered every bit of data about the horses (55 items per horse !)into a datbase and then developed a program to add a weighting to each bit of data and try to match the winners of the races...it took 2 years to do that !...when I finished I had a formula that actually worked in real time and I was able to cut the number of data items to just 8 per horse..my system still works 5 years later..it's the only way to do it uless you are prepared to pay big $ for someone elses work...good luck.

[crash]

Hi Gumboot,

So what are the 8 form facts just out of interest ?

Cheers.

[hermes]

I can't be more helpful either. I spent ages looking for such a database at a reasonable price. Costs plenty. Or it will cost you years in time to compile your own.

I once spent several months designing a software program that took in data and spat out selections. I added dozens of factors. It took forever to add the data and then, at the end of it all, the program usually selected....the favourite.

I wonder at the validity of testing ideas over 20,000 races anyway. My proceedure is to develop sound principles (strong handicapping ingredients) and test them on 150 recent races. Then batches of 50 races from previous years. If it holds up I test on at least 100 live races on paper. If it still holds, I start with small bets.

Its alarming the number of systems that work on recent and past races and then fail on live races.

In my own experience I find that bigger databases don't necessarily yield better results. The thing that helps is a better understanding of principles (handicapping rules) from the outset. For example, beginners at this will often look at win strike percentages and include that in their selection systems. But in fact, as experienced system developers know, place strike percentages are far more important and a more reliable guide to winning potential. No database however giagantic will turn win strike percentages into a meaningful statistic. If your theory is flawed a bigger database won't help.

Hermes

[gumboot]

Hi there Crash,
The 8 form items are :

1:Has it won at this distance
2:Has it won at this track
3:Is it a mudrunner
4:The rating (either from the form guide or from another good source.
5:Last race finishing position
6:2nd last race finishing position
7:3rd last race finishing position
8:4th last race finishing position

[lomaca]

Hi Benny!
I do have a database going back at least four years. It includes the form of horses, the race results (on the Vic TAB including the exotics, the pools and the price fluctuation like 15 minutes 10 minutes etc. as supplied by the TAB results). It is a huge database and it is in MSAccess format.
I do not know how much you know about databases and programming in general, so let me know what sort of a data you are looking for? I may be able to help you. We may come to an agreement!?

[crash]

Hi GB,

Very interesting final filters. You have the KISS principle [keep it simple stupid] well under control.
I am currently using a small system that I worked out that is based on almost the same principles your filters describe. I believe too many filters beyond these simple but powerful ones, can lead to erroneous results as can those back fitted over many years because the horses running now generally had nothing to do with them and the further back you go the worse it gets.

I don't bet beyond dead though and I am even wary about that [dead and good footing or dead and slippery ?]. I also have a pre-post price [using PP from: http://www.tabform.thewest.com.au/spcontent/TFFrameset.jsp ] filter of no more than $9 or must be in the first 4 favorites. I use price but first 4 favorites would do the same job of excluding unrealistic class jumpers and a last start filter of no more than 10 days since last start. These three filters add a powerful handicapping effect. Finally I 'do the form' on any runner my system selects looking for any obvious [only] reasons not to bet the selection.

The main thing I try to remember is that I am looking for consistent results that no, won't score the winner of every race but will consistently achieve acceptable POT return figures from flat stake win bets. I am not interested in big wins but have a bit of fun with long shot interest bets on the side from my own handicapping efforts using the Sportsman and $5 win bets only. Sportsman ratings are tried and true and the new Winning Post ones look good too. My selections usually fall in the top 4.

Cheers.


<font size=-1>[ This Message was edited by: crash on 2004-01-02 05:11 ]</font>

[darkydog2002]

GUMBOOT and CRASH.
Do you think you would improve your strike rate further by including on pace runners and field strength?.
Cheers.
darky.

[crash]

Darky.

No, it just complicates things for no practical reasons whatsoever [unless your system revolves around running style and field strength of course ]. Swings and slides come into play regarding filtering in general, and to achieve a better POT from a system using more filters can decrease the number of bets to the point that they no longer represent the majority of them.

With specific regard to running style and field strength, I'd point out that if a good system throws up a selection it is showing that it's running style and field strength are doing fine thank you, or it wouldn't be a selection.

The first question that affects filter numbers that I would ask myself would be: 'How many bets a week I am happy restricting myself too and especially on Saturday '?
The second question would be with regard to what $$$ amount per selection am I comfortable outlaying ? Too few bets would require too large an outlay for the same profit so I need a minimum of about 25 bets a week with about 6 or 7 on Saturday for the system to stay viable and maintain my interest. We all have to set our own comfort zones in this area however, so maybe more or even less filters would be required but avoid filters that do what another make unnecessary. An obvious eg. would be: 1. No maiden races. 2. Must have won it's last start. etc.

As I have my system now, if a selection makes it through to the betting stage it already has a top chance of winning with the filters I already have in place. The variables that are left are mostly in the hands of the gods or represent poor handicapping benefits anyway and further filters would only leave me wondering what day or perhaps even what week it will be before I can actually have a bet. We all know what happens when we put ourselves in that situation.

Cheers.



<font size=-1>[ This Message was edited by: crash on 2004-01-02 13:37 ]</font>

[gumboot]

On 2004-01-02 09:43, darkydog2002 wrote:
GUMBOOT and CRASH.
Do you think you would improve your strike rate further by including on pace runners and field strength?.
Cheers.
darky.


Hi Darkydog,
I figure that the rating given by whichever professional rater covers both the running style and the opposition horses' strengths.
And I'm not a morning person so the least info I have to come up with at 10 am the better .
Years ago I tried including every known variable..I ended up with 55 variables for each horse and just one race took an hour or so to evaluate...I'll stick with my 8 meager variables.

Cheers..Gumboot

[Shaun]

I use to be the same...had lots of variables took ages to work out a race....sometimes i came up with ripper prices but at what time cost....i started to think what are the main form factors to me the lasr 3 beaten margins are the first thing i look at...if you take the top 6 horses according to beaten margins and put them in order of lowest price to highest price you can find some good exotics then i like to add the class factor looking at what races they have been in lately and giveing them a value.....just with these 2 bits of info you can have a lot of fun

[Luckyboy]

Hi All,

Long time no post...

I have been using the following criteria as the basis of a system for about a year now with good results albeit reliant on the occasional 12/1 or 15/1 shot getting up. But it seems to do this regularly enough for flat stake betting. Here are the criteria:

1. Pre Post Price
2. Average Prizemoney
3. Racing Style (Leader, On Pace, Behind)
4. Sportsman Zipform
5. Weight
6. Barrier
7. Last Start Finish
8. 2nd Last Start Finish

I have found I can generally do 'the selections' in about an hour on Friday night for S, M, B, & A.

I then spend Saturday morning with a cup of coffee (right now) doing as 'crash' said "seeing why I shouldn't have a bet".

Good luck in 2004.


Lucky

[darkydog2002]

Hi Gumboot and Crash.
I agree .Too many variables and you exclude most horses.
For me I prefer to look at a couple of standouts in a 1200-1600 race in a smaller field.
That seems to work best for me.
Today for example there appears to be 1 race with 2 top chances .
I couldnt imagine myself betting 25 -30 horses a week but thats just me.Everyone to their own I guess.
Cheers.
Darky.

[gumboot]

Hi Darkydog,
I used to use the Sportsman zipratings too ,but the only newsagent for miles no longer stocks it..I was the only one who purchased it!
I've tried the Age newspaper ratings which is what I used to come up with the Moonee Valley picks for 3/1/04..then I thought better of it and used the tabracing ratings as I have been for a while now and came up with different picks.

Of the 6 picks I made using the Age ratings I got 2 winners from the 4 suitable races..2 were mares or filly races which I didn't notice at the time.
Using the tabracing ratings I got 3 winners from the 4 races I bet on.
Just goes to show..don't change a thing when it's working!.

But the Sportsman is so much better..if I can get it.

And Luckboy...I'm glad to see that someone else uses minimal info as well...I only use 3 of the same variables as you do but I will try out your 8 variables some time and see how the 2 lots compare over time.