OZmium Sports Betting and Horse Racing Forums

Quote:

Originally Posted by Dirk Gently

Tailwag, I also recognise that the changing state of the punting market will render once profitable systems useless. My thoughts at the moment are concerned with the information you CAN'T get off a PC as I think there is a new legion on smarty pants punters who think if they download enough they will turn a profit at punting.

Yes Dirk, I am with you there. I also believe that data obtained from non-internet sources are invaluable. I am not referring to newspapers or subscriptions to magazines or TV either. I am referring to other forms of information which is free providing you go where it exists i.e. the racetrack for one, a trainers complex for two and so on. Can't give up all my secrets, but you are certainly on the right track by incorporating data from non-traditional sources. These sources, in their own right, are as valid a part of the racing industry as any other aspect.

I know this is so corny to say, but the sum of all the parts is greater than the whole, meaning that ANY part is a valid part than can benefit the whole system. Even a sublime issue like trainer A differs from trainer B in certain aspects. These aspects can influence the result of a race and inturn, this directly impacts on you!!!

Now I hear a lot of groans out there saying you can't interview every trainer re every training aspect, what are you on about :-) Of course I agree with that, but there are certain constants that are adopted almost universally, knowing them is useful.

From a racetrack point of view, the heart and soul of any system because this is the domain of the industry you are trying to analyse, supplies an abundance of information. The local feel, transactions, colour, coat pullers, rumours, way they 'Vigorously' whipped the horse in the straight and a vast number of other factors can be gleaned on a regular basis by visiting the racetrack. Most of all you can see the real volume of money supporting horses, which is a better gauge than reading the starting price in the next days newspaper.

These factors can be quantified and entered into your algorithms and hey presto, its free and enhances your static, standard limited data collection from traditional sources i.e. Net, newspaper, etc. Any system is only as good as the data that goes into it. Here I go with another saying...sorry....'Garbage in - Garbage out'. This is true for a racing system or a cinema program, so if you have the most diverse forms of data, given that their is relevance, you can filter for a better result.

I sound like I know everything and can tell you I don't, I am only trying to share what I believe. I believe that a racing system should follow best system practices, just like any other system, coin collecting, video cataloguing, weather prediction. The system output should be clearly defined, and I doubt that many are. I read what other listers write and they have a collection of constraints that need to be met to yield a profit. For example, no 2 year olds and so forth, this to my mind is not a system, but a method of culling a large list into a smaller list.

My ideal system would evaluate every runner and compare them against each other and against a predefined ideal which would take into account every possible variable, such as weather, track condition, time of year, class of race, weight and so on. There are literally hundreds of possible variables from the numerous sources. Some of which would be static (known in advance or sourced from a historical standpoint) and dynamic sourced from a live stream of data either electronically or in human form on a track.

With very powerful computing power today in ever decreasing size, you can setup a virtual office anywhere and have real processing clout for a very small capital outlay. Most people (well some people) bet more in a single bet than the price of a good laptop :-)

So, the very first question to ask is, what do I want my system to provide (output), then what data do I need to supply (input), to get my desired result. The desired result is usually the grey area, most people seem to want a short list of horses at backable odds so they can have a bet. I however want my system to rate every horse and show me where the value lies, and in this day and age, it might not be in finding a winner, or placegetter, in fact there could be more value in knowing a horse that is legless and simply can't win.

I feel that in the future the best value might come from correctly identifying the 4th, 5th and 6th best horse in any given race. I can't go into why I believe that, but switched on people will probably agree that, that is where the greatest value may lie.

Okay, I'm off to chew my masters slippers (woof)...

Tailwag