- This topic has 39 replies, 15 voices, and was last updated 18 years ago by
.
- Posts
yes Carlisle – you could be right re-fitness. It is an important factor but assessing it is tricky. I’d say that a recent level of performance though would hint at a fitness ‘factor’. Howver rather than try to analyse everything your best chance of an edge, IMO, is more in-depth focus of a couple of factors which you have a feeling may be underbet. Deciding these factors is the key.
Carlisle, four of your five primary elements are included in my database ratings method but these are all minor factors in my book. The weightings I have been using over the last year or so have the raw ability of the horse accounting for 85% of the final rating for the race being assessed.
Thanks fellas
I have finally decided to start using Access2000 to hold my data. This thread isn’t about winning or finding an angle/edge. I have got plenty of them to bite on. The raw ingredients are my main concern.
Race Distance being a really knotty problem….. eg. 5 Furlongs @ Epsom on Firm going bares no comparison to 5 Furlongs @ Pontefract on the Heavy.
Any programmers out there? Lets lay the foundations and then go shopping for curtains….
byefrom
carlisleI wouldn’t trust my computer to make selections for me. It’s been processing horse-racing info for years and still just hasn’t got any gut-feeling for the game at all.
I used to allow it to print out my betting slips about 10-15 years ago, in fairness, it was quite proficient at that.
Race Distance being a really knotty problem….. eg. 5 Furlongs @ Epsom on Firm going bares no comparison to 5 Furlongs @ Pontefract on the Heavy.
You can compare the two runs but only as a guide (course/class/distance/conditions- pars) and use as "better than nothing" if no closer conditions have been recently run by this horse.
Backing two runners is the relentless pursuit of value. Backing each way is a shortcut to the poor house. Only 7% make a long term profit.
I give you a tip for free. Why on earth would you want to use Access2000? Get yourself something like MySQL as your back end data source. The likes of Access are only good for keeping dates of birthdays etc!
get your requirements sorted prior to tinkering with any database though..
Microsoft Access is an awkward piece of software, but once you get used to it it’s one hell of a tool. There’s also the added bonus that you don’t have to be a programming extraordinaire to start using the more advanced features (though having such ability is obviously useful, especially when getting into the realms of SQL and VB).
I’m assuming the man is going to keep data on race results etc…..therefore his datasource will soon expand to 10000s + rows. Access is a personal productivity tool and should not be used for analyzing data of that size and complexity.
Underscore
I focus solely on Flat handicaps, class 5 and above, and delete data over two years old. That leaves me with 20,000 or so rows with only 16 primary data cells each, which I handle comfortably within Excel, where my requirements are met by the formula function and a handful of macros. I can quite see that someone with more elaborate requirements might need Access or even MySQL. It is principally, as you say, a matter of being clear about one’s requirements.
An SQL based is a must for this type of project. Access would be a huge handicap. I have an SQL database with results for the last 12 years hosted on a dedicated web server and it is very fast. I like to have everything on a web server and do all work via a browser so I can access it from any web connection.
DaveJay is spot on but I suppose the easiest way to view past results and tinker with the analysis is to write the software in the first place.
As for the Access / MySQL argument – thats an easy one.
Microsoft upon launching SQL Server 2005 decided to give it away for FREE !
There are a couple of limitations but none that will even remotely hinder the requirements here being met. FYI these limitations are: –
No built-in management tools. Microsoft has a query tool called Express Manager that is available as a separate download, but as of this writing, it is an unsupported pre-release package.
* Supports only 1 CPU at a time, i.e. can’t run parallel queries.
* Does not allocate more than 1 GB of memory when executing queries.
* Maximum size of any one database cannot be > 4 GB.
* No analysis or reporting services. Data mining, Data Transformation Services (DTS) and reporting functions are not available for Express.
The loss of DTS (which enables you to easily import / export data to and from text files among other things) is the biggest hindrance but there are other ways to do this and, without wishing to sound rude, it’s not somethign most users would normally be aware of anyhow.
Quite simply if you use an RDBMS for storing anything you should be converting to SQL Server 2005 if you haven’t already.
P.S I have no affiliation with Microsoft at all BTW
Lee
Chester 15:15.
Compare recent results.
How will TEN DOWN’s recent a/w results pan-out on turf ?
This is NOT a tip. Just a question anyone playing in the race SHOULD know the ANSWER to.
goodbye from
quadrillaxxxxxxxxxxxxxxxxxxxxxxxxxx
Backing two runners is the relentless pursuit of value. Backing each way is a shortcut to the poor house. Only 7% make a long term profit.
The answer is –
Computer – a great chance on speed and form.
Non-Computer – turf from a/w, speed and form (and visa-versa) is either black or white.
Not today. This race did not carry any of my money – a few punters thought it has a chance tho’ and went against the crowd.
goodbye from
quadrillaxxxxxxxxxxxxxxxxxxxxxxxxxx
Backing two runners is the relentless pursuit of value. Backing each way is a shortcut to the poor house. Only 7% make a long term profit.
Hi gang
i don’t know much about Access2000 or SQL to really make a comment.
However i am not intending to do anything too complicated, i just want to hold my data and run a few filters/queries.My aim is to flag up the truely RELEVANT evidence. This would involve entering going, distance & course parameters. Enabling best speed & class figures to float to the top.
Yeah, i know computers can only be relied on as a work horse. The human brain must produce the interpretation of the key indicators, and the betting decisions.
I think of it as letting the picture evolve and take on increased levels of complexity. This is only possible, in my view, when the primary elements have been carved in stone (relative to one another)
Quadrilla i dont mix turf & aw form.
byefrom
carlislemarb, yes simple stuff for any reasonable commercial database. BUT not all commercial products are decent enough to enable this to be simple (if you know what I mean). Having a custom database is the real way forward. If you can code this or pay a developer you can have whatever you want.
Hi marb
your comments hit right at the main point, and in my view……
"There aren’t any worthwhile horse racing products on the market". My suspicion is that nobody wants to go to the trouble to offer one. Betting and betting products rely on vast amounts of LOSERS. Also any widespread winning system would quickly get most of the value squeezed out of it.
Therefore, in this thread, i am trying to see if anyone is interested in developing a baseline product that embraces the 5 primary elements.
With this in place more detailed layers could be bolted on.Hey….. we might have a truely useful piece of kit!
byefrom
carlisle
- You must be logged in to reply to this topic.