PDA

View Full Version : Info


UltimateBetter
02-13-2016, 08:43 AM
Anybody using homemade software

mikesal57
02-13-2016, 01:12 PM
Dude....start using "search" functions when your a newbie


http://www.paceadvantage.com/forum/showthread.php?t=125297

JJMartin
02-13-2016, 09:48 PM
Anybody using homemade software
Yes.

Speed Figure
02-14-2016, 02:12 AM
Yes, any questions?

UltimateBetter
02-14-2016, 10:37 AM
It's a simple question I plan to invest in software that a coder is planning to make and would like to get some ideas what kinda oh algorithim should be written.

UltimateBetter
02-14-2016, 10:39 AM
Dude Dude there is a part two on this question do your reading I plan to get a coder too write some homemade software need some ideas on algorithim.

raybo
02-14-2016, 11:06 AM
Hundreds of members here use home grown software. Algorithms, thousands of them. Good luck!

UltimateBetter
02-14-2016, 11:33 AM
Just need an idea don't have to share ant secrets lol...

hcap
02-14-2016, 01:18 PM
Just need an idea don't have to share ant secrets lol...Here is a not so secret secret.
Learn to handicap.

Southbaygent
02-14-2016, 01:19 PM
I have thousands of lines of code I've written. Vast portion of it has to do with:

1) where do I access data and in what format - entries, charts, trn/jock stats, etc
2) how to I import or use the data above in a spreadsheet/database?
3) what kind of historical review can I perform and report on (oh yeah report another huge hunk of code) on said data

Then you can start working on algorithms to work the data for handicapping future races.

whodoyoulike
02-14-2016, 04:40 PM
Yes, I do. Designed and programmed it myself.

Is this coder designing the program based on your algorithms?

Otherwise, you really need to consider hcap's advice.

raybo
02-14-2016, 05:05 PM
Just need an idea don't have to share ant secrets lol...

I wasn't being flippant about all the members here using home made software, nor the fact that there are, literally, thousands of possible algorithms that may or may not apply to your own style of handicapping, and your own goals.

But, to say that you are going to hire a programmer to write some software for you, and at the same time be asking us for algorithms, means that you have not paid your dues yet. You must learn to handicap races first, then you can worry about computer algorithms. Now, if this programmer of yours is an accomplished handicapper already, then he likely has lots of algorithms, or many ideas for algorithms, already. But, even if he writes his own algorithms, they may not suit your needs or style of handicapping/playing.

If you want a good piece of software written, one that will work well for you, and you can't create it yourself, then you need to learn how to play the game properly first, before thinking about hiring someone to write software for you.

UltimateBetter
02-14-2016, 05:17 PM
Wow it was question and too my friend about handicapping I been playing this game for awhile and yes its about handicapping but a lot counts about ROI my friend cause in long run from a great friend handicapping is fine but even the best handicappers take a loss I'm so sure you know how to handicap but what I want to know is do Avg At 35 percent not.Thats why alot of these warez are get saving data to Get ROI and then when the true handicapping happens . Intended to Hcp

whodoyoulike
02-14-2016, 07:04 PM
I'm sorry, I didn't understand any part of your last post.

Btw, I always attempt to help supposed newbies when I'm able because I still remember how frustrating it is to navigate the handicapping pitfalls. And, back when I started there really were few places to turn for assistance but don't expect being given anything in other words I'm willing to point you in another direction if I've encountered similar problems. And, you'd be surprised how often we go in the same or a certain direction in this game it's just at different times.

Speed Figure
02-15-2016, 03:39 AM
I think the 1st thing you need to know is 1. What's my style of handicapping? 2. How much $ do I want to spend? 3. What do I want the program to do? 4. What data do I want to use? 5. Once you start the more things you will want to add & that's going to up the price! custom programs are not cheap. There are many on this board that claim to have homegrown software, but they never ever show it!

UltimateBetter
02-15-2016, 09:53 AM
That's all I wanted to know but HCP comes out blasting and posting and telling me to start handicapping I just want an idea of what principles I can start thank you for your time.

raybo
02-15-2016, 10:11 AM
That's all I wanted to know but HCP comes out blasting and posting and telling me to start handicapping I just want an idea of what principles I can start thank you for your time.

Hcap wasn't "blasting". You asked for algorithms, but if your handicapping is up to snuff you will already have plenty of ideas for algorithms. Since you were asking, he assumed, as others here did also, that your handicapping is sub par.

UltimateBetter
02-15-2016, 01:05 PM
So it assumption so it's to say something about handicapping and not anybody else I never questioned anbodys handicapping skills my friend all I want is some idea of algorithms that's all wow

Speed Figure
02-15-2016, 01:37 PM
The idea's should come from your handicapping. No one is going to give you algorithms. Take a look at different programs to see what they include and start building from there. You sound like you don't have any idea of what your looking for and that's going to make the process even harder. You have to know your handicapping style and put that style into a software program. Have you looked around at the different programs on the market that may be on your style of handicapping? do you even have your own handicapping approach?

UltimateBetter
02-15-2016, 01:42 PM
I've Beaver Yeast and Contenders

raybo
02-15-2016, 02:02 PM
So it assumption so it's to say something about handicapping and not anybody else I never questioned anbodys handicapping skills my friend all I want is some idea of algorithms that's all wow

Ok, one last response and I'm out of here.

Creating handicapping software is all about your handicapping skills and your handicapping style and goals. Nobody can create that for you, unless you have a real idea of what should be included in the program, which means, what you want included in the algorithms, based on the way you handicap and the level of your handicapping skill. Algorithms just put your ideas to work, nothing more.

What any of us want in a program probably is not what you want in your program. If you are a skilled handicapper, you will already know what data should be included in the program, and what the algorithms should address and accomplish.

Nobody is trying to demean you, or your handicapping skill, but the fact that you asked for algorithms tells everyone here that you don't even know what you want a program to do, or how to do it. Take that as you wish. How you interpret what we have said is up to you. If you want to get offended and defensive by what we have stated, then that's your choice, but that doesn't mean that any of us offended you, you just decided to take it that way, yourself.

Good luck with your endeavor.

whodoyoulike
02-15-2016, 03:38 PM
I didn't mean to offend you with either of my previous posts or even this one.

As I stated, I'd really like to help you if I'm able because I do remember the frustrations everyone probably feels starting out whether learning to handicap or designing a program. I just hope you will realize later that everyone responding was or will be willing to assist you (if they can). It's very difficult to convey thoughts and processes using this type of communication medium and realize that some people will just never understand.

So many people get discouraged attempting to understand this game and end up quitting which I hope doesn't happen with you. This "game" needs more interested people not fewer.

Experience really does count in handicapping.

UltimateBetter
02-15-2016, 04:08 PM
Nicely put sir I inbox you and Raybo that remark learn how to handicap is sarcastic.But any ways my question has been answered by you and thank you for your time.

ranchwest
03-12-2016, 12:03 PM
As a retired professional programmer, I can tell you that if you do not have a good idea of what you want the programmer to do, you better have deep pockets.

UltimateBetter
03-12-2016, 03:20 PM
Money has never been an issue thank god.

kingfin66
03-12-2016, 05:11 PM
As a retired professional programmer, I can tell you that if you do not have a good idea of what you want the programmer to do, you better have deep pockets.

I don't recall you having posted here for quite a while. I hope you are well.

traynor
03-13-2016, 02:01 PM
Wow it was question and too my friend about handicapping I been playing this game for awhile and yes its about handicapping but a lot counts about ROI my friend cause in long run from a great friend handicapping is fine but even the best handicappers take a loss I'm so sure you know how to handicap but what I want to know is do Avg At 35 percent not.Thats why alot of these warez are get saving data to Get ROI and then when the true handicapping happens . Intended to Hcp

It sounds like what you need/want/are looking for is a data mining app (not a handicapping app). Start with: This is what I want to know (the output).

Next step: This is what I have going in (the input/data source).

Then (and only then) do you need to think about the algorithms that generate the output from the input.

Most "conventional" approaches to handicapping are little more than a computerized version of paper-and-pencil methods.

You might look at the app delta lover has/had on the software thread (or computer thread, whatever). If I remember correctly, it was specifically for data mining/data analysis.

It is pointless to hire someone to write code unless you can give them explicit (and relatively precise) instructions on exactly what you want. Fishing expeditions in programming tend to be very expensive, while providing little of real value.

traynor
03-13-2016, 02:13 PM
It is pointless to have meticulously detailed algorithms that produce "sophisticated" pace and speed comparisons/values/rankings if those comparisons/values/rankings--on a really good day--fail to generate a consistent ROI that even comes close to breaking even.

Doing the same stuff that everyone else (or most everyone else) is doing will only get the same results everyone else (or almost everyone else) is getting.

Unless you can find a process that generates a healthy ROI over time--not just in a handful of sample races--it is probably not too useful to worry about writing code when the end result will be no more than a complex, pseudo-scientific way to lose money.

If you are making money with your handicapping method, you already know the algorithms needed. If you are not, it is the data mining side of the fence that is important, not the handicapping side.

traynor
03-13-2016, 02:15 PM
As a retired professional programmer, I can tell you that if you do not have a good idea of what you want the programmer to do, you better have deep pockets.

Absolute Truth!

green80
03-13-2016, 08:28 PM
make some software to design the algorithms

traynor
03-15-2016, 05:57 PM
make some software to design the algorithms


Uhhhh. Hardcore gotcha. This is such a basic step that most model makers (in race applications) don't even consider mentioning it. It is like saying "First, turn on the computer."

Thanks for pointing out something that is so obvious it is often overlooked in recommendations) and, hence, in developing strategies.

jasperson
03-17-2016, 09:04 AM
It's a simple question I plan to invest in software that a coder is planning to make and would like to get some ideas what kinda oh algorithim should be written.
I attached a explanation of a program I use that trifaca mike propose in post some time ago. I haven't fully exploited this program because I use it for another purpose. It is easy to add any handicapping factors that you would like. To change the weight of each factor you just change the co-efficient of the factor. If you set the co-efficient to zero that factor is not considered. That can be used to compute the effect of just one factor if you set all co-efficients to zero except one.

raybo
03-17-2016, 11:53 AM
I attached a explanation of a program I use that trifaca mike propose in post some time ago. I haven't fully exploited this program because I use it for another purpose. It is easy to add any handicapping factors that you would like. To change the weight of each factor you just change the co-efficient of the factor. If you set the co-efficient to zero that factor is not considered. That can be used to compute the effect of just one factor if you set all co-efficients to zero except one.

It's really easy to create a program that allows weighting of single or multiple factors. I have done it in Excel. The tougher part is being able to automate batch file importing and processing so testing of factor settings can be recorded for analysis (I've done that too but requires some degree of programming language usage).

jasperson
03-17-2016, 05:11 PM
[QUOTE=raybo]It's really easy to create a program that allows weighting of single or multiple factors. I have done it in Excel. The tougher part is being able to automate batch file importing and processing so testing of factor settings can be recorded for analysis (I've done that too but requires some degree of programming language usage).[/QUOTE
In my oddsline program I write a evaluation file thethat has all my factors in it. I then import it to foxpro database program and then enter in the payoffs. In foxpro I can do all most anything with the data. Run a single factor like speed in claiming races,speed in sprint claiming races, speed and early pace in claiming race. I had to write a foxpro program to do this but it wasn't hard. I could do the same with this combined program if I wanted to.

betovernetcapper
03-18-2016, 10:54 PM
Sign up for HTR, It's free. I don't use it but many good handicappers do. It costs $125 for a months data. Use it every day & look at the factors you find workable or interesting. If at the end of the month, you still want to reinvent the wheel, you'll at least know how you want it shaped. You can tell the programmer you want X Y & Z.
Hope this helps. :)

Southbaygent
03-19-2016, 02:35 AM
Sign up for HTR, It's free. I don't use it but many good handicappers do. It costs $125 for a months data. Use it every day & look at the factors you find workable or interesting. If at the end of the month, you still want to reinvent the wheel, you'll at least know how you want it shaped. You can tell the programmer you want X Y & Z.
Hope this helps. :)
Plus HTR has great export to csv/comma delimited files of nearly all its data, so if you want to RYO with the data is very doable...
I think it's $119/month...