PDA

View Full Version : TSN Data Files. How do I get them?


thoroughbred
06-18-2004, 02:22 PM
A number of people have asked me whether CompuTrak can use TSN data files. Since it was designed for Bloodstock, comma delimited, single file input, I don't know the answer and would like to test it out.

Where, and how does one obtain TSN data files?

Thanks

cj
06-18-2004, 02:28 PM
http://www.tsnhorse.com

Looks a lot like Bris ;)

thoroughbred
06-18-2004, 03:30 PM
Originally posted by cjmilkowski
http://www.tsnhorse.com

Looks a lot like Bris ;)

Thank you. One of you was gracious enough to email a sample copy of a TSN file. I tried it on CompuTrak, and it works just fine.
It must be that the file structure of the TSN files is identical to that of Bloodstock.

Thanks again; I've learned something valuable.

BillW
06-18-2004, 05:09 PM
Originally posted by thoroughbred
Thank you. One of you was gracious enough to email a sample copy of a TSN file. I tried it on CompuTrak, and it works just fine.
It must be that the file structure of the TSN files is identical to that of Bloodstock.

Thanks again; I've learned something valuable.

Rube,

Both formats are the same. The difference is that the exclusively DRF data (SR/TV among other fields) are blank in the TSN files and likewise the exclusively Equibase data in the Bris files.

BTW the format specs are on the respective websites.

Bill

Tom
06-18-2004, 08:11 PM
Just don't mix the two if you plan to expost data or maintain a database wtih pace and speed figs-they are not equal. But for just using a data file for the day's card, you are golden.

Hosshead
06-19-2004, 02:15 AM
Track Variants are mentioned (on CompuTrak site) as part of the equation CompuTrak uses. (to be as good as it is said to be.)
It may run on TSN files, but it can't use that part of the equation, as there are no TV's in TSN.

Therefore, CAN CompuTrak perform as well without tv's, as it can WITH them in the equation? I'm sure the tv's were thought (and tested) to improve accuracy, or they would Not be part of the program.
This is the problem with using TV's in designing SW. It limits your data options, IF you want the Best Accuracy the sw can acheive.

The site says that tv's are used for internal fractions. Wouldn't that be part of some of the major functions of the sw? "Friction"?

Maybe the sw is so great that it can sacrifice some accuracy, and still show a +ROI. ??

thoroughbred
06-19-2004, 02:54 AM
Originally posted by Hosshead
Track Variants are mentioned (on CompuTrak site) as part of the equation CompuTrak uses. (to be as good as it is said to be.)
It may run on TSN files, but it can't use that part of the equation, as there are no TV's in TSN.

Therefore, CAN CompuTrak perform as well without tv's, as it can WITH them in the equation? I'm sure the tv's were thought (and tested) to improve accuracy, or they would Not be part of the program.
This is the problem with using TV's in designing SW. It limits your data options, IF you want the Best Accuracy the sw can acheive.

The site says that tv's are used for internal fractions. Wouldn't that be part of some of the major functions of the sw? "Friction"?

Maybe the sw is so great that it can sacrifice some accuracy, and still show a +ROI. ??

Hosshead,

This subject is extremely interesting. I will let all know the results of comparisons of the outputs using Bloodstock and TSN inputs.

Here's the crux of the track variant situation. CompuTrak, which as you know, was designed for the Bloodstock inputs, back calculates the variant from the Bloodstock Speed Rating.

When I ran the one TSN case, the output column which normally shows the Bloodstock Speed Rating, showed reasonable speed rating numbers. I need to find out if those numbers, coming from the TSN data fields, are equivalent to the Bloodstock Speed Ratings. If they are, then there is a variant that is calculated which would be accurate. If they are not, I will have to try to find out how TSN makes the calculation that appears as a Speed Rating, and, if possible, do a conversion between the numbers.

But, in the final analysis, if it doesn't work out with TSN, there is always Bloodstock to use and have everything be just fine.

I really appreciated your insight on this. It is exactly right.

Hosshead
06-19-2004, 03:34 AM
Thoroughbred, As you may know, the reason many people want to use Procaps TSN, is the price. At $60/mo. all tracks incl., it's affordable.

thoroughbred
06-19-2004, 11:29 AM
As you know, I want to compare running CompuTrak using the Bloodstock, comma delimited, single file format, (the file that CompuTrak was designed for), to using an equivalent TSN file.

I just downloaded a TSN file for HOL today. It's file name, (after unzipping), is: HOL0619.EG.

But when I was sent a sample TSN file from one of you the other day, the extension was ".PCS"

So, I have a question. What is the difference between TSN files that have the ".EG" extension and the ones with ".PCS"?

Also, where on the TSN site can I get a ".PCS" file?

The TSN site I am using is:

http://www.tsnhorse.com/

Thanks.

BillW
06-19-2004, 11:56 AM
The .pcs is a "Procap" file equivalent to the Bris "Multicap" file. These are "premium" files in that they contain a few extra bits of data over the standard data files. They are a strict superset so they should be benign in your application.

Procaps & Multicaps are free handicapping programs offered by Bris/TSN.



Bill

thoroughbred
06-19-2004, 03:21 PM
I downloaded today's (Saturday's) HOL data from both Bloodstock and TSN then ran CompuTrak using both.

While much of the output is identical, some of the important outputs do differ somewhat.

The relative values do seem to be consistent between the two, e.g., the highest and lowest values of each output appear at the same places.

For now, the thing to do is to continue to use the Bloodstock files, because CompuTrak was designed for them, and the results have been thoroughly tested over many years.

Thanks to all of you for your interest.