DRF Speed Ratings/Variant/BRIS Question

[podonne]

In my BRIS PPs (DRF Data Files) I see a few "speed figures" for each pp:

846- 855 BRIS Speed Rating NUMERIC 999 3
856- 865 DRF Speed Rating NUMERIC 999 3
866- 875 DRF Track Variant NUMERIC 99 2

Anyone know the details behind these? Is DRF Speed Rating the Beyer? They give you the track variant, is it already in the DRF Speed Rating or do you need to adjust the figure to account for it?

Thanks
Phil

File Format Description
http://www.brisnet.com/cgi-bin/static.cgi?page=drfsff

[Tom]

Hi Phil,
The BRIS speed rating is the large bold SR in the BRIS pps- it is no the Beyer which only appears in the DRF products.

The DRF SR and DRF TV are the "85-17" numbers you see in the racing form PPs.

The 85 is the raw SR and the 17 is the variant for that day. Add them together, use half the variant and add it to the SR, whatever - there are many ways to use it.

85-17 is "equal" to 90-12 in that both add up to 102.

[ranchwest]

I believe the DRF speed number is the number of lengths off of the best time in the last three years, with a length assumed to be equal to 1/5 second. At least that is the way it was computed in the past.

Before it was computed that way, it was computed against the track record.

[Tom]

Exactly, and the TV is calculated for sprints and for routes.

[podonne]

So could the rating or variant be negative?

[Greyfox]

Originally Posted by podonne

So could the rating or variant be negative?

No.

[Tom]

The variant can never go below 0.

However, the way you use it is to have some average TV number, say, 17, so that a TV of 23 would mean the track was slow 6, or slow 3 if you use half of the TV - I do it both ways. A TV of 11 would mean either Fast 6 or Fast 3, again, depending how you use it.

[podonne]

I don't understand, where do you get an "average track variant" to subtract the existing track variant from? Why wouldn't they just do this when they give you the figure, negative means slow, positive means fast, etc...

[Greyfox]

Originally Posted by podonne

I don't understand, where do you get an "average track variant" to subtract the existing track variant from? Why wouldn't they just do this when they give you the figure, negative means slow, positive means fast, etc...

I'm not sure that I get your question completely, but I'll take a stab.

On a given day a variety of races will be run. Some sprints, some routes and some on turf. To keep things simple, let's just do sprints.

Let's say that there are 5 sprint races.
And on that particular day, the average track variant is 15.
So for that particular day, the next time you see that horses running line,
in the DRF you will see the Track Speed + 15. (Of course in reality some of them might have been 12, some might have been 18, but the average for that day is 15.)

Now that variant will move about on a daily basis, dependent upon surface, quality of horses running (bigger purses = faster times) and so on. So on another day the average track variant for sprints might be say 18.

Chances are the track was, in general, slightly slower that day earning a higher variant. (Or it could have been cheaper horses, but we won't go there.)

And on it goes every day. Generally speaking for a fast course like Santa Anita that track variant will usually vary between 14 to 18 or so.
At other tracks it will go a lot higher.
If you want to individually compare one race to the three year time, you can make individual variants if you want. And many handicappers do.
They make their own.
Then some handicappers don't believe in using the variants at all.
They seem content to play from just raw times. (I don't.)

The reason they are not using +3 or -3 as you suggest, is you still have to know the variant for the day that they are running. In other words it is always + or - to something. You can't keep putting up plus and minuses unless you are also working out the average variant over the 3 year period as well. If you did this, every race line that the DRF put out would have to change every day as the average track variant would be changing, if only so slightly every day. This way, is cheaper, works well and they just report the variant average from that one fast time in the last 3 years.

Now I don't know if what I've said is of any help, but that's the best that I can say off the top of my head. Perhaps someone else can explain it clearer.

[socantra]

They could have done it many ways, but that is the way the DRF did it for many years, and the figure was put there to duplicate the DRF speed rating and track variant. That is why the BRIS single file data file is called a *.DRF file.

[Pell Mell]

Over the years I have tried numerous ways to use the DRF TV. I have done as Tom and used 1/2 and the whole number. Ray Talbout used the raw number and added 5 for a good track 10 for muddy or whatever, then Sartin came along and developed his method for the TV.

Whichever method one uses will average out to be just about equally effective. One day one thing works and the next day another. Talbout told me once that he only used SR as an indicater of condition but not as a tool of selection. Personally, I don't like to use any rating taken from an off track.

[Tom]

When I use the SR+TV, it is like salt in cooking - I use it various ways depending on the dish. Sometimes, I use half, sometimes the whole thin,g - all depends on the track nd what looks "right: at the time.
I couldn ever use it in a computer program, per se, because I am so inconsitent with it.

I use it two ways - to get contenders, and to adjsut pacelines for velocity calculations. As Brohamer explains in his book, when you use velocity, the effect of the variant is lessened on each fraction, so accuracy is not that imperative.
What I like about it is when I am on the road, I can pick up a form, take it to my motel, or airplane, and do races in my head, for entertainment, or even visit a track of betting parlor wherver I am, and actually win some races.

[podonne]

So if the track variant is the difference between the best time ever run (in the last three years) and the time that this race was run in, what happens when there is a new track record? Surely the track variant for that race would be zero (because the best time is = to this one, but the reference point has now been moved, so all the drf pps for that track I downloaded up to that point are now wrong?

[socantra]

Originally Posted by podonne

......... ..what happens when there is a new track record? Surely the track variant for that race would be zero (because the best time is = to this one, but the reference point has now been moved, so all the drf pps for that track I downloaded up to that point are now wrong?

I believe the three year best times are updated at the beginning of each year. Until that time, the speed rating would just show up as 100 plus a point for each fifth of a second by which the previous best time was topped.

The variant for that race might be zero, but the variant printed would be an average of the races in that category for that day (durt sprints, dirt routes, etc.)

[Greyfox]

Originally Posted by Tom

When I use the SR+TV, it is like salt in cooking - I use it various ways depending on the dish. Sometimes, I use half, sometimes the whole thin,g - all depends on the track nd what looks "right: at the time.
I couldn ever use it in a computer program, per se, because I am so inconsitent with it.

I use it two ways - to get contenders, and to adjsut pacelines for velocity calculations. As Brohamer explains in his book, when you use velocity, the effect of the variant is lessened on each fraction, so accuracy is not that imperative.

The SR+ TV,
You and I absolutely agree.

But when you say:
"I couldn't ever use it in a computer program,"...???
Why not?
(Your honesty on this board is profound.
I use that fig all the time in my algorithm's,
I suspect you do as well.) Probably a little bit I suspect.