BRIS speed ratings

[Tom]

What is the value of a point on the BRIS speed rating scale?

I looked at lots of old threads all morning, and came to the conclusion...who knows?

So looking at current data, consider the BRIS speed by Circuit for this week at Monmouth......

9/13 - two 1m16 races, #1 and #4 and two 6.0F races, #3 and #8
9/14 - four 6.0F races, #1, #3, #5, and #9

I plotted the SR point s and raw times and came up with a point at 1m16 is equal to .19 seconds, essentially a 1/5 of a second.
At 6.0F, a point was worth .131, .134, or .135, depending on how I treated the data from 9/14. On that day, races #1 and #5 did not line up on the trend line, while races #3 and #9 did, so it was a split variant.

Whichever, here is the data.....what do you think a point of SR is worth?

[Tom]

Here is what I want to do -

Two pace lines:

6F....110.2....92
6F....111.1....98

The second race was the faster race, but by how much?
how many tenths of a second is that 6 points worth?
What is the adjusted time of the first race? Is is 112.0?

[Clocker]

HOW MANY SPEED POINTS REPRESENT ONE LENGTH ?
The BRIS Speed Ratings points-per-length scale is based on the long accepted premise that as distances increase the value of a length decreases. On the BRIS Speed Ratings scale one length is approximately equal to 1 1/2 pts in sprints and approximately 1 point in routes.

Source: BRIS (http://www.brisnet.com/cgi-bin/static.cgi?page=speed)

[Tom]

That is lengths - I want time - how many points is 1/10 second worth.

[Speed Figure]

Would it not be 0.16 points per 1/10 of a second for 6 furlongs? so 6 points would be about 3.5 lengths.

[Clocker]

The times for each horse other than the leader at each call is an estimate based on the leader's time adjusted for beaten lengths. We don't know the actual times for horses other than the leader. If 1 length is assumed to be 1/5 second, and if 1 length is 1.5 Bris speed points, then 1/10 sec. equals 0.75 points.

If the numbers in Tom's second post represent the winning times and the winner's speed ratings, then the time difference is 4/5 seconds, or 4 lengths. And the 6 point difference in speed figs., at 1.5 points per length, would also be 4 lengths.

[Speed Figure]

I use the chart of 10/6 = 1.66 points per length for 6 furlongs.

[Tom]

I am lot looking at beaten lengths - time.
I want to adjust one pace line to another.
The 10/distance works for BLs, as does using that optin in the BRIS PP Generator - I substitute pace of race E1, E2, LP and SR for the race times.
But I want to compare final times of races.

[Billnewman]

I just ran some numbers on FL 1 mile and 40 yards and came up with one point of Bris speed # =.16 seconds i'll do a 6f after I help my boy with his geometry homework

[Light]

Here is what I want to do -

Two pace lines:

6F....110.2....92
6F....111.1....98

The second race was the faster race, but by how much?

I'm assuming these are final times not internal pace times.

Since the 2nd race is faster by 6 points, and 1 length = 1.66 points (at 6f) Then:
6 points/1.66 points =3.61 lengths difference between the 2 races.This is confirmed when multiplying 3.61 lengths * 1.66 =6 points.

how many tenths of a second is that 6 points worth?

1 speed points = 1/10 second = 1/2 length. 6 speed points = 6/10 seconds = 3 lengths

What is the adjusted time of the first race? Is is 112.0?

Adjusted first race time = 110.2 +6/10ths = 110.8 Tenths.

Don't forget that the SR incorporates the variant and will cause some discrepancies.

The races you posted could not have been from the same day on the same track at the same distance or else the first race would have a higher SR.

[Exotic1]

Tom,

Don't know if this helps.


From the archives...

http://216.92.33.211/forum/printthread.php?t=42306&pp=40



Tom 01-07-2008 03:22 PM
Been working on this and came up with some new values for a point of BRIS speed.
At 6 furlongs, I get 1 point = .12 seconds and at 1 mile, 1 point = .17
I had been using 0.1 for all distances, but got to many really odd times.
Theses two values are not really that constant either - lots of variation, or BRIS splits variants almost every other day.

I need to feed some more days in to my db to do the other distances.

[Tom]

Time and beaten lengths are not the same value.
I'm trying to figure out time.

And I can't find something repeatable.

[Robert Goren]

Time and beaten lengths are not the same value.
I'm trying to figure out time.

And I can't find something repeatable.If time was repeatable, every pool would be negative. You can spend the rest of your days trying to find a way to make time repeatable and you would die a broken man.

[Exotic1]

Time and beaten lengths are not the same value.
I'm trying to figure out time.

And I can't find something repeatable.

This sounds like time:

At 6 furlongs, I get 1 point = .12 seconds and at 1 mile, 1 point = .17

[Tom]

If time was repeatable, every pool would be negative. You can spend the rest of your days trying to find a way to make time repeatable and you would die a broken man.

I'm talking about how much a point of SR is worth. I sit .13, .16? .2?

[Clocker]

I'm talking about how much a point of SR is worth. I sit .13, .16? .2?

Isn't that going to change from day to day depending on the track variant? In your example, a 6f race at 1.11.1 had a speed figure of 98. That same final time on another day might get a speed figure of 96 if the track was faster that day.

[Tom]

No, it won't change.
A SR point is worth X
In Quirin, 1 point is 1/5 of a second for final time, 1/10 second is 1 point for pace calls.

This is what I want to fill in, so I can then see what each point equals.

[Light]

I'm talking about how much a point of SR is worth. I sit .13, .16? .2?

So you don't believe me when I said its worth .1 (one tenth of a second)? Why not?

[Tom]

Because when you look at the data, it doesn't work.
If you use .1 in my worksheet, which winners times and SR for the same day, the RED is what you get.
I wish it were .1 - it make life easier.

[cj]

Because when you look at the data, it doesn't work.
If you use .1 in my worksheet, which winners times and SR for the same day, the RED is what you get.
I wish it were .1 - it make life easier.


Based on those it is around .135 hundredths of a second per point at that distance. A few more points might narrow it down a little, but it won't matter much.

[Tom]

That is what I get, but on other days, I get different values.
I have to assume some are due to spit variants, but sometimes, if I get 3-4 races on the same card, the variation is a lot. Enough to put doubt in my mind.

[Light]

I have no idea what your chart means or where its from. But you asked how much 1 speed point is worth using Bris and if you look at tandem races in their PP's you will see over and over again that if a horse loses to another horse by 1/2 length, they get 1 point off their SR. And it has a value of .1. If they lose by 1 length they get 2 points off their SR's and it has a value of .2 tenths or 1/5. That's the facts, period. Whether or not its accurate or what you are trying to do, is a different issue. But your original question has been answered.

[Tom]

Light, you are talking about beaten lengths. I am talking about time.
If 112 gets a 96, what does112.1 get? 111.3? 109.4?

Race 1 today goes in 112. What SR does it get?
RAce 2 goes in 113. What SR does it get?

That chart is actual aces at Monmouth and the SR those race times earned.

[HUSKER55]

If I understand your problem right,...

You could make a raw time only parallel time chart. It is my understanding that published speed fig have several variances built in so I have no idea how you could determine that part but perhaps that would solve your problem.

just trying to help...

[Light]

Light, you are talking about beaten lengths. I am talking about time.
If 112 gets a 96, what does112.1 get? 111.3? 109.4?

I already equated beaten lengths,time and sr's used by Bris.

To answer your question above, The horse with a 112.1 (assuming that is tenths) will get a 95 if the 112 for that day at the same 6f distance got
a 96. I already said how and why twice. This isn't rocket science.

[Tom]

Light, look at that chart I put up - that is the SR scale, along with 4 6F races on the same day next to the SR that time got.

69.96 got a SR of 89.
71.26 got a SR of 79.

There are no beaten lengths.
These are two winners on the same day.
71.26-69.96 = 1.3, or 13 tenths.
But the difference in SR points is 10.

That contradicts your example, right?
I am not trying to be argumentative here. It is close to .1, but using .1 or .135 is a big difference - 35%.

[Tom]

Here a few more examples.
All of these 5 samples are from a single day of racing, at 6 furlongs.
The SR is first, the time rounded down to a 10th is next to it.
All are the winner's times and SR, no beaten length adjustments.

[TexasDolly]

Here a few more examples.
All of these 5 samples are from a single day of racing, at 6 furlongs.
The SR is first, the time rounded down to a 10th is next to it.
All are the winner's times and SR, no beaten length adjustments.

Hi Tom,
Just to make sure I understand ; the 1st race and the 4th races both went in 71.5 and the 1st had a SR of 87 and the 4th had
a SR of 97 ? What could be going on to justify a 10 point difference in your opinion ? I am puzzled frankly.
TD

[Light]

It's no surprise to me that the numbers don't come out as Bris would have you think. I was just explaining what they supposedly do.

I just randomly checked Aqu on 1/1/14. They had 3 6f races that day.

The first was in 12.03 and the winner got a 99

The next was in 13.8 and the winner got an 81. This is right on the money correct using 1 SR point for each .1 of time (compared to the first race).

The 3rd was in 11.88 and got a 96! A faster time than the first race but a lower SR!

Obviously they are doing something they're not telling. I have called them in the past on things like this and you get nowhere.

For me the figs do not have to be that precise, just in the ballpark. My main concern is will the horse put out an effort today or not because it makes the numbers irrelevant if he won't.

My most recent beef with Bris was their Class pars listed in the upper corner. It changes with time which can be as short as a week. Calling them was a dead end. So I had to determine for myself which of the various pars they list for a specific class is actually correct. The way I did this was amassing the data for each specific class at a specific track into an excel spreadsheet and finding out for myself.

[Tom]

Hi Tom,
Just to make sure I understand ; the 1st race and the 4th races both went in 71.5 and the 1st had a SR of 87 and the 4th had
a SR of 97 ? What could be going on to justify a 10 point difference in your opinion ? I am puzzled frankly.
TD

No, each set of numbers is one day - and the 6 furlong races I could find in my DB. Five different days. I have to use one day at a time because the variants will be different. I am only looking for the differences in that day to get the value of a point of SR. So the two races with the same time are from different days with different variants.

[Tom]

Thanks light.
I hadn't noticed the pars changing.

[Big Bill]

Tom,

Click here for a link to a 2005 thread (http://www.paceadvantage.com/forum/showthread.php?p=214439#post214439) that might contain the info you are looking for.

Big Bill

[Secretariat]

Maybe this is simplistic but the time difference between the two is .9. The point differential is 6 points. Therefore in this scenario the value of a Bris point is .9/6 or .15 seconds per Bris point.

Your question interests me though as Bris has spoken in generalities about this for years. For example does BOTH lengths and a fifth of a second equate to the formula (10/distance) for Bris points? Or is it only for lengths?

I have wanted to take a few Bris result charts run on the same day with their final lengths provided by Bris, and their reported horse Bris times, and see if the actual relationship of the (10/distance) formula is equal for a fifth of a second and a length universally. Has anyone done that?

[Secretariat]

Posting something from further investigation using the Bris PP Past Performance Generator.

A race in which the winner ran a 72.2 second 6F race
A horse 4.5 lengths back from the same race ran a 73.05 second 6F race.

Both of these are as displayed using the Bris PP Past Performance Generator not my own calculations.

The winner scored an 82 Bris. The horse 4.5 lengths back scored a 75 Bris.

A 7 point Bris differential. The difference in time was 73.05-72.2 or 0.85. If one takes the .85 time difference divided by the 7 point Bris differential one gets .85/7 or approximately .12 seconds per Bris point at this 6F difference.

My investigation was not based on a formula, but the actual reported numbers in the Bris PP Generator. However, the result kind of matches the chart posted before here on the board which rises .01 per increase in each half furlong. (so a 4.5F race would be .09 and a 12 f race would be .24)

5f = .10
5.5f= .11
6f = .12
6.5f= .13f
7f = .14
7.5f= .15
8f = .16
8.3f=.1667
8.5f=.17
9fr =.18
9.5f=.19
10f =.20

[Tom]

Hey Sec!
Where you been, guy?

Thanks for the input.

[Tom]

Seems to work on every race I have tried it out on so far. You get a 1 point error every so often, probably due to rounding the figs. But it does work.
As far as pace, 1 pace point = 1/10 of a second.

[Secretariat]

Was just working on the 2nd race from DEL on 07/11/15 which was a 6F race.

There were four horses who had run in the same race against each other on 06/27 again at 6F and at DEL.

Do Ya Wanna had a 72 Bris and was one length back in that race.
Ebony Two Star turned in a 69 Bris and was two lengths back.
Vanna Go's Point had a 66 Bris and was four lengths back
Frosty Whirl turned in a 56 Bris and was ten lengths back.

Now the speeds and lengths:

72-1
69-2
66-4
56-10

Using the top speed horse and the speed length differentials we get this.
0-0
3-1 or (3 Bris points behind and one length back of the 72 speed horse)
6-3 or (6 Bris points behind and 3 lengths back of the 72 speed horse)
16-9 or (16 Bris points behind and 9 lengths back of 72 speed horse)

So for the horse who had a 69 he lost "3" points per length
The horse who had the 66 lost (6/3) or "2" points per length
Lastly, the horse who had a 56 lost (16/9 )or "1.78" points per length.

This kind of wide discrepancy in their beaten lengths is frankly inexcusable. The place doesn't even post a beaten length chart, and I've been told by them that the formula of 10/(distance in furlongs) for beaten lengths is not correct. Apparently Churchill bought Bris and the communication between the old and the new is lacking. Very reticent at this point to rely too much on these figures.