Trakus Beaten Lengths - Horse Racing Forum - PaceAdvantage.Com

Tom · 12-09-2009, 12:15 PM

Anyone ever look closely at the trakus charts at DelMar or Kee? (Not sure WO had them?)

Not too far into it yet, but I am finding the time for a BL is not consistent and wondered if anyone else has seen this. I've only looked a few races, but I've seen enough variation to make we really question it.
As an example, DelMar, R5, 9/9/09 Turf, 1m16....Victory Pete ran third timed in 141.06, one length ahead of Get Funky, times in 141.26, so 1 length = .20.
Get Funky is one length ahead of Buenos Dias, timed in 141.41, so 1 length here = .15.

I have so far seen 1 length = .15 up to .23....if I round all out to .20, then that says 1/5 = 1 length is correct, but isn't the generally dismissed as an inaccurate rule of thumb? So much for timing in hundredths?

Horseplayersbet.com · 12-09-2009, 12:20 PM

Quote:

Originally Posted by Tom

Anyone ever look closely at the trakus charts at DelMar or Kee? (Not sure WO had them?)

Not too far into it yet, but I am finding the time for a BL is not consistent and wondered if anyone else has seen this. I've only looked a few races, but I've seen enough variation to make we really question it.
As an example, DelMar, R5, 9/9/09 Turf, 1m16....Victory Pete ran third timed in 141.06, one length ahead of Get Funky, times in 141.26, so 1 length = .20.
Get Funky is one length ahead of Buenos Dias, timed in 141.41, so 1 length here = .15.

I have so far seen 1 length = .15 up to .23....if I round all out to .20, then that says 1/5 = 1 length is correct, but isn't the generally dismissed as an inaccurate rule of thumb? So much for timing in hundredths?

Maybe there was a strong wind at their backs in the stretch that day

Backstretch Pirate · 12-09-2009, 12:23 PM

As the horses approach the wire, their rates of decceleration will vary widely.
One horse may still be going reasonably well, while the other one is essentialy stopping dead.

Horseplayersbet.com · 12-09-2009, 12:47 PM

I just checked a race. Trakus has the second horse running faster than the winner:

Start 1/4 1/2 3/4 Mile Finish

Tap Show
1Neck
3:14.45

Lifes Not a Breeze
23/4
3:14.41

Include Us
3Head
3:14.61

Mack's Blackhawk
4Head
3:14.61

Paullinus
56 1/2
3:14.61

Criminal Power
64 1/2
3:15.88

Deputy Lor
71 3/4
3:16.77

Faye's Gray
81/4
3:17.15

Silent Knight
9Head
3:17.25

Regal Storm
101/2
3:17.26

Christo
114
3:17.34

Parabola
1218
3:18.29

Aggressive Joe
1317 1/2
3:22.78

King Cayenne
14
3:27.62
Race Notes: Win and Fourth Photo.

[yt=5th race WO Dec.6]3tB5OhahYBE[/yt]

Backstretch Pirate · 12-09-2009, 12:50 PM

Clearly then, trakus is out to lunch.

BillW · 12-09-2009, 12:54 PM

That's well within the error of the Trakus system.

Quote:

Originally Posted by Horseplayersbet.com

I just checked a race. Trakus has the second horse running faster than the winner:

Start 1/4 1/2 3/4 Mile Finish

Tap Show
1Neck
3:14.45

Lifes Not a Breeze
23/4
3:14.41

46zilzal · 12-09-2009, 01:16 PM

Quote:

Originally Posted by Horseplayersbet.com

I just checked a race. Trakus has the second horse running faster than the winner:

I any horse has a wider trip, and thereby covers more ground, there is a chance that it could run faster than another and yet be further back at the wire.

Horseplayersbet.com · 12-09-2009, 01:27 PM

Quote:

Originally Posted by 46zilzal

I any horse has a wider trip, and thereby covers more ground, there is a chance that it could run faster than another and yet be further back at the wire.

How exactly does that work? Trakus times aren't adjusted to distance covered by the horse are they? All the other times in the race are in the right order of finish.

Tom · 12-09-2009, 01:38 PM

That was my point - if time is not used to calculate the BL, then what is?

BillW · 12-09-2009, 01:50 PM

Quote:

Originally Posted by Horseplayersbet.com

How exactly does that work? Trakus times aren't adjusted to distance covered by the horse are they? All the other times in the race are in the right order of finish.

No they're not. You have to understand that Trakus has an error. It can't pinpoint a horse to the exact .01 inches. It's error is +/- 2 feet and is sampled every 1/30th of a second (as of the last time I talked to them - maybe 2 years ago) which adds additional error (the software can possibly compensate for the sampling error). In addition the position of the trakus device is variable, i.e. it is not glued on the point of the nose. For finishes, the photo is much more accurate. With Trakus it is entirely possible to show the order of any 2 horses reversed from reality.

BillW · 12-09-2009, 01:53 PM

Quote:

Originally Posted by Tom

That was my point - if time is not used to calculate the BL, then what is?

I think that the problem you are seeing is that the photo is used to calculate BL's (as it should) but not used to correct for the Trakus error. i.e. the Trakus data you are looking at is the raw data.

BTW, using Trakus for intermediate BL's should be more accurate than the chart caller because they don't have the benefit of a photo at each call.

Show Me the Wire · 12-09-2009, 02:02 PM

I would guess beaten lengths would vary according to ground loss related to the wideness from or nearness to the front horse.

kenwoodallpromos · 12-09-2009, 02:05 PM

1/5 or .20 seconds X 660 = 132 seconds, or 2:12.00 minutes
1 "length" (8') X 660 = 1 mile
No equine runs that slow.'
Do not believe any distance that is based on time. It is always inaccurate, even Trakus. Convert to your own ' per second based on inner rail distance. All the bumping in the race will not cause a horse to be pushed wide enough toactually lose much distance. If the horse goes further in the same race as an opponent it is the fault of the horse and/or connections. All you can do is measure the Trakus feet per second if you so wish.

DJofSD · 12-09-2009, 02:07 PM

Velocity (or speed, if you must) is the change in distance over a period of time:

v = delta distance / delta time, or v = dx/dt .

Algebra allows these to be written:

dx = v * dt .

Trakus is giving you the dt, delta time or the difference in time. Unless you know the value for v, velocity, you can not determine the change in distance.

As another poster has already pointed out, the runners are decelerating, that is, their velocity is changing, so, that value you want for v is going to be changing albeit by a small value.

Given enough Trakus data and a small enough difference in the time differences, you can make good estimations for the values.

For me, this represents the Holy Grail of pace handicapping and will allow expansion into other areas of investigation.

cj · 12-09-2009, 02:23 PM

I agree, this will be very good stuff once the chart callers become transcribers of Trakus info at all tracks.

However, there is no excuse to report times with the second finisher faster than the first. You have to build some error checking in to fix minor issues like this.