Horse Racing Forum - PaceAdvantage.Com - Horse Racing Message Board

Go Back   Horse Racing Forum - PaceAdvantage.Com - Horse Racing Message Board > Thoroughbred Horse Racing Discussion > General Racing Discussion


Reply
 
Thread Tools Rate Thread
Old 08-20-2020, 06:44 PM   #31
MooseDog
Registered User
 
Join Date: Aug 2004
Location: Northern California
Posts: 166
First of all let me admit I'm a fan of using more technology in racing. I've been learning to use CJ's video techniques on many races and I have to say the accuracy is highly overstated. For verifying final times, yes, good. But the internal fractions are tough. The angles and which the horses pass the various poles is extremely challenging and the presence of large screens and trees only complicates things further. I will say my in trip handicapping game has improved a lot...but I digress.

It's really curious how Randy Moss has embarked on a Trumpian-level Tweetstorm against GPS timing (or Equibase, it's clear he's no fan).

Randy's pecking order is laughable. I challenge anyone to tell me where to manually start timing for 1M at LRL. Actually most chute starts are difficult but this one is extremely tricky.

Here's a screenshot of the start as the horses leave the gate. That pole to the far right is not the 1M marker, it's holding up the fence behind the gate. And the camera is over 1/4 mile away. That picture is extremely compressed.

The second picture is the first in the video feed showing the pan view. Can you tell me - are they past the 1M marker pole? How many feet out of the gate?

The marker is not in any frame of the video. Run-up is stated at 20ft. What do you do? One jump out of the gate? Two? Very scientific.

I know people who time workouts, and some of them have worked as backup timers. They all will admit to making guesses at these extreme angles. Turf chute starts are also problematic. They've tried to show me how to do this. It's very hard.

Those guys and gals are amazingly good at what they do but none of them would ever claim accuracy to the 1/10th all the time.

I'm only guessing that Trakus and Gmax have an imaginary "start line", with GPS for sure it should be easy to know exactly where that line should be drawn. I don't imagine they guess. Starting the time too early will result in a slower final time, starting too late will result in a faster one.

So someone please explain to me how hand/video timing is better in this case? Moss is making a blanket statement that GPS doesn't work. Maybe he means the Gmax system is flawed, yet in today's statement from Del Mar (quoting from the Paulick Report):

"The times produced by the Equibase GPS System for dirt races have proven to be highly accurate and will continue to be provided"

Attached Images
File Type: jpg Screen Shot 2020-08-20 at 2.46.32 PM.jpg (153.7 KB, 11 views)
File Type: jpg Screen Shot 2020-08-20 at 2.47.02 PM.jpg (150.9 KB, 9 views)
MooseDog is offline   Reply With Quote Reply
Old 08-20-2020, 07:46 PM   #32
steveb
Registered User
 
Join Date: Jan 2008
Location: Melbourne Australia
Posts: 900
Quote:
Originally Posted by Jeff P View Post

Don't know who Randy Moss is, but that is about the right order from my experience.
steveb is offline   Reply With Quote Reply
Old 08-20-2020, 10:57 PM   #33
cj
@TimeformUSfigs
 
cj's Avatar
 
Join Date: Jan 2002
Location: Moore, OK
Posts: 46,816
Quote:
Originally Posted by MooseDog View Post
First of all let me admit I'm a fan of using more technology in racing. I've been learning to use CJ's video techniques on many races and I have to say the accuracy is highly overstated. For verifying final times, yes, good. But the internal fractions are tough. The angles and which the horses pass the various poles is extremely challenging and the presence of large screens and trees only complicates things further. I will say my in trip handicapping game has improved a lot...but I digress.
Fractions are most certainly tougher, but with practice you can get just as good final time. You have to use races you trust are timed correctly and then use those as baselines. Once you build those baselines, it isn't that tough.

Quote:
Originally Posted by MooseDog View Post
It's really curious how Randy Moss has embarked on a Trumpian-level Tweetstorm against GPS timing (or Equibase, it's clear he's no fan).

Randy's pecking order is laughable. I challenge anyone to tell me where to manually start timing for 1M at LRL. Actually most chute starts are difficult but this one is extremely tricky.

Here's a screenshot of the start as the horses leave the gate. That pole to the far right is not the 1M marker, it's holding up the fence behind the gate. And the camera is over 1/4 mile away. That picture is extremely compressed.

The second picture is the first in the video feed showing the pan view. Can you tell me - are they past the 1M marker pole? How many feet out of the gate?

The marker is not in any frame of the video. Run-up is stated at 20ft. What do you do? One jump out of the gate? Two? Very scientific.
You can time these races with a great deal of accuracy if you believe the run up is correct. Particularly for the mile at Laurel, there is nowhere else to put the gate so it is always in the same place. It is true you can't really see when they pass the pole, but you can time from the gate. I do this all the time and it immediately shows you races that have a problem. If you get run up times of 1.48, 1.57, 1.43, 2.11, I guarantee you the 2.11 race is mistimed by a lot. What I do is just use the gate times for these races. If I'm going to time from the gate, I may as well use the best time possible, right?

Quote:
Originally Posted by MooseDog View Post
I know people who time workouts, and some of them have worked as backup timers. They all will admit to making guesses at these extreme angles. Turf chute starts are also problematic. They've tried to show me how to do this. It's very hard.

Those guys and gals are amazingly good at what they do but none of them would ever claim accuracy to the 1/10th all the time.

I'm only guessing that Trakus and Gmax have an imaginary "start line", with GPS for sure it should be easy to know exactly where that line should be drawn. I don't imagine they guess. Starting the time too early will result in a slower final time, starting too late will result in a faster one.
The GPS tracks were surveyed I'm sure and yes I'm also sure they have an imaginary start line. But that doesn't mean it is as accurate as the beam system at determining when the horses reach that start line. Much like Trakus, most of the problems I've found involve the start of the race. From the first call onwards I don't find near as many or as large of problems.

If workouts were on video and I cared to time them, I could absolutely get accurate times. It would take a lot of work but for sure it could be done. I'm pretty sure I'd rather light myself on fire than spend my day doing that though.

Quote:
Originally Posted by MooseDog View Post
So someone please explain to me how hand/video timing is better in this case? Moss is making a blanket statement that GPS doesn't work. Maybe he means the Gmax system is flawed, yet in today's statement from Del Mar (quoting from the Paulick Report):

"The times produced by the Equibase GPS System for dirt races have proven to be highly accurate and will continue to be provided"
Randy had thoughts on that. I'm not discussing GPS timing any longer, but I'm happy to talk about how I video time with anyone. And yes, for the vast majority of races I can easily get within 0.10 for the final time and any points of call in between if the race distance (including temp rails) has been run before and was accurately timed in the past. Here is the thing, even if you debate how accurate you can get from video, what video timing gives you is consistency. If I time three races from video at 6f on dirt and get 1:10, 1:11, and 1:12, I can 100% use that to make accurate speed figures. It doesn't matter if the real, perfectly accurate times are 1:10.10, 1:11.10, and 1:12.10. What I can't use is races reported as 1:10, 1:11, and 1:12 that are really 1:09.50, 1:11.50, and 1:11.50.
cj is offline   Reply With Quote Reply
Old 08-20-2020, 11:12 PM   #34
cj
@TimeformUSfigs
 
cj's Avatar
 
Join Date: Jan 2002
Location: Moore, OK
Posts: 46,816
Here is a good recent example from Gulfstream (NOT GMAX, Trakus). They use a head on shot of the mile dirt races. The gate is always in the same place. I can't see the pole for sure but I know I can time from the gate and know if there was a problem, which there often is.

8-16:

R3
Official Time 1:35.93
Gate Time 1:37.10
Run up Time 1.17 seconds

R7
Official Time 1:38.77
Gate Time 1:39.06
Run up Time 0.29 seconds

There is no way in hell there can be that much difference between run ups when they are basically starting half a stride out of the gate. I've seen worse. There was one from Thursday where the run up time was actually a negative time, the clock started before the gate even began to open.

But in any case, lets pretend we're making speed figures and that we're using crude pars of 80 for R3 and 60 for R7. These numbers really don't matter, we could use anything.

R3 Raw Beyer figure from official time: 104, Par 80, Variant 24
R7 Raw Beyer figure from official time: 74, Par 60, Variant 14

Again, very crudely average the variants together to get a 19 fast variant, it gives us figs of 85 for R3 and 55 for R7.

Now lets use the real times:

R3 Raw Beyer figure from video time: 91, Par 80, Variant 11
R7 Raw Beyer figure from video time: 71, Par 60, Variant 11

Average the variants together to get an 11 fast variant, it gives us figs of 80 for R3 and 60 for R7. What looked like a 30 point difference between the races was really 20, but even worse it misrated both races. This is why figure makers care about accurate times.

Funny note, I completely made those numbers up and didn't realize the video time figures would work out as neatly as they did. That really wasn't the point, I knew the difference would be reflected because of how far off the times were. The "official" gap is 2.84 seconds, but the real gap is only 1.96 seconds. That is quite a difference. And the problems usually don't just affect one race if not caught, they affect multiple races.

Last edited by cj; 08-20-2020 at 11:24 PM.
cj is offline   Reply With Quote Reply
Old 08-21-2020, 03:17 AM   #35
iamt
Registered User
 
Join Date: Sep 2014
Posts: 310
Quote:
Originally Posted by MooseDog View Post

I'm only guessing that Trakus and Gmax have an imaginary "start line", with GPS for sure it should be easy to know exactly where that line should be drawn. I don't imagine they guess. Starting the time too early will result in a slower final time, starting too late will result in a faster one.

So someone please explain to me how hand/video timing is better in this case? Moss is making a blanket statement that GPS doesn't work. Maybe he means the Gmax system is flawed, yet in today's statement from Del Mar (quoting from the Paulick Report):

"The times produced by the Equibase GPS System for dirt races have proven to be highly accurate and will continue to be provided"


I'll start by saying I'm a believer in idea of these systems, that if they can work accurately you can produce perfect times, point of calls and like other countries workouts. All of which has been promised to tracks when installed and advertised, but they just can't quite do what they claim.



The issue isn't inserting in a start line (or the line for any other point of call), it is a relatively simple thing to do, although for turf and rail placements they still seem to struggle. The issue with all of these systems is that they aren't very good at tracking/placing the horses to the level of precision they need to be, very often they're one or two lengths off.


For races with minimal run-up this can mean that a runner in the barriers is actually ahead of the start line while waiting for the gates to open, making accurate timing impossible.


Until they can actually place the horses close to perfectly, the systems are unreliable and not as advertised.
iamt is offline   Reply With Quote Reply
Old 08-21-2020, 10:53 AM   #36
classhandicapper
Registered User
 
classhandicapper's Avatar
 
Join Date: Mar 2005
Location: Queens, NY
Posts: 20,528
These and other problems with timing races and changes in track speed seem so widespread and significant, unless you have a staff of people checking this stuff all day, it seems like it makes sense to break races out for fairly consistent older horses more liberally than I used to think. As SKJ implied, you will "back into" the right answer more often even if the time is wrong.

I'm growing more skeptical every day of strong opinions about horses based on time differences that are usually less than the margin of error. That goes 100 times over if you are comparing horses from different eras. Doing that is going to be laughably wrong all the time.

I don't have a great solution. I've been wrestling with this for decades in my gambling. All the possible solutions have other problems. What I know for me personally is that it's way harder to bet a significant amount of money when I'm not even sure the figures I am looking at are correct. Ignorance is bliss.
__________________
"Unlearning is the highest form of learning"

Last edited by classhandicapper; 08-21-2020 at 10:59 AM.
classhandicapper is offline   Reply With Quote Reply
Old 08-21-2020, 04:12 PM   #37
TonyK@HSH
Registered User
 
Join Date: Apr 2001
Posts: 498
Quote:
Originally Posted by cj View Post
I've actually gotten a lot better at this since I made that. Previously it was easy to spot problem races using multiple races at the same distance, but getting accurate times from a single replay was much harder. I can do that now for most races.
Hey CJ,

First of all thanks for your hard work and contribution to this great game. We are lucky to have you.
I have a couple of questions regarding your findings in this area as they pertain to Timeform US.

1) Are you finding GPS timing discrepancies at all tracks using the GMAX system?
2) When you uncover a timing error with the GMAX tracks is this error corrected in the Equibase data?

I appreciate your feedback. I realize that you have no control of Equibase data.

Thanks

TonyK
TonyK@HSH is offline   Reply With Quote Reply
Old 08-21-2020, 04:23 PM   #38
cj
@TimeformUSfigs
 
cj's Avatar
 
Join Date: Jan 2002
Location: Moore, OK
Posts: 46,816
Quote:
Originally Posted by TonyK@HSH View Post
Hey CJ,

First of all thanks for your hard work and contribution to this great game. We are lucky to have you.
I have a couple of questions regarding your findings in this area as they pertain to Timeform US.

1) Are you finding GPS timing discrepancies at all tracks using the GMAX system?
2) When you uncover a timing error with the GMAX tracks is this error corrected in the Equibase data?

I appreciate your feedback. I realize that you have no control of Equibase data.

Thanks

TonyK
I'm working together with Equibase on this currently. We have a good relationship and I'll leave it there.
cj is offline   Reply With Quote Reply
Old 08-21-2020, 05:33 PM   #39
Tom
The Voice of Reason!
 
Tom's Avatar
 
Join Date: Mar 2001
Location: Canandaigua, New york
Posts: 112,470
Quote:
Originally Posted by MooseDog View Post
It's really curious how Randy Moss has embarked on a Trumpian-level Tweetstorm against GPS timing (or Equibase, it's clear he's no fan).
Nothing curious about it - HE is timing EVERY race that uses GMAX using video equipment using a method that has compared very closely to doing the same thing with races timed by a beam And he is seeing a third of the races are significantly OFF, either fast or slow. Not a fan, a figure maker who relies on accurate timing for accurate figs. Say Horse A and B run the exact same time on the exact same track an hour apart. A's race is off by +2/5 and B's race is off by - 2/5. Actual times were, say, 1.11

A gets a Beyer of 86 while B gets a Beyer of 98.

100% unacceptable.
__________________
Who does the Racing Form Detective like in this one?
Tom is online now   Reply With Quote Reply
Old 08-21-2020, 08:01 PM   #40
MooseDog
Registered User
 
Join Date: Aug 2004
Location: Northern California
Posts: 166
CJ - thanks much for the feedback. Practice, that's exactly what the clockers told me
MooseDog is offline   Reply With Quote Reply
Old 08-21-2020, 08:07 PM   #41
cj
@TimeformUSfigs
 
cj's Avatar
 
Join Date: Jan 2002
Location: Moore, OK
Posts: 46,816
Quote:
Originally Posted by MooseDog View Post
CJ - thanks much for the feedback. Practice, that's exactly what the clockers told me
For sure, I've learned plenty the past couple years timing from video. Lots of trial and error. And there are still some that are very hard to time for some of the reasons you mention. But you can still get reasonably close, certainly better than hand timing.
cj is offline   Reply With Quote Reply
Old 08-21-2020, 09:27 PM   #42
Psychotic Parakeet
Sartin Methodology Fan
 
Psychotic Parakeet's Avatar
 
Join Date: Mar 2007
Location: Earth
Posts: 328
Quote:
Originally Posted by cj View Post
Here is a good recent example from Gulfstream (NOT GMAX, Trakus). They use a head on shot of the mile dirt races. The gate is always in the same place. I can't see the pole for sure but I know I can time from the gate and know if there was a problem, which there often is.

8-16:

R3
Official Time 1:35.93
Gate Time 1:37.10
Run up Time 1.17 seconds

R7
Official Time 1:38.77
Gate Time 1:39.06
Run up Time 0.29 seconds

There is no way in hell there can be that much difference between run ups when they are basically starting half a stride out of the gate. I've seen worse. There was one from Thursday where the run up time was actually a negative time, the clock started before the gate even began to open.

But in any case, lets pretend we're making speed figures and that we're using crude pars of 80 for R3 and 60 for R7. These numbers really don't matter, we could use anything.

R3 Raw Beyer figure from official time: 104, Par 80, Variant 24
R7 Raw Beyer figure from official time: 74, Par 60, Variant 14

Again, very crudely average the variants together to get a 19 fast variant, it gives us figs of 85 for R3 and 55 for R7.

Now lets use the real times:

R3 Raw Beyer figure from video time: 91, Par 80, Variant 11
R7 Raw Beyer figure from video time: 71, Par 60, Variant 11

Average the variants together to get an 11 fast variant, it gives us figs of 80 for R3 and 60 for R7. What looked like a 30 point difference between the races was really 20, but even worse it misrated both races. This is why figure makers care about accurate times.

Funny note, I completely made those numbers up and didn't realize the video time figures would work out as neatly as they did. That really wasn't the point, I knew the difference would be reflected because of how far off the times were. The "official" gap is 2.84 seconds, but the real gap is only 1.96 seconds. That is quite a difference. And the problems usually don't just affect one race if not caught, they affect multiple races.
Things like this have made me put less-and-less value on times and speed figures. What an utter nightmare that this is not getting fixed.
__________________
"And there they go! It's Toupée going on ahead, Long Underwear has fallen behind, Toothpaste is being squeezed out on the rail as Banana joins the bunch, and Cabbage is trailing by a head."
Psychotic Parakeet is offline   Reply With Quote Reply
Old 08-22-2020, 09:17 AM   #43
classhandicapper
Registered User
 
classhandicapper's Avatar
 
Join Date: Mar 2005
Location: Queens, NY
Posts: 20,528
Quote:
Originally Posted by Psychotic Parakeet View Post
Things like this have made me put less-and-less value on times and speed figures. What an utter nightmare that this is not getting fixed.
Everyone knows how important the times of races are, but not everyone knows all the technical and subjective difficulties involved in making accurate figures.

That's why it has become a pet peeve of mine when people declare with confidence that "so and so" was better than "so and so" based primarily on figures. Put 4 brilliant figure makers into the same room and they won't agree about how fast a lot of horses ran yesterday. let alone 10 years apart.

IMO, it's more important to find the figure maker that thinks like you do about certain issues. CJ's thinking is most similar to mine because he incorporates pace into the process.
__________________
"Unlearning is the highest form of learning"
classhandicapper is offline   Reply With Quote Reply
Old 08-22-2020, 12:55 PM   #44
MooseDog
Registered User
 
Join Date: Aug 2004
Location: Northern California
Posts: 166
Flag Man?

The talk about start lines got me thinking, does anyone remember the use of the flag man, as in "the flag is up"?

Wondering the last time anyone remembers a track using one.

For you young-uns, back in the day a guy used to stand at what we have been calling the start line. As the last horse would load into the gate he would raise a brightly colored flag, at which point the announcer would say "the flag is up".

As the horses were released from the gate, the flag man would drop his flag quickly as the horses passed him, signaling the timer to start the clock or watch.
MooseDog is offline   Reply With Quote Reply
Old 08-22-2020, 01:10 PM   #45
Augenj
Top Horse Analytics
 
Augenj's Avatar
 
Join Date: Mar 2013
Location: Las Vegas, NV
Posts: 12,303
Quote:
Originally Posted by MooseDog View Post
The talk about start lines got me thinking, does anyone remember the use of the flag man, as in "the flag is up"?

Wondering the last time anyone remembers a track using one.

For you young-uns, back in the day a guy used to stand at what we have been calling the start line. As the last horse would load into the gate he would raise a brightly colored flag, at which point the announcer would say "the flag is up".

As the horses were released from the gate, the flag man would drop his flag quickly as the horses passed him, signaling the timer to start the clock or watch.
Hollywood Park in the 70's with Harry Henson saying "The flag is up", bright orange as I recall.

Last edited by Augenj; 08-22-2020 at 01:11 PM.
Augenj is online now   Reply With Quote Reply
Reply




Thread Tools
Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

» Advertisement
Powered by vBadvanced CMPS v3.2.3

All times are GMT -4. The time now is 09:26 PM.


Powered by vBulletin® Version 3.8.9
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright 1999 - 2023 -- PaceAdvantage.Com -- All Rights Reserved
We are a participant in the Amazon Services LLC Associates Program, an affiliate advertising program
designed to provide a means for us to earn fees by linking to Amazon.com and affiliated sites.