Trakus

[hracingplyr]

How are some of the players on here using Trackus? I see how many feet further horses traveled in a race. For example a horse travels 30 more feet during the race, would you adjust his speed rating next race? If yes how many points?

[ultracapper]

What's 30 ft, about 2 strides? No adjustment.

[Tom]

That's 3 - 3.75 lengths.

[raybo]

I don't think one can use feet traveled, alone. If a horse travels further than others by running wider in the turns one should factor in the fact that it is usually easier to run a wider radius in turns, because of less centrifugal force. If a horse runs further without running wide in the turns (moving in and out for position, or varying the path in the straights) then that is excess energy expenditure.

[whodoyoulike]

How are some of the players on here using Trackus? I see how many feet further horses traveled in a race. For example a horse travels 30 more feet during the race, would you adjust his speed rating next race? If yes how many points?

This is a good question. Because I've found using the the data as provided difficult in it's format, I'm currently only using it in viewing the chicklets during a race.

I have used their data when they became available years ago (which I still use) in developing a beaten length adjustment for sprints and routes.

How do you use their data?

[Cratos]

This is a good question. Because I've found using the the data as provided difficult in it's format, I'm currently only using it in viewing the chicklets during a race.

I have used their data when they became available years ago (which I still use) in developing a beaten length adjustment for sprints and routes.

How do you use their data?
You are correct, Trakus data format leaves a lot to be desired.

[whodoyoulike]

You are correct, Trakus data format leaves a lot to be desired.

Btw, let the others figure the bl adjustment out on their own. I saw you once publish it in a response and it made me very uncomfortable. I don't mind if others come up with a figure as long as they've done the work there's no need to give it to them. And besides we may be wrong but your response did provide to me an independent reassurance of it's validity.

[raybo]

Btw, let the others figure the bl adjustment out on their own. I saw you once publish it and it made me very uncomfortable. I don't mind if others come up with a figure as long as they've done the work there's no need to give it to them. And besides we may be wrong.

Yeah, as long as they all keep using the same BL adjustment (time per length) for all races and times, I'll be happy. Static time per length is so wrong, it's simply absurd.

[whodoyoulike]

Yeah, as long as they all keep using the same BL adjustment (time per length) for all races and times, I'll be happy. Static time per length is so wrong, it's simply absurd.

That isn't the way I calculated it.

[Cratos]

I don't think one can use feet traveled, alone. If a horse travels further than others by running wider in the turns one should factor in the fact that it is usually easier to run a wider radius in turns, because of less centrifugal force. If a horse runs further without running wide in the turns (moving in and out for position, or varying the path in the straights) then that is excess energy expenditure.
Correct, feet travelled alone is a nebulous metric, but when integrated into the aerodynamic drag formula, the COF formula, the surface resistivity formula, and the surface wind resistance formula you will have a powerful result.

A good example is to take Honor Code's last 3 races using Trakus data and look at them normalized at the 1-1/8m distance. He ran on a 1 mile track at CD, a 1-1/2 mile track at BEL and a 1-!/8 mile track at SAR and if you normalize his performances using Trakus data you will find a very revealing conclusion.

Also in using Trakus data, you should use weather data at time of the race and track configuration data.

My point of view in using Trakus is that it is best used in a statistical model using Bayesian statistics.

[cj]

As of now Trakus is showing bubkus for the 9th at Saratoga today. It was only a G1 race for probably the best sprinter in the country. This happens more than people realize. Trakus is nice, but there are issues.

[raybo]

That isn't the way I calculated it.

Didn't say you did. I said that anyone using "static" beaten length multipliers, without regard for how fast the horses are traveling will most of the time adjust the raw time in error. The only times they will be right is in those instances when the horses happen to be traveling at a speed that corresponds to that static multiplier, and that would be pretty rare.

[Robert Fischer]

Trackus is an ingredient or a raw material or a commodity.

[whodoyoulike]

Trackus is an ingredient or a raw material or a commodity.

I think it's an end product which can also be used as an ingredient or a raw material if you think you know what you're doing.

[Cratos]

Trackus is an ingredient or a raw material or a commodity.
Trakus is what DRF should have evolve too, but they didn't.

[Cratos]

As of now Trakus is showing bubkus for the 9th at Saratoga today. It was only a G1 race for probably the best sprinter in the country. This happens more than people realize. Trakus is nice, but there are issues.

Yes, Trakus has "issues", but they are exponentially ahead of Equibase and DRF in giving time and distance for the horserace.

According to Equibase/DRF, American Pharoah lost by 3/4 length, but if you use the Trakus data and set 1 length to equal 8 feet; he would have loss by half as much.

[cj]

Yes, Trakus has "issues", but they are exponentially ahead of Equibase and DRF in giving time and distance for the horserace.

According to Equibase/DRF, American Pharoah lost by 3/4 length, but if you use the Trakus data and set 1 length to equal 8 feet; he would have loss by half as much.

I agree, but we can't have races with no data.

[Magister Ludi]

What's 30 ft, about 2 strides? No adjustment.
30ft =~ 1.5 strides =~ 3.75 lengths

[Tom]

According to Equibase/DRF, American Pharoah lost by 3/4 length, but if you use the Trakus data and set 1 length to equal 8 feet; he would have loss by half as much.

And if you look at the photo, clearly is was not a less than half a length.


Trakus is what DRF should have evolve too, but they didn't.
Trakus evolved into DRF, unfortunately.
.

[raybo]

And if you look at the photo, cleary is was not a less than half a length.

Yeah, and if you consider that AP is a very long horse, that 1/2 length would probably be 3/4 length for most other horses - LOL.

[Cratos]

30ft =~ 1.5 strides =~ 3.75 lengths
The Jockey Club estimates the average stride for a thoroughbred race horse to be approximately 25 feet and the British Horse Authority standardize a length at 8 feet.

[Robert Goren]

A length in horse racing is a unit of time and has been for ages. That unit is determined by the speed of photo finish camera or the timer used by Trakus. With the old photo finish cameras, a length was 0.164 part of second. This was a number given to me by photo finish camera operator in the late 1980s. I have no idea if that number is still accurate. A "length" is sort the reverse of "Light Year" which is a unit of distance.
Maybe British horses are not as long as American horses. 8 feet is pretty small for a mature thoroughbred. I would say 10 feet is closer to average. Maybe one of the trainers who post here will venture their opinion.

[raybo]

A length in horse racing is a unit of time and has been for ages. That unit is determined by the speed of photo finish camera or the timer used by Trakus. With the old photo finish cameras, a length was 0.164 part of second. This was a number given to me by photo finish camera operator in the late 1980s. I have no idea if that number is still accurate. A "length" is sort the reverse of "Light Year" which is a unit of distance.
Maybe British horses are not as long as American horses. 8 feet is pretty small for a mature thoroughbred. I would say 10 feet is closer to average. Maybe one of the trainers who post here will venture their opinion.

I use 9' in my raw time adjustment calculations, the average between 8' for shorter horses and 10 feet for longer horses. 0.164 seconds, or any other static measurement, is only accurate when compared to other horses that are traveling at the same rate of speed when the measurement is taken.

[Cratos]

A length in horse racing is a unit of time and has been for ages. That unit is determined by the speed of photo finish camera or the timer used by Trakus. With the old photo finish cameras, a length was 0.164 part of second. This was a number given to me by photo finish camera operator in the late 1980s. I have no idea if that number is still accurate. A "length" is sort the reverse of "Light Year" which is a unit of distance.
Maybe British horses are not as long as American horses. 8 feet is pretty small for a mature thoroughbred. I would say 10 feet is closer to average. Maybe one of the trainers who post here will venture their opinion.
Wrong, a length is a unit of DISTANCE not TIME and you can make it any distance you like. Yes, in the old days before high technology, horsemen would attempt to estimate the distance between horses with something called at the time a “horse length” and subsequently a length became synonymous with a one-fifth second measurement.

All of this became naught with Trakus introduction of the RDIF chip into racing. The problem with Trakus data is that racing (both fans and management) has not embraced it and hopefully the joint venture between Equibase and Statts will change that.

Incidentally, the length in horseracing and the unit of length in astronomy does the same thing: they measure distance.

[steveb]

Wrong, a length is a unit of DISTANCE not TIME and you can make it any distance you like. Yes, in the old days before high technology, horsemen would attempt to estimate the distance between horses with something called at the time a “horse length” and subsequently a length became synonymous with a one-fifth second measurement.

All of this became naught with Trakus introduction of the RDIF chip into racing. The problem with Trakus data is that racing (both fans and management) has not embraced it and hopefully the joint venture between Equibase and Statts will change that.

Incidentally, the length in horseracing and the unit of length in astronomy does the same thing: they measure distance.


he is not wrong.....YOU ARE.

[highnote]

he is not wrong.....YOU ARE.

I don't know if the following is relative to the conversation, but these thoughts occurred to me when I read this thread...

According to Einstein, Time is relative. It depends on the observer.

Einstein used an example like this to demonstrate relativity:

If a person stands on a train platform and watches a slow moving train travel by at, say, 2 miles per hour and there is a man walking on the train in the opposite direction at 2 miles per hour, the man walking on the train appears to be not moving to the person on the train platform. The man walking on the train platform sees himself moving at 2 miles per hour. Who is correct? They both are. It depends on the observer.

So in one sense a length is a unit of time and in another sense it is a unit of distance. It depends on how you are using it.

In horseracing, Beyer divides the distance of the race by the length of a horse to determine the value of a length.

But also, the time it takes to run one length is relevant.

[Cratos]

he is not wrong.....YOU ARE.
Prove how it is done; I am very open-minded.

If you want to say that a length is distance converted into time that make sense, but to say a length is time doesn’t make sense.

However before you go off on the deep end, can we agree that time is “the indefinite continued progress of existence “ and a length in horseracing is not indefinite, but you might know something that I am not aware of and I would appreciate your explanation.

[Cratos]

I don't know if the following is relative to the conversation, but these thoughts occurred to me when I read this thread...

According to Einstein, Time is relative. It depends on the observer.

Einstein used an example like this to demonstrate relativity:

If a person stands on a train platform and watches a slow moving train travel by at, say, 2 miles per hour and there is a man walking on the train in the opposite direction at 2 miles per hour, the man walking on the train appears to be not moving to the person on the train platform. The man walking on the train platform sees himself moving at 2 miles per hour. Who is correct? They both are. It depends on the observer.

So in one sense a length is a unit of time and in another sense it is a unit of distance. It depends on how you are using it.

In horseracing, Beyer divides the distance of the race by the length of a horse to determine the value of a length.

But also, the time it takes to run one length is relevant.

Your explanation is a simple understanding of relativity.

[whodoyoulike]

I don't know if the following is relative to the conversation, but these thoughts occurred to me when I read this thread...

According to Einstein, Time is relative. It depends on the observer. Who is correct? They both are. It depends on the observer. ...

So in one sense a length is a unit of time and in another sense it is a unit of distance. It depends on how you are using it.

In horseracing, Beyer divides the distance of the race by the length of a horse to determine the value of a length.

But also, the time it takes to run one length is relevant.

This makes a lot of sense and it should prevent an argument.

[steveb]

Prove how it is done; I am very open-minded.

If you want to say that a length is distance converted into time that make sense, but to say a length is time doesn’t make sense.

However before you go off on the deep end, can we agree that time is “the indefinite continued progress of existence “ and a length in horseracing is not indefinite, but you might know something that I am not aware of and I would appreciate your explanation.

forget the lessons in whatever, you are too clever by half.
the simple fact of the matter; is that the margins ARE DERIVED FROM TIME.
is it wrong....yes of course it is, but that does not change the fact that that is how they do it.

for somebody as clever as you are undoubtedly then it's hard to understand why you would even bother working in lengths when they are an inaccurate measure derived from time.

and i am not interested in proving ANYTHING to you.
i know how it's done as does any amount of other people on here.

mr goren when he states margin is a function of time IS correct.

[Cratos]

forget the lessons in whatever, you are too clever by half.
the simple fact of the matter; is that the margins ARE DERIVED FROM TIME.
is it wrong....yes of course it is, but that does not change the fact that that is how they do it.

for somebody as clever as you are undoubtedly then it's hard to understand why you would even bother working in lengths when they are an inaccurate measure derived from time.

and i am not interested in proving ANYTHING to you.
i know how it's done as does any amount of other people on here.

mr goren when he states margin is a function of time IS correct.
First of all he didn't say that a length was a "function" of time; he said a "unit" of time. Additionally, since the advent of Trakus my model, Merlin only use time metrics except when they are not possible to be gotten.

Hang in there, your point is taken without malice.

[cj]

There are differences between the way the charts you see in Equibase and those for Trakus are calculated. For Equibase, when the leader reaches a point of call, the chart caller estimates how far behind the other horses are in lengths. That is a measure of distance, albeit it an inaccurate one. At the finish, the times of each horse are recorded as they hit the wire by the photo finish company. The difference in times is converted to lengths, but it is a time measurement.

With Trakus, all points of call are taken like the finish is at Equibase. If horse A reaches the 1/4 mile in 22 seconds, and horse be reaches it in 23, that one second is converted to a distance, but is in fact a measurement of time.

[steveb]

First of all he didn't say that a length was a "function" of time; he said a "unit" of time. Additionally, since the advent of Trakus my model, Merlin only use time metrics except when they are not possible to be gotten.

Hang in there, your point is taken without malice.


don't get mired in complexities.
regardless of exactly what he said, the meaning of what he said is very very easy for the dumbest of minds to understand.

[Cratos]

There are differences between the way the charts you see in Equibase and those for Trakus are calculated. For Equibase, when the leader reaches a point of call, the chart caller estimates how far behind the other horses are in lengths. That is a measure of distance, albeit it an inaccurate one. At the finish, the times of each horse are recorded as they hit the wire by the photo finish company. The difference in times is converted to lengths, but it is a time measurement.

With Trakus, all points of call are taken like the finish is at Equibase. If horse A reaches the 1/4 mile in 22 seconds, and horse be reaches it in 23, that one second is converted to a distance, but is in fact a measurement of time.
I agree and this discussion is really about a difference in technology which will probably change again in the future.

[highnote]

Your explanation is a simple understanding of relativity.


Tell it to Einstein. That's the example he used. :D

But seriously, you are right. It is an overly simple explanation.

[Tom]

So at non-Trakus tracks, the distance of a length is whatever the chart caller thinks it is. I've seen races at major league tracks where the caller says the leader is out by 3, then out by 5, then out by 7 and the margin never changes.

[Valuist]

Why does Trakus use the winner as a benchmark for a race? Numbers are given plus or minus in feet relative to the winner. But unless you know what kind of trip the winner had, what good is it?

Wouldn't it make more sense to use the distance of the race (adding in runup) as the benchmark? A mile race on the grass with a 120 foot runup would start with a benchmark of 5400 feet, and then we can see the true ground loss for each horse.

[Tom]

That info is probably out there - just not for us to use.

[cj]

Why does Trakus use the winner as a benchmark for a race? Numbers are given plus or minus in feet relative to the winner. But unless you know what kind of trip the winner had, what good is it?

Wouldn't it make more sense to use the distance of the race (adding in runup) as the benchmark? A mile race on the grass with a 120 foot runup would start with a benchmark of 5400 feet, and then we can see the true ground loss for each horse.

The info is there except for the run up distance. However, Trakus overestimates the actual ground traveled by about seven feet per two furlongs. So for a mile race, the benchmark is 5280 feet + 28 for the adjustment. I'm not going to get into specifics on this again since I have in the past, but it is reliable.

The run up is another story. Trakus only uses the data from the finish and works backwards the official distance of the race. Anything before that is run up and excluded from the public. There is no ground loss during that segment at least.

[Valuist]

The info is there except for the run up distance. However, Trakus overestimates the actual ground traveled by about seven feet per two furlongs. So for a mile race, the benchmark is 5280 feet + 28 for the adjustment. I'm not going to get into specifics on this again since I have in the past, but it is reliable.

The run up is another story. Trakus only uses the data from the finish and works backwards the official distance of the race. Anything before that is run up and excluded from the public. There is no ground loss during that segment at least.

OK, thanks. I wonder why they don't use this (actual feet plus the 7 feet per 2 furlong adjustment) as the benchmark? Nothing more frustrating than seeing a grass race where the winner had a wide trip; and everybody else has negative numbers.

I'm going to try using this as a benchmark. When we handicap, so much easier to see horse A had 25 feet of ground loss in its last, and 38 in its race before that, as opposed to a number relative to the winner.

[VigorsTheGrey]

Just a little side question for our physics team here.

It is known that a horse travels for a segment of its running motion with all 4 hooves off the ground. How far does a horse fly from the time the last hoof leaves to when the first hoof touches the ground?

[classhandicapper]

OK, thanks. I wonder why they don't use this (actual feet plus the 7 feet per 2 furlong adjustment) as the benchmark?

There were some discussions about this at the time I found the issue, but I never heard a satisfactory explanation for why this occurs to begin with and never bothered to research it. I just adjusted my spreadsheets. Perhaps CJ and/or Cratos have inquired.

[classhandicapper]

Just a little side question for our physics team here.

It is known that a horse travels for a segment of its running motion with all 4 hooves off the ground. How far does a horse fly from the time the last hoof leaves to when the first hoof touches the ground?

I think Cratos may have hinted at something along those lines, but I can't recall exactly what he said other than defending the technology strongly.

[garyscpa]

Just a little side question for our physics team here.

It is known that a horse travels for a segment of its running motion with all 4 hooves off the ground. How far does a horse fly from the time the last hoof leaves to when the first hoof touches the ground?

As Cratos himself will tell you, this isn't actually flying. Due to the laws of physics, gravity just hadn't caught up with the horse yet.

[cj]

There were some discussions about this at the time I found the issue, but I never heard a satisfactory explanation for why this occurs to begin with and never bothered to research it. I just adjusted my spreadsheets. Perhaps CJ and/or Cratos have inquired.

You found it? Really? Maybe my memory is slipping.

That said, I think Valuist is asking why not just tell us how much "extra" ground each horse covered than the prescribed distance instead of relating it to the winner and having +s and -s. For example, why not for a mile race just have all the horses listed as + and whatever the distance greater than a mile traveled is.

It could very well be that the extra seven feet per two furlongs would stick out and they would rather avoid that, but that is only speculation.

[cj]

I think Cratos may have hinted at something along those lines, but I can't recall exactly what he said other than defending the technology strongly.

Pretty sure he was asking a totally different question which has nothing to do with what was being discussed...not that there is anything wrong with that.

[woodbinepmi]

Was just on Trakus' website and looked at the charts for the first last week at Happy Valley, a 1000 metre race. They have the winner running 952 metres and the second place finisher at 949 metres. How can you take them serious when they are 160 feet WRONG in the distance of a race? Tried to use them to make figures at Woodbine (hence my handle name), but gave up due to them missing way too many races on a card, sometimes up to 3 a day. Don't get me wrong, I love the idea, but we are gambling here with real money and it's hard to justify using a product that has a track record of constant mistakes.

[cj]

Was just on Trakus' website and looked at the charts for the first last week at Happy Valley, a 1000 metre race. They have the winner running 952 metres and the second place finisher at 949 metres. How can you take them serious when they are 160 feet WRONG in the distance of a race? Tried to use them to make figures at Woodbine (hence my handle name), but gave up due to them missing way too many races on a card, sometimes up to 3 a day. Don't get me wrong, I love the idea, but we are gambling here with real money and it's hard to justify using a product that has a track record of constant mistakes.

I just looked at the current thoroughbred meet and through four days there are two races with no data. One was the first race of the meet, one the 9th race on Sunday.

I have seen cases where data will show up a few days later for missing races, but how reliable it is I don't know. I'm skeptical of it for sure.

[cj]

Interesting note on the Hong Kong racing website:

Due to the frequent usage of mobile phones at the racecourses, the accuracy of signals receiving by Trakus may be affected. Therefore, HKJC’s Race Replay Video is provided in parallel with Aerial Virtual Replay for customers’ reference. Every effort is made to ensure the information is as accurate as possible, but the Club assumes no responsibility for it.

[cj]

I can't even find Trakus on the Hong Kong site any longer. The link from Trakus just takes you to the home page and I've looked everywhere. Anybody have a link?

[Valuist]

You found it? Really? Maybe my memory is slipping.

That said, I think Valuist is asking why not just tell us how much "extra" ground each horse covered than the prescribed distance instead of relating it to the winner and having +s and -s. For example, why not for a mile race just have all the horses listed as + and whatever the distance greater than a mile traveled is.

It could very well be that the extra seven feet per two furlongs would stick out and they would rather avoid that, but that is only speculation.

That's exactly right. Just imagine if your Timeform US figs were based strictly on each horse relative to the race winner (not necessary leader)?

[woodbinepmi]

I can't even find Trakus on the Hong Kong site any longer. The link from Trakus just takes you to the home page and I've looked everywhere. Anybody have a link?
I don't think they have the Trakus info on their website, I go directly to Trakus.com (have an account with them). I have never understood why you have to have an account with them to get information off their website. It takes sometimes up to a week for the charts from Hong Kong to show up, have phoned them about it and their response is that the stewards hold onto it before they turn it over to Trakus.

[classhandicapper]

You found it? Really? Maybe my memory is slipping.

That said, I think Valuist is asking why not just tell us how much "extra" ground each horse covered than the prescribed distance instead of relating it to the winner and having +s and -s. For example, why not for a mile race just have all the horses listed as + and whatever the distance greater than a mile traveled is.

It could very well be that the extra seven feet per two furlongs would stick out and they would rather avoid that, but that is only speculation.


Yes. I did. I contacted you privately and after you checked it out you gave me credit for the discovery in a post somewhere on the forum. After that it turned into a debate over why and you started researching it further.

[cj]

Yes. I did. I contacted you privately and after you checked it out you gave me credit for the discovery in a post somewhere on the forum. After that it turned into a debate over why and you started researching it further.

Told you I'm slipping :)

[classhandicapper]

Told you I'm slipping :)

No biggie. I went to you to verify it and help me figure it out. :lol:

The whole thing blows my mind. I just add an adjustment into the spreadsheet I sometimes use for biases.

[Si2see]

Quote:
Interesting note on the Hong Kong racing website:

Cj I am with you here. If the Hong Kong statement is accurate to what is happening, then cell phone usage could cause problems with trakus at any and all race tracks ?

Jason

[Cratos]

Just a little side question for our physics team here.

It is known that a horse travels for a segment of its running motion with all 4 hooves off the ground. How far does a horse fly from the time the last hoof leaves to when the first hoof touches the ground?
It is not when the last hoof leaves the ground because in 1872 photographer Eadweard Muybridge photographed a horse in motion for Leland Sanford the governor of California at the time showing that when a horse goes “airborne” all 4 hooves are off the ground and bended inward; this is where the term “horses can fly” came from.

To answer your question of “far does a horse fly” is more complicated because it depends on the environmental conditions, the horse’s mass and its load; and the exerted force applied by the horse.

[Cratos]

The info is there except for the run up distance. However, Trakus overestimates the actual ground traveled by about seven feet per two furlongs. So for a mile race, the benchmark is 5280 feet + 28 for the adjustment. I'm not going to get into specifics on this again since I have in the past, but it is reliable.

The run up is another story. Trakus only uses the data from the finish and works backwards the official distance of the race. Anything before that is run up and excluded from the public. There is no ground loss during that segment at least.
Trakus does not “overestimates the actual ground traveled by about seven feet per two furlongs”. This is an incorrect assertion with no factual basis.

Trakus uses a race chip timing system which use RFID transponder chips to precisely measure the horses’ times. Horses wear the RFID tags in their saddle cloths during the race.

When compared to the traditional Teletimer timing Trakus is recording both the time and the distance traveled of the horse during the race.

This is done with point-to-point measurement. That is, Trakus establishes X-Y coordinates along the race route and measure at point of intersection.

This becomes a series of tangent vectors that refers to "the rate at which the horse changes its position during the race."

The reason that Trakus doesn’t measure the “run-up” is because it is measuring displacement not distance; run-up is not part of the race specified distance.

[Cratos]

Cj I am with you here. If the Hong Kong statement is accurate to what is happening, then cell phone usage could cause problems with trakus at any and all race tracks ?

Jason
In an electronic signal environment the operative word is “susceptibility” because of conducted and radiated EMI; the fix to this problem is shielding which can be very expensive; especially when a company like Trakus is implementing its technology into a legacy environment that is an outdoor open air environment.

[Cratos]

Was just on Trakus' website and looked at the charts for the first last week at Happy Valley, a 1000 metre race. They have the winner running 952 metres and the second place finisher at 949 metres. How can you take them serious when they are 160 feet WRONG in the distance of a race? Tried to use them to make figures at Woodbine (hence my handle name), but gave up due to them missing way too many races on a card, sometimes up to 3 a day. Don't get me wrong, I love the idea, but we are gambling here with real money and it's hard to justify using a product that has a track record of constant mistakes.
I am curious, how do you know that the course measurement is measured accurately to 1000 meters or how do you know that the race was started correctly at 1000 meters from the finish line?

I am not suggesting that Trakus measurements could not have been in error, but what I am suggesting is that you have not given sufficient info for creditable substantiation to what is right or wrong.

[Cratos]

Why does Trakus use the winner as a benchmark for a race? Numbers are given plus or minus in feet relative to the winner. But unless you know what kind of trip the winner had, what good is it?

Wouldn't it make more sense to use the distance of the race (adding in runup) as the benchmark? A mile race on the grass with a 120 foot runup would start with a benchmark of 5400 feet, and then we can see the true ground loss for each horse.
I would advise you to revisit the Trakus charts and look at the “Distance Run” metric and the “Off the Rail” metric. The intersection of those two metrics independently establishes the horses’ displacement position from start of race; it is not race leader dependent.

[VigorsTheGrey]

It is not when the last hoof leaves the ground because in 1872 photographer Eadweard Muybridge photographed a horse in motion for Leland Sanford the governor of California at the time showing that when a horse goes “airborne” all 4 hooves are off the ground and bended inward; this is where the term “horses can fly” came from.

To answer your question of “far does a horse fly” is more complicated because it depends on the environmental conditions, the horse’s mass and its load; and the exerted force applied by the horse.

5280\8=660
660\12 secs=55fps
Or 5.5 feet per 1\10th of a second.

Does a horse stay airborne for 1\10th of a second?

My guess is for 1\20th of a second....so about 2.75 ft each stride. Am I way off?

How many strides are there in 1 furlong? 660\25 ft?

I also wonder how jockeys time their charges in route? I bet they count the pulses of each completed full stride?

[woodbinepmi]

I am curious, how do you know that the course measurement is measured accurately to 1000 meters or how do you know that the race was started correctly at 1000 meters from the finish line?

I am not suggesting that Trakus measurements could not have been in error, but what I am suggesting is that you have not given sufficient info for creditable substantiation to what is right or wrong.

Because there is a huge sign that says '1000 metres' next to the starting gate. Hong Kong unlike here doesn't just put the gate where they want to, everything is measured and timed from the gate, there is no run-ups. If you don't believe me, here is the video to the first race that was measured at 952 metres: http://racing.hkjc.com/racing/video/play.asp?type=replay-full&date=20160413&no=01&lang=eng

Here is the video for the 7th race on the same evening that is measured over 1000 metres: http://racing.hkjc.com/racing/video/play.asp?type=replay-full&date=20160413&no=07&lang=eng

Is that enough evidence or do you need more?

[Cratos]

Because there is a huge sign that says '1000 metres' next to the starting gate. Hong Kong unlike here doesn't just put the gate where they want to, everything is measured and timed from the gate, there is no run-ups. If you don't believe me, here is the video to the first race that was measured at 952 metres: http://racing.hkjc.com/racing/video/play.asp?type=replay-full&date=20160413&no=01&lang=eng

Here is the video for the 7th race on the same evening that is measured over 1000 metres: http://racing.hkjc.com/racing/video/play.asp?type=replay-full&date=20160413&no=07&lang=eng

Is that enough evidence or do you need more?
First of all your sarcasm buys you nothing with me. I questioned the validity of your earlier post statement, but I never said you were wrong.

I asked for validation you came back with a "fixed sign" contrasted against a Trakus video.

Again you might be right and if you feel that strong about it, contact Trakus directly and get their answer.

[woodbinepmi]

I am not contacting Trakus because I could care less, have found too inconsistencies in the past to care about.
As you can see by the videos, the gate are positioned at the same location, for them to be off by 61 metres in their charts shows that there are serious issues with their product.
Also, I am not selling you anything, I wasn't being sarcastic, it was a question.

[Cratos]

5280\8=660
660\12 secs=55fps
Or 5.5 feet per 1\10th of a second.

Does a horse stay airborne for 1\10th of a second?

My guess is for 1\20th of a second....so about 2.75 ft each stride. Am I way off?

How many strides are there in 1 furlong? 660\25 ft?

I also wonder how jockeys time their charges in route? I bet they count the pulses of each completed full stride?
You asked some good question.

How long a horse stays airborne will take some calculations because you have a lot going on to include the horse's applied force against the g-force and the trajectory of its air travel is an arc.

I will think about it or maybe someone else have the answer.

Strides per furlong would estimate to be about an average of 30 given that a horse's stride is 21-23 feet in length.

A g .= 22, 660/22 = 30

[Dave Schwartz]

Wasn't there a video somewhere that claimed Secretariat had a 28' stride?

[Si2see]

In an electronic signal environment the operative word is “susceptibility” because of conducted and radiated EMI; the fix to this problem is shielding which can be very expensive; especially when a company like Trakus is implementing its technology into a legacy environment that is an outdoor open air environment.

Thanks for the explanation ( though it's way over my head :lol: )

What you are saying still makes me believe that if the Hong Kong statement is accurate, then the same occurrences could be happening with any and all tracks using Trakus, therefore Trakus could very well be extremely inaccurate or inconsistent like other posters are stating?

Jason

[Cratos]

Wasn't there a video somewhere that claimed Secretariat had a 28' stride?
I agree there was published info that Big Red had the stride you suggested; however I can't recall a confirmation, but when it come to Secretariat's qualities I am a believer.

[Valuist]

I would advise you to revisit the Trakus charts and look at the “Distance Run” metric and the “Off the Rail” metric. The intersection of those two metrics independently establishes the horses’ displacement position from start of race; it is not race leader dependent.

Then why have the plus/minus numbers on the far right relative to the winner?

[Cratos]

Thanks for the explanation ( though it's way over my head :lol: )

What you are saying still makes me believe that if the Hong Kong statement is accurate, then the same occurrences could be happening with any and all tracks using Trakus, therefore Trakus could very well be extremely inaccurate or inconsistent like other posters are stating?

Jason
No, I don't agree with your suggestions because I have gone through several new product developments and new product introductions; errors and mistakes were made that were not found in the lab or field testing.

I say this because many products (software and hardware) which we all use today went through many revisions before they became fully customer acceptable.

Therefore Trakus will have a tenure in the marketplace to eliminate it's missteps and grow or new competition will come with a better and more acceptable product; and Trakus might be replaced.

[cj]

Trakus does not “overestimates the actual ground traveled by about seven feet per two furlongs”. This is an incorrect assertion with no factual basis.

Trakus uses a race chip timing system which use RFID transponder chips to precisely measure the horses’ times. Horses wear the RFID tags in their saddle cloths during the race.

When compared to the traditional Teletimer timing Trakus is recording both the time and the distance traveled of the horse during the race.

This is done with point-to-point measurement. That is, Trakus establishes X-Y coordinates along the race route and measure at point of intersection.

This becomes a series of tangent vectors that refers to "the rate at which the horse changes its position during the race."

The reason that Trakus doesn’t measure the “run-up” is because it is measuring displacement not distance; run-up is not part of the race specified distance.

If a horse runs two furlongs on the rail around a turn or on the straightaway, it should say the horse ran 1320 feet. That is what horseplayers want to know. They don't want extra distance built in.

You are dead wrong about run up, sorry. Trakus does in fact measure run up. It is measured and it is timed. It is not, however, reported in the T-Charts. I've seen it with my own two eyes.

[cj]

I am curious, how do you know that the course measurement is measured accurately to 1000 meters or how do you know that the race was started correctly at 1000 meters from the finish line?

I am not suggesting that Trakus measurements could not have been in error, but what I am suggesting is that you have not given sufficient info for creditable substantiation to what is right or wrong.

LOL.

[cj]

First of all your sarcasm buys you nothing with me. I questioned the validity of your earlier post statement, but I never said you were wrong.

I asked for validation you came back with a "fixed sign" contrasted against a Trakus video.

Again you might be right and if you feel that strong about it, contact Trakus directly and get their answer.

He is right, hard as it might be for you to admit.

[cj]

Thanks for the explanation ( though it's way over my head :lol: )

What you are saying still makes me believe that if the Hong Kong statement is accurate, then the same occurrences could be happening with any and all tracks using Trakus, therefore Trakus could very well be extremely inaccurate or inconsistent like other posters are stating?

Jason

I like Trakus a lot, but there are a lot of errors that slip through the cracks. The team they have in place is probably too small, most likely for financial reasons.

[steveb]

I am curious, how do you know that the course measurement is measured accurately to 1000 meters or how do you know that the race was started correctly at 1000 meters from the finish line?

I am not suggesting that Trakus measurements could not have been in error, but what I am suggesting is that you have not given sufficient info for creditable substantiation to what is right or wrong.

there is NOT a shadow of doubt that the error is trakus, because the speed of that race was within one length of expectations.
if it was far too short, there is no way it would not have been very very fast speed, and my computers would have been in meltdown because something was miles outside the expected parameters, and they weren't.

not to mention hong kong racing is the best on the planet as far as accuracy goes.
and maybe that is why there is no link to trakus on their site!!!
and i don't think cj is right where he says it has now gone.....was it ever there in the first place?
i frequent that site and can never recall hkjc linking to trakus.

[cj]

there is NOT a shadow of doubt that the error is trakus, because the speed of that race was within one length of expectations.
if it was far too short, there is no way it would not have been very very fast speed, and my computers would have been in meltdown because something was miles outside the expected parameters, and they weren't.

not to mention hong kong racing is the best on the planet as far as accuracy goes.
and maybe that is why there is no link to trakus on their site!!!
and i don't think cj is right where he says it has now gone.....was it ever there in the first place?
i frequent that site and can never recall hkjc linking to trakus.

I don't know if it was ever there. I just know the Trakus site lists them as a partner track but doesn't link to the T-Charts like the others, it just goes to the home page.

[Cratos]

Then why have the plus/minus numbers on the far right relative to the winner?
I am assuming you are speaking of the "Delta Column" which gives the difference in feet travelled (+/-).

It is a convienent column, but it changes nothing about the displacement.

You could do the same yourself by adding the segment displacements of each horse and subtract the difference at the end.

[Cratos]

there is NOT a shadow of doubt that the error is trakus, because the speed of that race was within one length of expectations.
if it was far too short, there is no way it would not have been very very fast speed, and my computers would have been in meltdown because something was miles outside the expected parameters, and they weren't.

not to mention hong kong racing is the best on the planet as far as accuracy goes.
and maybe that is why there is no link to trakus on their site!!!
and i don't think cj is right where he says it has now gone.....was it ever there in the first place?
i frequent that site and can never recall hkjc linking to trakus.
Steve,
I am going to admit to something Malcolm Forbes once said and it was:"I make no apologies for being a capitalist."

Having said that, I believe in "price discrimination" and "product discrimination."

Therefore if Trakus is not your product; product discriminate and don't use it.

[cj]

Steve,
I am going to admit to something Malcolm Forbes once said and it was:"I make no apologies for being a capitalist."

Having said that, I believe in "price discrimination" and "product discrimination."

Therefore if Trakus is not your product; product discriminate and don't use it.

You have no idea if he uses Trakus or not, so your comment has no place here. Whether he does or not he is free to relay knowledge he as about it, especially real life knowledge and not wishful thinking.

[steveb]

then if you were to compare the sections from hkjc to trakus for any meeting/race you care to, they will always have variations, EXCEPT, the one constant is the final time!!
i have no doubt trakus times are 'fitted', and i have done much research on them for various jurisdictions(not usa)

it's a great idea, but just like every other method i have tested...stridemaster, equitime and whatever, they are full of errors.


now i have also decided to put this here.....
the numbers are all taken from trakus.
moderator can remove it, if need be.
why are the numbers taken a week after the meeting, different to numbers taken at a later date?
it is a question, i would love answered, although i don't really care these days.......

[Cratos]

He is right, hard as it might be for you to admit.
It is not what you or me believe; it is what is right and he hasn't proven anything right, but it doesn't mean he is wrong; it mans it is Just conjecture.

[davew]

I have not looked at them often, but see they had a different winner in the 4th at AQU today. The shorter horse got beat by a nose, but I guess the saddlecloth got to the wire first.

[cj]

then if you were to compare the sections from hkjc to trakus for any meeting/race you care to, they will always have variations, EXCEPT, the one constant is the final time!!
i have no doubt trakus times are 'fitted', and i have done much research on them for various jurisdictions(not usa)

it's a great idea, but just like every other method i have tested...stridemaster, equitime and whatever, they are full of errors.


now i have also decided to put this here.....
the numbers are all taken from trakus.
moderator can remove it, if need be.
why are the numbers taken a week after the meeting, different to numbers taken at a later date?
it is a question, i would love answered, although i don't really care these days.......

This is fine, no need to remove anything.

[cj]

It is not what you or me believe; it is what is right and he hasn't proven anything right, but it doesn't mean he is wrong; it mans it is Just conjecture.

It isn't conjecture. It was an obvious error and anybody that halfway understands racing would know it right away.

[Cratos]

You have no idea if he uses Trakus or not, so your comment has no place here. Whether he does or not he is free to relay knowledge he as about it, especially real life knowledge and not wishful thinking.
You are correct, but I never said that he use Trakus; what I suggested was to product discriminate.

What that could mean is stop using it or don't start using it.

[steveb]

Steve,
I am going to admit to something Malcolm Forbes once said and it was:"I make no apologies for being a capitalist."

Having said that, I believe in "price discrimination" and "product discrimination."

Therefore if Trakus is not your product; product discriminate and don't use it.

that is a nonsense reply cratos.
the idea is great, the reality is something not so great.
i do note that when offered the chance to see those errors for yourself, you chose not to.
i guess you did not want to fog up your beautiful rose coloured glasses!

i got where i am today, as far as my racing life is concerned, by not taking anything at face value.
trakus simply failed those tests.

[Cratos]

that is a nonsense reply cratos.
the idea is great, the reality is something not so great.
i do note that when offered the chance to see those errors for yourself, you chose not to.
i guess you did not want to fog up your beautiful rose coloured glasses!

i got where i am today, as far as my racing life is concerned, by not taking anything at face value.
trakus simply failed those tests.
First let me say this to you.

Although we disagree,I admire you for the success you have had in wagering on horseracing.

Secondly, I hate losing with a passion, but I do lose and I pick myself up and keep going.

Lastly, I see nothing through "rose-colored glasses"; i am tough-minded and a die-heart New Yorker; that might not sit well with some, but I don't give damn.

[steveb]

First let me say this to you.

Although we disagree,I admire you for the success you have had in wagering on horseracing.

Secondly, I hate losing with a passion, but I do lose and I pick myself up and keep going.

Lastly, I see nothing through "rose-colored glasses"; i am tough-minded and a die-heart New Yorker; that might not sit well with some, but I don't give damn.


and you have chosen to say sweet nothing about trakus?

as for not sitting well with some; i don't actually give a toss if you are tough minded, and a die-heart new yorker, because it is meaningless twaddle, in the context of trakus.
everybody is from some place, or has a place that is dear to them, but it has sweet sweet nothing to do with trakus, or horse racing.

why don't you do something useful and tell me why trakus changes?
and what do you do in such instances??

[Dark Target]

They certainly seem fitted. Out of curiosity, what does your computer say about the time for race 4 on 29th Jan @ Singapore?

[steveb]

They certainly seem fitted. Out of curiosity, what does your computer say about the time for race 4 on 29th Jan @ Singapore?

dt, i stopped keeping data for singapore last xmas, so i don't know.
considering i spend most of my time doing other things these days, i could not justify the time, so stopped.

EDIT: but being the inquisitive person i am, i just checked. :confused:
and it is too slow.
if race 1 is correct at 73.07(i get 72.7), then race 4 should be circa 65.03(i get 65.4), not 66.08

[Cratos]

and you have chosen to say sweet nothing about trakus?

as for not sitting well with some; i don't actually give a toss if you are tough minded, and a die-heart new yorker, because it is meaningless twaddle, in the context of trakus.
everybody is from some place, or has a place that is dear to them, but it has sweet sweet nothing to do with trakus, or horse racing.

why don't you do something useful and tell me why trakus changes?
and what do you do in such instances??

I play NYRA (except Aqueduct) and there is nothing to tell; me my associates have enjoyed Trakus at NYRA and will continue with them because their data fits our model.

Your complaints should be addressed to Trakus directly or to the management of the Trakus racetrack in question; if not you are just blowing hot air (if that can be done) on an Internet forum.

[Dark Target]

Thanks mate.

[steveb]

I play NYRA (except Aqueduct) and there is nothing to tell; me my associates have enjoyed Trakus at NYRA and will continue with them because their data fits our model.

Your complaints should be addressed to Trakus directly or to the management of the Trakus racetrack in question; if not you are just blowing hot air (if that can be done) on an Internet forum.

the hot air consists of a file with undoubted differences unless you believe,... i made it up.
and if you now looked at trakus for that meeting you would see it is again different!
and i don't actually have any complaints, i long ago decided trakus is not fit for purpose.

it must be wonderful having those associates, and decimating the nyra totes(except aquaduct!:lol:), with those 200k bets i read about.
i was just looking at a file i have from woods and associates which would probably be close to biggest race investors on the planet.
the biggest outlay over a 6 month period was 1.67 million for 87 races at flemington victoria on win pool, which is a meagre 19k odd for each race.
and that was when the pools were bigger than they now are.
for all pools it is about 38k, so you are in rarefied air.
how do you stop your own bets destroying the value?

i am also of the opinion that you are not telling all.
you spend so much time defending trakus that one could be forgiven for thinking you have an ulterior motive.

[steveb]

Thanks mate.

dt, my first impression would be that they were hand timed or something, because when you time races yourself off video, then invariably official time is SLOWER than the time you come up with.
if official time is faster than yours, then it will usually be because of 'reaction time',....that is the gates spring open and the guy presses his stop watch or whatever, but the reaction takes time, and so they(times) are too quick.

i timed r1 and r4 from video
race 1 i got 72.7 but official was slower at 73.07
r4 i got 66.08 and official was FASTER at 65.4.

i think i got it arse up before.
if 73.07 is right for r1, then race 4 should be about 66.08 + .37 is 66.45

but as i said, that is dependent on r1 being correct and all others bar race 4 too, but i did not check those others.

[sammy the sage]

well Trackus did or maybe still (corrected now)?does have/had one THIRD of the field in the Fla. Derby doing 49mph at one point...which we ALL know is not possible... :lol:

[Dark Target]

dt, my first impression would be that they were hand timed or something, because when you time races yourself off video, then invariably official time is SLOWER than the time you come up with.
if official time is faster than yours, then it will usually be because of 'reaction time',....that is the gates spring open and the guy presses his stop watch or whatever, but the reaction takes time, and so they(times) are too quick.

i timed r1 and r4 from video
race 1 i got 72.7 but official was slower at 73.07
r4 i got 66.08 and official was FASTER at 65.4.

i think i got it arse up before.
if 73.07 is right for r1, then race 4 should be about 66.08 + .37 is 66.45

but as i said, that is dependent on r1 being correct and all others bar race 4 too, but i did not check those others.

Yes i thought it would be strange that R4 was slow, as it seemed a fair bit faster than i expected, so i checked trakus, and sure enough the time was exactly the same...

[steveb]

Yes i thought it would be strange that R4 was slow, as it seemed a fair bit faster than i expected, so i checked trakus, and sure enough the time was exactly the same...

thanks for the chuckle dt.
i just went to trakus and had a look at the winners sections
the race time was 65.4(and wrong as in too fast) but if you add all the section together it comes to an even quicker 65.24!
its cumulative time to last 100 was 58.98, so if you take that away from the winning time (65.4) then you get 6.42 seconds, but trakus says it went the last 100 in 6.31 seconds a massive .11 diff.

of course cratos and his associates have it all figured, and are killing them in new york with their carefully selected humungous wagers. :lol:
that's what happens when you are a whizz at physics!

[Dark Target]

Yes they seem to be pure fantasy.

[castaway01]

thanks for the chuckle dt.
i just went to trakus and had a look at the winners sections
the race time was 65.4(and wrong as in too fast) but if you add all the section together it comes to an even quicker 65.24!
its cumulative time to last 100 was 58.98, so if you take that away from the winning time (65.4) then you get 6.42 seconds, but trakus says it went the last 100 in 6.31 seconds a massive .11 diff.

of course cratos and his associates have it all figured, and are killing them in new york with their carefully selected humungous wagers. :lol:
that's what happens when you are a whizz at physics!

First of all, very funny.

Second, we unfortunately end up in this place in every discussion of Trakus. It's a good concept, but there are obviously flaws and mistakes made with their numbers, many MANY of which have been pointed out on this site. The difference is that some of us look at Trakus as a program that has flaws, and others (well, one person) treat any complaints about it like you insulted their child. Ulterior motives indeed.

[Kash$]

https://youtu.be/m2PQhXDBfTA

2:28 mark

[cj]

https://youtu.be/m2PQhXDBfTA

2:28 mark

This stuff happens ALL THE TIME. (Sorry Cratos)

[Cratos]

https://youtu.be/m2PQhXDBfTA

2:28 mark

Is this on the published Trakus chart? Also please show where either Equibase or DRF ever gave speed data? Yes, historically both have given "time data," but never speed data.

Therefore anyone who is deriving "speed" from Equibase/DRF data is incorrectly doing so. The best that can be done is an estimated speed for the specific race distance and even then that would be incorrect.

Like Trakus or dislike Trakus data; it is the best there currently is for deriving speed and vector analysis for thoroughbred handicapping.

Can it be improved? Yes, but I will take its data today over Equibase.

[cj]

Is this on the published Trakus chart?

Probably not. I'm sure the guy doctored it himself just to denigrate Trakus while promoting them at the same time. :rolleyes:

[woodbinepmi]

Probably not. I'm sure the guy doctored it himself just to denigrate Trakus while promoting them at the same time. :rolleyes:
Just checked the chart on the Florida Derby from the Trakus' website:

At 3/16 (1000 ft) from the gate:
Nyquist 39.0 Peak (mph)
Mohaymen 39.8 Peak (mph)

At 1/4 (350 ft later)
Nyquist 49.7 Peak (mph)
Mohaymen 49.9 Peak (mph)

That's one hell of a speed increase by a pack of horses.

[Tom]

That will be compensated for by an estimation of the weights of the horses coupled with wind speed data from airports miles away. It all works out.

[whodoyoulike]

I found this interesting info regarding Trakus. Hope it helps.

GFub7d2ACyQ

[Cratos]

The most detrimental environmental force that impedes a racehorse's rate of motion (pace) is aerodynamic drag (air resistance).

Wind forces are taken locally to the racetrack and are measured at the surface level for resistivity.

The least and most popular is surface resistance which is measured by the coefficient of kinetic friction between the horse's hooves and the racetrack's surface through GRF.

[fiznow]

I really like the idea of Trakus, it can be a very useful tool. I'll give them some time to fix bugs and hope it will be available for more tracks soon.