How do you use Trakus?

[Tom]

The big surprise to me has been the overhead vie - really nifty to see where the horses were during the race. I like that the even show the main turf course on the other side of the rail! Detail.

As far as the ground covered and times.....any suggestions?

I'm downloading the summary of the race time for now and saving them, while I play around with a few ideas.

Biggest thing I have using it for at other tracks is turf races....basing the late pace rating on actual ground covered.

[Robert Fischer]

The big surprise to me has been the overhead vie - really nifty to see where the horses were during the race. I like that the even show the main turf course on the other side of the rail! Detail.

As far as the ground covered and times.....any suggestions?

I'm downloading the summary of the race time for now and saving them, while I play around with a few ideas.

Biggest thing I have using it for at other tracks is turf races....basing the late pace rating on actual ground covered.

Consider the Pace when looking at ground loss.

Ground loss works best when comparing like running styles.

[Robert Fischer]

EXAMPLE: Kentucky Derby.

A quick look at the "Race Summary" shows that closers dominated the finish positions of the derby
indicating a bias. The very fast for distance "Trakus Times" at the top confirm this.

The trio of closers was then indicated with blue, while the 1 pace horse showing was indicated in red.
http://www.fotoshack.us/fotos/78189TrakusDerby1.jpg






Now we have to go to a Pace Call. Here is the 1/2 Mile.
The early pace group is "red", the trio of closers "blue".
Comparisons within like running styles
= Of the Early group, Vyjack lost a ton of ground, and Oxbow saved a ton of ground.
= Of the Late group, Orb lost a good bit of ground,
while Golden Soul and Revolutionary saved a ton of ground.
http://www.fotoshack.us/fotos/38167TrakusDerby2.jpg






Finally, we examine the "finish positions".
Of note = Orb ran 80! more feet than the meaningful like running style horses (Golden Soul,Revolutionary).

Oxbow was the only Early Pace horse to finish anywhere near the front.
Also of note from the Early Pace Group, Vyjack ran 50-70 more feet.
http://www.fotoshack.us/fotos/75967trakusDerby3.jpg

[Robert Fischer]

I don't know how clear a job I did with that (sometimes pictures help and sometimes they are a headache to read).

Obviously, it's easier to back-fit an example, but you did pretty well if you used Orb, Oxbow, Palace Malice, Verrazano, Vyjack, and Goldencents going forward , and played against Golden Soul and Revolutionary.


If pace is not an issue in a given race, you can concentrate more on AVG(mph). [Then you wouldn't need to separate into like running styles for that race.]


You want to look at the turns as well. In a 2 turn race like the Derby, You can pick something like 4f like the example, so that you see both the 1st turn and the pace call.

Even if pace is not a factor you still want to know if the horse got killed on the first turn. It doesn't matter if it's exact as long as it's sometime after the 1st turn and before the 2nd turn(so it's nice to capture both the pace and the turn in one look like the 4f of the Derby).

In the example, Vyjack really had a nightmare 1st turn. Unfortunately he never came back in a good spot, and is off for the year, reportedly for ulcers. He stopped badly (along w/ Goldencents = >9seconds final furlong!), but the fact that he was both against the grain(on the hot pace) and that specifically his 1st turn was very wide (33feet wider than Palace Malice!) makes him a good example of significant 1st turn data. His AVG(mph) will also reflect that. [40.3mph, Vyjack ran the fastest 1/2 in the Derby per Trakus.].

[classhandicapper]

Trakus is a great tool, but I suspect just as many people are going to misuse it (especially initially) as use it well. The math/geometry behind ground loss is certain, but the impacts are not. IMO if you combine uncertain ground loss impacts with uncertain pace impacts in a literal way you are often going to get lead badly astray. I tired that years ago (using manual ground loss notes for NY only) and the results were not pleasant. Granted, I know more now than I did back then, but it's probably a mortal lock that other people are going to take the exact path I did and be forced to reverse course and look at it another way.

[cj]

Classhandicapper asked me a question which led me to do some research and ask a few questions. There is clearly a small issue with the ground covered part of Trakus. It is pretty small, but still matters

Ground loss is being overestimated. As an example, look at the 2nd race at Belmont yesterday, a 7f dirt race where the entire first 2f is run on a straight. Every horse is listed as running between 1327-1329 feet. That is a lot of extra ground for a straight.

I realize that even on a straight, horses don't run straight lines, but it would take a huge amount of drifting or swerving to make a horse run an extra 9 feet. Here is an example using the Pythagorean Theorem:

2f straight line: 1320 feet
Horses drifts: 100 feet
Total distance run: 1323.78

So even with 100 feet of drifting, the horse only runs 4 extra feet, not 7-9.

To actually run 1329 feet as one of the horses is listed to have done, the horse would have had to have drifted and weaved 154.41 feet. This is just one horse. It can happen, but the odds are pretty remote. Trakus has ALL the horses running extra.

This isn't a big knock on Trakus. I love it still. But if it is off by 7 feet every 2f, that can add up over the course of a race. The good news is it appears pretty consistent from horse to horse, so comparisons on ground traveled between horses should still be applicable. It would be a much bigger problem if the ground issue varied by location on the track, or from track to track. That doesn't appear to be the case.

[classhandicapper]

Yea, it's still useful, but I may have to do some adjustments in my spreadsheet once I figure out exactly what's going on. I was doing some research and the numbers didn't make sense.

[bbopjz]

Please someone tell me how to use Trakus in my handicapping?

[devilsbag]

Please someone tell me how to use Trakus in my handicapping?

Install Trakus in your home, setting up points in the bathroom, the kitchen, bedroom, and the location where you do your handicapping. Sew a pouch for the transmitter in your tighty whities (or whatever your preferred undergarment may be).

Then, time yourself on how long it takes for you to get from your handicapping location to these other points in the house. You'll see how much time you're wasting elsewhere. Ultimately, I altered my routine so that all my handicapping now takes place in the bathroom, better known as handicrapping. Less time wasted means more winners.

I sold my Trakus setup on eBay and having been cashing tickets ever since.

[Valuist]

Best use for Trakus? Helps out considerably when watching replays in a big field. I wish the Fair Grounds had it; the split screen makes it very difficult to differentiate horses when they race into the far turn.

[cj]

Please someone tell me how to use Trakus in my handicapping?

Other than for help watching live or replays as Valuist notes, the use is limited until it is provided in a more user friendly format. Trakus would be great for data manipulation, but you can't do it without investing considerable time given the current format.

[GaryG]

When you use it for internal and final fractions you don't have to try and calculate the beaten lengths. I use it mostly for closing fractions in grass races.

[iceknight]

I use Trakus for some amount of trip handicapping. But there are always other issues with "S" type horses so the combination of better trip and a good price still remains hard to find.

[Tom]

I like the overhead replay function, watching with the charts in hand.

[PhantomOnTour]

Install Trakus in your home, setting up points in the bathroom, the kitchen, bedroom, and the location where you do your handicapping. Sew a pouch for the transmitter in your tighty whities (or whatever your preferred undergarment may be).

Then, time yourself on how long it takes for you to get from your handicapping location to these other points in the house. You'll see how much time you're wasting elsewhere. Ultimately, I altered my routine so that all my handicapping now takes place in the bathroom, better known as handicrapping. Less time wasted means more winners.

I sold my Trakus setup on eBay and having been cashing tickets ever since.
You can love Devils Bag or you can hate Devils Bag - call him a troll or not call him a troll...but don't EVER say the man ain't funny as hell.
(and yes - he has gotten me too) :lol:

[DJofSD]

Classhandicapper asked me a question which led me to do some research and ask a few questions. There is clearly a small issue with the ground covered part of Trakus. It is pretty small, but still matters

Ground loss is being overestimated. As an example, look at the 2nd race at Belmont yesterday, a 7f dirt race where the entire first 2f is run on a straight. Every horse is listed as running between 1327-1329 feet. That is a lot of extra ground for a straight.

I realize that even on a straight, horses don't run straight lines, but it would take a huge amount of drifting or swerving to make a horse run an extra 9 feet. Here is an example using the Pythagorean Theorem:

2f straight line: 1320 feet
Horses drifts: 100 feet
Total distance run: 1323.78

So even with 100 feet of drifting, the horse only runs 4 extra feet, not 7-9.

To actually run 1329 feet as one of the horses is listed to have done, the horse would have had to have drifted and weaved 154.41 feet. This is just one horse. It can happen, but the odds are pretty remote. Trakus has ALL the horses running extra.

This isn't a big knock on Trakus. I love it still. But if it is off by 7 feet every 2f, that can add up over the course of a race. The good news is it appears pretty consistent from horse to horse, so comparisons on ground traveled between horses should still be applicable. It would be a much bigger problem if the ground issue varied by location on the track, or from track to track. That doesn't appear to be the case.
cj, I am not a Trakus user -- I have no idea what the data looks like let alone what are the supplied data points versus what is being computed.

With that said, if Trakus is using continuous data, they probably are measure the distance along the path the horse takes, an instantaneous measure, instead of a straight line distance from point A to B. (Think integral calculus.)

[HuggingTheRail]

Perhaps a silly question, but I ask them often...

If Trakus uses a chip in the saddlecloth to track each horse, is there any up and down motion of the saddlecloth that may account for a very small amount of the distance travelled?

[DJofSD]

Perhaps a silly question, but I ask them often...

If Trakus uses a chip in the saddlecloth to track each horse, is there any up and down motion of the saddlecloth that may account for a very small amount of the distance travelled?
Good question.

I think in part it depends on the embedded chips. Whether they're using accelerometers with either 2 or 3 axis and what data is being captured.

For example, I use an app on an iPod Touch called Golf MTRx. It measures the speed and position of your hips relative to (1) the initial position in space as you address the ball, and (2) when the sound of the strike of the golf ball is detected. The app displays a lot of good data for improving your swing but it does so in 2-space and not 3.

[Robert Goren]

Trakus for most of us recreational bettors is very difficult for us to use until they put out DRF or BRIS style PPs. I think they are coming and it is only matter of when.

[Cratos]

cj, I am not a Trakus user -- I have no idea what the data looks like let alone what are the supplied data points versus what is being computed.

With that said, if Trakus is using continuous data, they probably are measure the distance along the path the horse takes, an instantaneous measure, instead of a straight line distance from point A to B. (Think integral calculus.)

I agree with your assumption that Trakus is calculating instantaneous speed and if I remember my calculus correctly, the horse’s speed would be written in the language of calculus with speed as the first derivative of distance with respect to time.

[Cratos]

Trakus for most of us recreational bettors is very difficult for us to use until they put out DRF or BRIS style PPs. I think they are coming and it is only matter of when.

Your are correct Trakus does have a formatting issue

[DJofSD]

Yep, dx/dt.

Trying to get my son to understand what he's learning right now (basic analytic geometry and slopes) is what he'll need to know for doing physics and calculus.

[Cratos]

Good question.

I think in part it depends on the embedded chips. Whether they're using accelerometers with either 2 or 3 axis and what data is being captured.

For example, I use an app on an iPod Touch called Golf MTRx. It measures the speed and position of your hips relative to (1) the initial position in space as you address the ball, and (2) when the sound of the strike of the golf ball is detected. The app displays a lot of good data for improving your swing but it does so in 2-space and not 3.

I am not sure, but I believe Trakus is using RFID tags which are small battery-less chips attached to the horse’s saddle cloth. RFID technology is typically used to track the movement of objects and RFID tags are passive.

[Cratos]

Please someone tell me how to use Trakus in my handicapping?

Please go to the Gulfstream Park website and you will find a good illustrated view of how Trakus is set-up; I will send to you a PM of some very good, but simple uses of Trakus.

[DJofSD]

I am not sure, but I believe Trakus is using RFID tags which are small battery-less chips attached to the horse’s saddle cloth. RFID technology is typically used to track the movement of objects and RFID tags are passive.
If you're correct, that changes a lot of things.

[classhandicapper]

It's a welcome addition in some ways, but I did about a month of research into how I might use it and abandoned the project. I don't believe in "literal" ground loss adjustments to figures for a variety of reasons. So it was only of limited benefit to me personally in that way. I also tried to use it to help with bias determination at tracks I don't follow so I didn't have to watch so many replays, but I ran into issues there also.

At this point I am hoping more people use it to adjust their figures for literal ground loss because I think that will create value for me.

[Cratos]

It's a welcome addition in some ways, but I did about a month of research into how I might use it and abandoned the project. I don't believe in "literal" ground loss adjustments to figures for a variety of reasons. So it was only of limited benefit to me personally in that way. I also tried to use it to help with bias determination at tracks I don't follow so I didn't have to watch so many replays, but I ran into issues there also.

At this point I am hoping more people use it to adjust their figures for literal ground loss because I think that will create value for me.

As I understand your use of “literal ground loss” you are referring to true ground loss. Am I correct?

Ground loss is an assumptive conclusion made by many horseplayers which I find to be false. Horseplayers will look at the ground (distance) travelled by one horse when compared to another and the difference is considered to be “ground loss” which again in my opinion is fallacious.

Trakus help us understand that in a given race what the distance is travelled by each horse and the time it took the horse to travel that distance.

Therefore if you normalized time travelled with respect to distance travelled you will find that the fastest horse doesn’t always win the race because it travelled too much distance in the race and that is not ground loss, but “longer distance” travelled and the aforementioned normalized metric will take you to the “if condition” which says if my horse had run that rate of speed over the same distance travelled as the winner, it would have won, but that is conjecture not fact.

[PaceAdvantage]

You can love Devils Bag or you can hate Devils Bag - call him a troll or not call him a troll...but don't EVER say the man ain't funny as hell.
(and yes - he has gotten me too) :lol:I agree... :lol:

[raybo]

I thought that the main use for Trakus data would be that all horses in a race would have exact times at the calls, which alleviates the inherent errors we make when adjusting the leader's time by "estimated" beaten lengths and estimated distance of a length. At least that is the way I would use the data (I would not use ground loss at all). Is this thinking wrong?

Regardless, until we get the Trakus data in PP data files, it will have no value for me.

[thaskalos]

Beyer had written in his book "Beyer on Speed" -- which was published in 1993 -- that the technology already existed whereby each horse in the race would be timed individually...and we would have the individual fractions of each horse in a race -- instead of basing our pace calculations on the timing of the leaders at every call. Was Trakus the technology that Beyer was talking about in that book?

If so...I am not too impressed by the strides that this technology is making. I somehow thought that this sort of technology would be targeted more towards the bettors than towards the racing industry insiders.

[Cratos]

Beyer had written in his book "Beyer on Speed" -- which was published in 1993 -- that the technology already existed whereby each horse in the race would be timed individually...and we would have the individual fractions of each horse in a race -- instead of basing our pace calculations on the timing of the leaders at every call. Was Trakus the technology that Beyer was talking about in that book?

If so...I am not too impressed by the strides that this technology is making. I somehow thought that this sort of technology would be targeted more towards the bettors than towards the racing industry insiders.

I am not sure what technology Andy Beyer was referring too in his 1993 book “Beyer on Speed” which you referenced in your post, but tracking technology goes back at least 50 years before 1993.

Tracking technology use in horseracing might be very recent, but if I remember correctly the Trakus group originally designed their technology for hockey; not horseracing.

Additionally, it is not the technology that is the problem, but the application of the technology. The people at Equibase/DRF have been collecting and distributing data to horseplayers for many years and it should be them with their expertise to get the tracking application technology correct.

However the best laid plans doesn’t always work because if they did Microsoft who had the cellphone technology before Apple would be dominant in that market today and Xerox would the “king of the hill” in the PC market today.

Also if the racetracks would list the horses’ body weight; and if Trakus would add surface resistance, wind force, and air resistance algorithms to their software, speed handicapping would move into the stratosphere.

In summation, if you are a speed handicapper the Trakus data is the best out there today (although limited in its distribution).

[DJofSD]

Another little tidbit to add to the litany: Kodak had digital image technology for a long time but did nothing with it. Where are they now?

[cj]

Anybody else read about Nakatani ripping his transmitters out? What is the deal with that?

[DJofSD]

Protecting the family jewels?