Chart Callers

[chickenhead]

thought I'd start a new thread, the other one is getting too long.

sent emails to about 15 tracks, will post any responses here.

[chickenhead]

Charts at all recognized North American racetracks are called by Equibase
employees. Some may have a side job working for the track (such as making
the morning line or writing comments for the program though that is NOT the
case here at Arlington Park.

Charts are called horse-by-horse at each point of call, such as, "1 by 2
(lengths), 3 by 1, 4 by 2, 7 by a head, etc...

The finish is confirmed by the photo finish camera and most chart callers
will refer to the video tape of the race to double check their work at other
points of call.

Lengths are subjective and called by the chart caller except at the finish
which can be verified by a camera placed directly on the line.

It's a visual estimation, and the tail of the horse is not included when
saying one length.

Field supervisors make several trips a year to check on the chart callers to
make sure they are doing a good job.

I do not believe there is a "written" handbook.

[cj]

What is the point of the live "chart call?" Just go straight to the video. I don't get it.

[chickenhead]

yeah, I like that, "most chart callers will go back and verify...."

It would be nice to know which do and which don't.....

[BillW]

The final is subjective also as a "length" has to be assumed at some point. Did they say what that was?

The photo finish data at HOU has an assumption it uses to calculate beaten lengths, but the placing judges didn't know what it was, as they weren't really interested in the beaten lengths. They told me that the Equibase caller didn't use that data anyway, so I didn't continue to press for an answer.

The live call must be a "preservation of the art" kind of thing. :rolleyes:


Bill

[chickenhead]

Here is my actual email:

> Hello,
> Please forward this mesage to the applicable party.
>
> I have questions regarding how specifically the results chart is called at
> each point of call.
>
> 1.) Is it handled by an Equibase Employee, or is the chart caller employed
> by your track directly?
> 2.) Do you know if all tracks operate on the same basis, or do some
> employ/some not employ their own chart callers.
> 3.) The chart caller calls the distance ahead each horse is at the time that
> horse passes through the point of call, correct?
> 4.) At the finish, or at any point, are the lengths taken from video
> playback, or are they taken live at all points?
> 5.) If they are taken from video at any point, are the lengths based on a
> visual estimate or calculation, or are they extrapolated from the time it
> takes for the next horse to hit the call? If they are taken from the time,
> what time unit is used to estimate one length? 1/5th of a second?
> 6.) In all cases of visual estimation, what is used as a guideline for 1
> length? Is it a freefrom calculation, or do the chart callers use point of
> reference such as the uprights, or do they do it any way they please?
> 7.) Do you test for accuracy of the chart callers?
> 8.) Is there any documentation or handbook or guide for chartcallers in
> general? If so is it available to the public? If so can I have a copy?
>
> I apologize for the length and scope of these questions, but I would be
> extrememly grateful if you could provide answers to any of them,
>
> Regards,


I sent him back one already asking about the finish, from Game Theory's post as well as my own recollections, I thought the photo finish field of view was too small to see far enough back to a horse that is say 10 lengths back, so I asked him whether it is actually the photo finish, or the video at the finish.

He didn't answer specifically what the "length" is they are using, but from his comments I get the feeling it is a rough visual estimate of the length of a horse. Since he said specifically the tail is not used (no kidding), I guess that means that guestimate is based on a fully extended horse, tip o' the snout to rump.

[BillW]

The equipment I saw was digital. The whole field was caught in the data. The software was set up such that the operator could scroll through the data and click the mouse on the nose of each horse and enter the associated number. From there it had the times and calculaued an (apparently unused) beaten length figure.

The "photo finish" isn't really a photograph (snapshot) but rather a narrow apperture continuous "picture" of the finishline.

Bill

[chickenhead]

that's what I don't understand, he says the chart caller uses the photo finish pictures to verify the finish lengths, but what you said, and what he himself says means they can't be using the photo finish pictures, as they are just a continuous picture of a point in space.

The picture you saw in Hou, that was an actual snapshot of when the winner passed the line? A point in time? If they took those as each horse crossed the line, then the chart callers could use them I suppose.

So, I still don't know what the chart caller uses to get the data that shows up in the form.

[Figman]

The system described by Bill W is what is also used at the New York Racing Association tracks. The timing and the photo finish system is operated in unison by the same company. Times and beaten lengths are captured by the photo finish system at the finish line. Times are recorded to the one/thousandth of a second. Beaten lengths are determined by computer by pulling each horse through the finish picture that is determined through an aperature that is surveyed directly on the finish line of each race course. NYRA at Belmont Park has four cameras, one for the main dirt track, one for the Inner Turf Course and one for the Widener turf course and one backup camera.

The Equibase people receive the final splits and final time and beaten lengths at the finish from the photo finish operator.

The Equibase chart caller determines by "eyeballing it" and later looking at replays in determining the positions and lengths behind at all other calls other than the finish. Needless to say, this archaic system should be modernized to match the state of the art finish line system!

[rokitman]

Count how many times you get "subjective" or "estimation" or their equivalent in the responses.

Next time I get pulled over by a cop for speeding and he asks the proverbial "Do you know how fast you were going?" I'm going to give him my estimate in Telephone Poles Past Per Minute. He'll have to handicap my estimation of what a minute is for himself.

[GameTheory]

The way it has been traditionally done for the finish is that the time from one nose crossing the wire to the next is taken from the photo finish camera -- the time is known because the speed of the camera is known (the camera doesn't take frames, but runs continuously). This time is then converted to lengths based on a rule of thumb (1/5 = 1 length, for instance. Older equipment didn't even have an explicit timer -- they used a chart that converted the photo finish "pictures" -- which are made up of hundreds of small fragments to times/lengths.). This rule of thumb is not neccessarily the same for all tracks, and the photo finish equipment isn't either. The actual distances between horses are never measured by any equipment -- only the time between one horse and the next. Which is why I say beaten lengths are time, not distance, at least at the finish. All this was true 10 years ago or so and as far as I know that is still the basic procedure.

For the calls other than the finish the beaten lengths are whatever the chartcaller says they are.

[chickenhead]

this is truly fascinating to me, I'm sorry if it is basic for others, and i'm sorry it has taken me so long to understand it, I just never knew any of this about the finish.

GT, if what you say is right, then the most accurate way to time beaten horses at the finish is to use plain ole 1/5 sec. per length (or whatever they actually use), and at all other points of call use velocity and some number of feet per length to convert to time. Since they are called different ways we should account for them different ways.

It had never occured to me before I started these threads that that would be the case, but it sure looks like it is.

[BillW]

Originally posted by chickenhead
that's what I don't understand, he says the chart caller uses the photo finish pictures to verify the finish lengths, but what you said, and what he himself says means they can't be using the photo finish pictures, as they are just a continuous picture of a point in space.



Correct ... the old analog systems were the same, hence the dirt on the track appearing to be in motion. i.e. the apperture is opened when the lead horse approaches and data is continuously collected until the last horse finishes. The finished product looks like a strip photo with the horses aligned across it in relative finish position.



The picture you saw in Hou, that was an actual snapshot of when the winner passed the line? A point in time? If they took those as each horse crossed the line, then the chart callers could use them I suppose.



It was on a computer screen (i.e. software collected the data from a digital camera interfaced to the computer) and scrollable across the whole field. This could have been easily networked to the chart caller who was next door, about 30 feet away. They (a placing judge) told me that the chart caller didn't use that data though. I assume he knew that as fact.


So, I still don't know what the chart caller uses to get the data that shows up in the form.

Yes I agree with you, what I saw wasn't conclusive. It was difficult to get much information as I was observing this while these guys were working. I got lucky and caught a placing judge out in the hall heading to the restroom. I haven't been able to arrange a get together with the chart caller yet. Quarter horses are running now and I'm not even sure if the same caller is working, but I'm going to continue trying.

I believe the use of the tracks photo finish data is a matter of personal taste and convenience though. Possibly the software, as typical with alot of packages in Windows was using a propriatary data format that did not easily lend itself to network access.

Bill

[chickenhead]

p.s. GT, BillW and others, I will try and call Equibase and see if I can't find out what time constant they currently use.

I did get confirmation from the Arlington guy that, as Figman says, the chart caller has nothing to do with the finish, lengths or otherwise, it all comes off that picture you saw Bill. The chart caller only has his hand in the internal calls.

[GameTheory]

The problem is that something is lost in the conversion when rounding off to lengths. Either that, or as BillW says, the chart caller doesn't actually use the photo finish data (I don't think I believe that, and I have been told otherwise, but it could vary from track-to-track. Certainly the chart has to correspond with the official order of finish, so the chartcaller must at least double-check that will the photo finish guy.)

Anyway, if you were to time the gaps between horses (I have) and then look to see what is in the charts, you will find no consistent time value per length that will fit just right, even at the same track.

[NoDayJob]

The only way you can take subjectives, like a length and exactly where the horse is at a certain point in the race, is to use an electronic monitor for each starter. I believe this was tried in Japan a few years ago. Since handicapping isn't an exact science we'll just have to get along with educated guesses and averages. I don't know about you but, it works for me.

NDJ

[BillW]

Originally posted by chickenhead
this is truly fascinating to me, I'm sorry if it is basic for others, and i'm sorry it has taken me so long to understand it, I just never knew any of this about the finish.


It's not just you. As this thread has demonstrated there are a lot of questiong that need to be understood in order to properly use the data. Until six months ago I assumed that beaten lengths were just measured out on the finish video until I saw the contorted pictures of thoroughbreds in full stride. No way in hell :D .

The quest continues ...

Bill

[chickenhead]

GT, thinking about what you said about rounding off....they round to a quarter of a length in the chart when beyond a neck, right?

If I get the constant can see if the error fits within that 1/4 length window.

Am I right in my thinking?

[GameTheory]

Originally posted by chickenhead
this is truly fascinating to me, I'm sorry if it is basic for others, and i'm sorry it has taken me so long to understand it, I just never knew any of this about the finish.
Actually, almost no one thinks about this. Take Beyer -- he claims that the value of a length changes depending on the length of the race because they'll be going slower at the end of longer races (which is true, logic-wise). But if they use the photo finish camera and it always runs at the same speed, then the Beyer method of making figures is built on a foundation of sand...

[chickenhead]

I know, that is what is amazing to me! I thinks it's quite exciting to think that most people, even a lot of the experts, might not be using the data correctly. Like you said, screwed up data is good if you know how it is screwed up.

As a Radio Engineer, I guess I'm just used to this kind of thing, you have to adjust data all the time to account for different types of measurements, but you have to know how the measurements are made and what the margins of error are.

Since I'm thinking about trying to become a little more serious about handicapping, it just made sense to me to start with the basics, what are the PP's, how are they built, how are these things measured.

I think this is great!

[Latin Qtr]

I thought Beyer said that the value of a length changes w/distance because 1 length means LESS proportionately
as the distance increases.

[Figman]

In a previous life, I was a "charter" at a harness track with very old equipment. The photo finish was projected down to the judges on a "light box." If the judges couldn't make out the finish projection (too close) they call for a "print"(hard copy).

After the race was declared official by the judges, the photo finish man called down to me the order of finish. Then he called to me the number lengths each horse beat the next successive one. His call might consist of "a length and a quarter, neck, neck, head, three and a half, seven, etc. I had to add them up to make the chart. In this case, the winner won by a length and a quarter and the second horse was beaten by a lenth and a quarter while the third horse was beathen by a length and a half, etc.

This was the given....a 1/4 length equalled a neck; two heads equalled a neck; three noses equaled a head.

[cj]

Beyer uses time, not beaten lengths, to construct his chart. Of course there will be an error built in when adjusting for beaten lengths, but that is not the foundation. I agree on the beaten lengths. As a matter of fact, when I make figures, I do use the Beyer scale, but as I say on my site, I don't agree at all with his beaten lengths scale and substitute my own. This is part of the reason.

[GameTheory]

Originally posted by Latin Qtr
I thought Beyer said that the value of a length changes w/distance because 1 length means LESS proportionately
as the distance increases. I don't actually remember the Beyer method very well, but he does change the time value per length depending on distance, doesn't he? (At least sprint vs. route.)

[BillW]

Figman,

Interesting post.

At that time, did Equibase (Am I making a bad assumption that you worked for Equibase?) have a standard set of procedures describing this process to insure uniformity from track to track or was it up to you and the host track to work out the details of how this was done?

Bill

[chickenhead]

Brohammer uses 10 ft. per length everywhere in MPH, if lengths are measured in time at the finish, then depending on the velocity of the horse he could be off by a fair bit. I know he was worried about velocity and not time, but since he got the velocity from the time.....

I just can't think of why you'd "throw away" accuracy where it does exist, considering how hard it is to come by. I would guess most of it comes from the legacy of doing it all by hand, it was just too hard.

To me, like I said earlier, the most intriguing thing is that since the internal calls are done differently than the finish, you can account for them two different ways and come up with better numbers. I have NEVER read that in a book, could be I've been reading the wrong books.

But then even on top of that, like I talked about in my other thread, by using the charts you can can get a better idea for the internal calls than you can using the PP's by calculating the velocity for all horses. I'm not sure anyone followed my thinking on that.

And then on top of that I think there is still some valuable work to be done characterising the average distance that is represented by the chart callers "lengths" at the internal fractions. This is another thing I haven't heard anyone do before, maybe because it is too tedious, maybe someone did do it but never talked about it because the results were too ambiguous. Even if individual chart callers are erratic, if you average them all individually, and there were some stand out offsets between the averages, like one guy on average called them a bit longer than another guy, that would be extremely valuable.

Maybe I'm nuts, but I want to make sure I have the most accurate data before I worry about how to analyze it.

[cj]

GT,

The figure for each race is based on the time of the winner. The others are then based on this numbers, minus an adjustment for BL. He has a chart for this that attempts to convert BL to points (time really.) This is where I think he errs a small bit. But to be fair, it doesn't make much of a difference until you get into big numbers of beaten lengths. At that point, the figure doesn't mean much anyway. I use my own chart mostly for the pace calls, though having better BL adjustments at the finish is an added bonus.

[rokitman]

Your nuts for math, it seems, which puts you in the right avocation.But I think you need to do a cost-benefit analysis on the mountain of work your looking at doing. The sophisticatos, some of which your conversing with in here, are already making adjustments to the odds line well beyond the common PP/charts/fig information. Do you have specific reason to believe there is large enough unknown innacuracy to be meaningful? The answer to that is almost always no in the end. The charts/pps/figs may have their innaccuracies but the pools are VERY efficient. Know how the work your looking at doing would translate into a significant separation from the betting line. In the end, that's the only fig that matters. Take it from somebody who has done a ton of promising research which yielded nuthin, as have plenty of others in here.

[GameTheory]

Originally posted by rokitman
Your nuts for math, it seems, which puts you in the right avocation.But I think you need to do a cost-benefit analysis on the mountain of work your looking at doing. The sophisticatos, some of which your conversing with in here, are already making adjustments to the odds line well beyond the common PP/charts/fig information. Do you have specific reason to believe there is large enough unknown innacuracy to be meaningful? The answer to that is almost always no in the end. The charts/pps/figs may have their innaccuracies but the pools are VERY efficient. Know how the work your looking at doing would translate into a significant separation from the betting line. In the end, that's the only fig that matters. Take it from somebody who has done a ton of promising research which yielded nuthin, as have plenty of others in here. Yeah, I should have mentioned that getting it all right (which is impossible anyway) won't really get you much. If you want to make money a hell of a lot quicker, forget all about times and study trainers & pedigree.

[BillW]

Originally posted by GameTheory
Yeah, I should have mentioned that getting it all right (which is impossible anyway) won't really get you much. If you want to make money a hell of a lot quicker, forget all about times and study trainers & pedigree.

It's a lot easier to brush something off as inconsequential that you understand.

Bill

[chickenhead]

Thank you for the perspective rokitman, I undertand what you're saying. The first two things I'm talking about are no extra work at all. The last one, well, that is probably something I do not have time for in any case.

But it's hard to know ahead of time whether something makes much of a difference until you try it. Considering how archane the method is for making charts, and considering how much (most)everyone relies on them as fact, it could be there is something there, no one will know till they try I guess.

Since it's such a basic, underlying thing we're talking about, how fast, and in what way, did a horse run...at face value it looks promising.

I would guess most likely is at some point, they will eventually present much more accurate data. It seems like the biggest cost/benefit analysis varaible might be how long are they going to continue doing things the same way.......

thanks everyone for their input, you have all helped me a lot.

[chickenhead]

Originally posted by GameTheory
Yeah, I should have mentioned that getting it all right (which is impossible anyway) won't really get you much. If you want to make money a hell of a lot quicker, forget all about times and study trainers & pedigree.

Those come next :)

I figure since this is a zero sum game, or actually a minus 20% sum game, if I have any chance at all, I have to start by looking at every single thing in isolation, try to understand it, and look for ways to improve upon it. If I were smarter I could brush things aside, but I'm not.

I get excited sometimes, don't hold it against me everyone, I absolutely respect all of your knowledge immensely, that is why I'm here, to learn. Thank you all for your patience.

[kenwoodallpromos]

This is a great thread! Only the winner is timed; the most common unit of race measure, the length, cannot be readily explained by chartcallers; almost everyone uses lengths to handicap but cannot define them; lengths are off by 25%; race trimes are useless by far; when you figure it all out it rains or they dig up the track!
Chickenhead-the most common barrier to a horse winning is purse size. That is how trainers place their horses.

[Tom]

Originally posted by kenwoodallpromos
This is a great thread! Only the winner is timed; the most common unit of race measure, the length, cannot be readily explained by chartcallers; almost everyone uses lengths to handicap but cannot define them; lengths are off by 25%; race trimes are useless by far; when you figure it all out it rains or they dig up the track! .........


Geez, Ken, you make it real easy for me to sit down and work on my speed figs tonight!:(

[JackS]

If racing began using overhead holographic imaging and doubled or tripled the points of call and switched to the metric system using millimeters to meters, we would all have a much more accurate picture of the race. Unfortunatly, everyone would have the same information. This would put us back into the exact same handicapping situation we're into today. I like the universal inaccuracy. This allows for some independant thinking rather than a slide rule. It also allows for many of the uneducated in horse racing to donate to our intense cause. The cause is to get our own hands on their money. Anyone truly interested in racing should be expected to learn their own lessons using their own money and dedicating months , years and indeed a lifetime to learn.

[kenwoodallpromos]

So tell me again why horseracing is not the #1 sport among young people?LOL.
Tom- Beats studying chess. Beats science research- horses are bigger test animals for our research than white mice.
I think the metric system was dropped after big business figured out they had to give more quantity in an litre than a quart and more in a meter than a yard. less profits. At least the betting windows take our metric money.

[JackS]

Dedicating a lifetime to inaccuracy might eventually lead to some independant thinking and the very educated "hunch".

[kenwoodallpromos]

Of course, and overcoming inaccuracy better than the next bettor.
My conclusion is: objects (losing horses) are closer (by 25% elapsed time) than they appear (to chart callers).

[chickenhead]

combing the deep dark basement of PA on other subjects entirely, found this thread....no big surprise but this topic has been covered quite a few times....

http://www.paceadvantage.com/forum/showthread.php?threadid=5901&perpage=15&highlight=deceleration&pagenumber=2

Mr. MV McKee apparently actually did what I was talking about doing, don't know if you're still around or not.....interesting thread all the way around

I so want to ask why you didn't go the next step, and turn those times back into feet??????Your times will vary, even at the same call, all the horses are running different speeds! The more lengths the bigger the error, even if the chart caller was perfect! Do you still have the source data in a notebook somewhere?

[so.cal.fan]

No matter what the accuracy or lack of is.....there is always the factor of what effect did this race and the subsequent time in between have on the condition of the horse.
That is the biggest X factor.....always has been, always will be.

[Latin Qtr]

There are some great past threads & posts here. Whose property is the past threads & posts? I assume PA has the
rights?

[chickenhead]

Originally posted by chickenhead
I so want to ask why you didn't go the next step, and turn those times back into feet??????Your times will vary, even at the same call, all the horses are running different speeds! The more lengths the bigger the error, even if the chart caller was perfect! Do you still have the source data in a notebook somewhere?

I shouldn't have put it that way, you didn't convert to feet, because that was not what you were looking for. You got what you needed. I don't have to ask.

[chickenhead]

Originally posted by so.cal.fan
No matter what the accuracy or lack of is.....there is always the factor of what effect did this race and the subsequent time in between have on the condition of the horse.
That is the biggest X factor.....always has been, always will be.

I agree completely.

[JackS]

Pray Dear Lord, God Almighty that this never happens. I'm keeping the faith. Amen...... I'd rather have one jockey caught with a "buzzer" now and then, than ten jockeys carring GPD's in every race.

[chickenhead]

I agree, I hope you read the other posts in that thread, Ralphs wasn't the one I was referring too.

I'd rather give all jockeys buzzers than have GPS, we could chart the voltage/FPS ratio.....

[GameTheory]

Originally posted by Latin Qtr
There are some great past threads & posts here. Whose property is the past threads & posts? I assume PA has the
rights? Actually, no. PA couldn't publish a book of posts, for instance. The rights belong to each individual author of each individual post. Obviously by posting a post you are giving implicit permission (implied license for limited use) for it to be published on this web site, archived, etc. Posting on the internet is very much like sending a letter to the editor of a newspaper...

[kenwoodallpromos]

You may be right, but why are some letters made into books (Penthouse)? There may be an implied copyright, but someone earlier said handicapping systems are not copyrightable.
I do know that the exact form something is on is copyrightable as a complete "compilation"
I think PA is up on all that.

[chickenhead]

got a real nice, long email back from the Bay Meadows chart caller Darrel Hove. He invited me up to the box next time I'm down, so I'll definately take him up on that.

He said he is "next to certain" that for thoroughbreds the finish is calculated at 1/5 sec=1 length.

For fractions, sounds from him like they start calling quite a ways before the pole depending on field size, so they have time to get the last couple of horses before they pass the pole. I'm assuming this is exaggerated most the earlier the call, the faster they are traveling.

Length is a visual (to him) length of a horse, says non chart callers over-estimate lengths 99% of the time.

He says a human call will give you a better idea of how a horse actually ran than a camera at each pole. I don't know where he is coming from with that, so I'll reserve my disagreement.

[BillW]

Originally posted by chickenhead
got a real nice, long email back from the Bay Meadows chart caller Darrel Hove. He invited me up to the box next time I'm down, so I'll definately take him up on that.

He said he is "next to certain" that for thoroughbreds the finish is calculated at 1/5 sec=1 length.


I assume he is talking about "length behind" data taken from the photo finish gear?


For fractions, sounds from him like they start calling quite a ways before the pole depending on field size, so they have time to get the last couple of horses before they pass the pole. I'm assuming this is exaggerated most the earlier the call, the faster they are traveling.

Length is a visual (to him) length of a horse, says non chart callers over-estimate lengths 99% of the time.


Interesting. I've noted a lot of times that a race caller will call a horse in the lead by 4 lengths when he looks like he's 6 or 7 to me :D


He says a human call will give you a better idea of how a horse actually ran than a camera at each pole. I don't know where he is coming from with that, so I'll reserve my disagreement.

Ask him to clarify .. I would be interested in hearing his reasoning.

Another question that would be interesting to get answered is re. the uprights of the inside rail .... are they spaced to act as a visual guage for calling lengths?

Good info chickenhead. Hope you enjoy the tour of the booth. Try to get a look at the photo finish process. You may also find it interesting.

Bill

[kenwoodallpromos]

So BM guy eyes it; At least 84 tracks have 8' rail sections; Tv today they said Belmont has 9' rail sections to figure Sec's 31; more information dispensed to bettors without the slightest interest by the racing industry of setting standards or anybody really picking winners.
That is why I do not use speed figures and only use beaten lengths to figure generally gains late in the race.
Many college math professors have gone into racing (Sartin, etc) and written books and/or systems, and many gave up and gone back into teaching. I think Ainslie even gave up.
Meanwhile, after many decades, the public and public handicappers still hit no more than 35% winners and have a negative ROI.
No thank you. For now I will keep researching all aspects of racing but stick to my methods.
By the way, I have a couple of emails out to racetrack railmakers and will let you know what I find.

[JPinMaryland]

Copyright issues:

"No," Ken. FOrms are not copyrightable. Not under any circumstances. This is on of the most basic, and one of the most agreed upon aspect of copyright law.

A form meaning some blank sheet w/ spaces for where you put numbers or whatever. Like a form for an accountant, or a tax form, or a questionairre, whatever.

I dont know if that is what you meant in the post above, but forms as I understand the term are not copyrightable.

[kenwoodallpromos]

By form I meant "arrangement". Sorry- long time since I copyrighted stuff.

[JPinMaryland]

Okay you are talking about a compilation of posts from an internet thread then, yes?

Hmm. You could copyright this compilation, but I think individual posts might have to get permission for their use. ON this latter pt. that would be some issue to litigate though, I dont think it is at all clear.

For instance, they published excerpts from the letters of JD Salinger. His papers were kept in some library. Certain short passages e.g. less than 200 words may not be copyrightable on the reasoning that you cant just copyright every time someone utters a sentence.

"Hey look at that horse." Copyright 2004, jpinMaryland.

See what I mean? YOu can just copyright a sentence. Or two or three.

I.e. large a passage does it have to be to get a copyright? Answer soemthing more than 100-200 words.

If that reasoning is good then certain short posts would not be protected for the individual that wrote it but a large post would be.

Coypright jpinMaryland. 2004.

[kenwoodallpromos]

You should check with Copyright office sight. I understand exact wording in works is protected but not paraphrase if different enough. Also, excerpts for purpose of reviews, quotes to prove a point, etc. are allowed. "Substantially the same" is protected.
On this forum, all information posted is in the public view already and the legal copyright is PA's for content, but asking permission of originator is polite.

[JPinMaryland]

Your last sentence makes no sense to me. You seem to suggest that because this forum is a public forum the copyright protection is lost. What statute or legal decision is the basis for that?

[GameTheory]

When you post you are giving implied license for the post to be distributed/copied in the normal ways one would expect when posting to an internet forum -- that it will be downloaded (automatically) and read by whomever happens by, maybe it will be printed out, etc. You do retain your copyright, however, and you are giving no implied license for anyone to do anything that would not normally be expected, like publishing your post elsewhere (in a book, or even on another forum). Obviously there is gray area here, but all of copyright is a gray area...

[JPinMaryland]

I agree w/ Game Theory's post and I think my comments are in agreement. His summary makes sense to me.

As he points out, lots of issue in copyright would need to be litigated before a final conclusion can be drawn. As in many legal cases, the actual facts sometimes are more important than what the legal "experts" say. So as he says, it is somewhat grey area of the law.

Does anyone have a published legal case that can confirm or deny our conclusions?

[sjk]

I don't care about the copyright issues, but the original charting issues have been bugging me since the thread was started.

It is amazing to me that the charts can be assuming 1/5 sec per length at the finish since that is totally contrary to visual reality.
Any further insight into what is being used in chart creation would be appreciated.

[BillW]

Originally posted by sjk
I don't care about the copyright issues, but the original charting issues have been bugging me since the thread was started.

It is amazing to me that the charts can be assuming 1/5 sec per length at the finish since that is totally contrary to visual reality.
Any further insight into what is being used in chart creation would be appreciated.

sjk,

The problem we as handicappers have is assuming a length is anything but what the chartcallers define it as. Whether it is 1/5 of a second or not is an academic argument.

Bill

[sjk]

My interest is in what the chartcallers do. Your earlier post indicated that the Hou chart caller did not use the photo video.

It is obvious that horses in a high class sprint are moving at a much different speed than horses in a low class route so if there is an assumption that lengths always equate to the same speed, it is patently abusrd.

I have always translated lengths back into time based on the speed the horses were moving at the finish. This is not totally accurate, but for it to be close, it relies on the lengths in the chart to be close to the visually observed lengths.

For the chart information to be based on a totally absurd and incorrect assumption would not be beyond belief. If it is true, it is comical that tracks would purchase digital equipment to fine-tune data-collection which anyone could better with one eye closed.

[GameTheory]

Since the photo finish camera only records a tiny slice of actual space (the finish line itself -- just an inch or two, maybe less) and builds up its pictures from hundreds of tiny such slices, the actual lengths behind (distance wise) are not recorded. Only the time between one horse crossing the wire and the next is known, so this must be converted to "lengths" somehow (it should just be reported in actual time and all this nonsense would end). I am quite sure they are not calculting velocities of individual horses. However, it has been suggested that the speed of the camera is altered with the distance of the race because in order to get clear pictures of the horses crossing the wire, the camera speed needs to be in sync with the speed of the horses at the finish (because it doesn't take frames like a normal camera, it runs continuously). Whether that is true or not about the speed changes I don't know. What the effect of digital technology has on all this I don't know. They probably don't use actual film anymore, or maybe they do.

I do know that if you time the horses and then use the beaten lengths in the chart, you will not find any neat value like 1/5 second that will make all your times match up as they should. At least I couldn't find any such value, even at the same track. I believe MV McKee has said he has found constant time-per-length values for certain tracks, but no constant for all tracks. I haven't found the former or the latter, and have gotten more accurate times by calculating individual velocities as suggested above. However, I did have to put some ad-hoc adjustments in the formula to make it work well....

[sjk]

GT,

Thanks for the reply.

In a high class sprint race, 1/5 second is close to 3/4 length as judged visually (2/5 is close to 1.5 lengths; 3/5 close to 2.25 lengths). You would have to think that if these were routinely reported as 1 length, 2 lengths, 3 lengths, there would be people who would notice and complain.

If different tracks calculate their lengths through different methods, it would be worth understanding which is which.

[BillW]

sjk,

I'm expecting to reach a conclusion that some chart callers use the photo finish data and some don't (this won't leave me with much by itself)

I don't disagree with your techniques. They make sense. The only question I want answered is "is my (your?, any?) technique close enough to reality (what the chart caller really does) such that it cannot be differentiated from that reality. I'm not even close to any conclusions. Hopefully we can get enough data to put this puzzle together. Who knows, it may turn out to be the most absurd answer. :eek: or it may turn out to be just not that critical. I sure would like to at least know though.

The .01 second precision reported in Quarterhorse racing sure would be a nice alternative although I cringe every time I think of the general public getting more information handed to them on a platter. This game is challenging enough for me as it is. :)

Bill

[chickenhead]

Originally posted by sjk
GT,

Thanks for the reply.

In a high class sprint race, 1/5 second is close to 3/4 length as judged visually (2/5 is close to 1.5 lengths; 3/5 close to 2.25 lengths). You would have to think that if these were routinely reported as 1 length, 2 lengths, 3 lengths, there would be people who would notice and complain.

If different tracks calculate their lengths through different methods, it would be worth understanding which is which.

At 10 ft. per length you're saying stakes sprinters are finishing at 37.5FPS (7.5X5)? Seems a little slow.

Here are McKee's final time adjustments by track, all over the map as you say GT. But, all between 1/5 and 1/6 sec/l

Emerald Downs .17 per length
BayMeadows .21 per length
Golden Gate .20 per length
Hol, SA .18 per length
Portland Meadows .16 per length

Don't know if these are all distances averaged or what.

[sjk]

chickenhead,

I have always used 8 1/2 ft per length.

I figure 660 ft per last 1/8 mile = 77.6 lengths

If they finish at a rate of 12 sec./1/8 mile you get about 6 1/2 lengths/sec.

If they finish at a rate of 15 sec./1/8 mile you get about 5 lengths/sec.

That is about the range you would normally see.

[chickenhead]

ok. you must have been talking a different kind of lengths in your example. my bad.

[sjk]

Chickenhead,

You're quite right; I got my example upside-down.

To correct:

In a high class sprint race, 1/5 second is close to 1.3 lengths as judged visually (2/5 is close to 2.6 lengths; 3/5 close to 3.9 lengths). You would have to think that if these were routinely reported as 1 length, 2 lengths, 3 lengths, there would be people who would notice and complain.

[chickenhead]

none of the chart callers I've talked to seem to really know how the finish is calculated. I think to get that info have to talk to the photo finish operator. But, some of them may not know for sure.

One question I have for you guys, when calculating the velocity, do you use average velocity, or do you assume some kind of straight line accel or deccel based on trend, so your velocity number is always above or below the average?

[kenwoodallpromos]

I think official statistic makers and recorders refer to handicappers as "mushrooms".
It is good you are interested in researching, but you have to decide what is out there to be found and how much value it has to your handicapping.
I have found that there is certain information the racing industry is not interested in standardizing and/or diseminating to the public until enough people start using it. Then they want everyone to use it so things even out.
In their mind it is ok if horsemen are privy to information but not a small % of the public. This is historically true of workout info, generally reliable speed figures. and timing of all horses in races.
When too high % of the bettors pick the winning horse the industry has to find a way for lesser horses to win races and be picked. I can give you a list of the ways this is done.
It is a matter of how much time you want to devote to exactly measuring the space between 2 horses' finishes. ***Remember to look for patterns, not just raw individual numbers. Otherwise you will end making the common mistake of only looking at a horse's most recent activity.

[kenwoodallpromos]

I am not a chartcaller, but it is obvious that the finish has to be calculated the same way as the fractioal beaten lengths- by counting the posts. Each rail post represents a section of the rail and that is the only reasonable way to measure lengths.
I have an email in to the people that make rails as to which tracks use which length rail. As I stated, at least 85 in Northa America use 8' rails and Belmont uses 9'.
For all races except fast sprints: at Belmont lengths x .16= seconds beaten; all others use lengths x .14= seconds beaten. For fast sprints multiply x .15 at Belmont and .13 at all other tracks.
I will let you know if I find out anything further.

[chickenhead]

Most people in this thread seem mainly interested in final time, which makes sense from a speed figure point of view.

I'm more interested in energy and pace, so I'm equally interested in all points of call, including final time.

[sjk]

chickenhead,

You ask what velocity we use. I use the race velocity from previous point of call; i.e. 1st frac/1st feet at 1st call, (2nd frac-1st frac)/(2nd feet-1st feet) at 2nd frac, etc. This overstates the velocity especially for the stragglers, but it has worked well for me.

[chickenhead]

ok, a few more tidbits, and then I'm done.

BillW -- I've asked this guy twice if he uses uprights, he hasn't answered me directly either time, but says 10 ft. is a good rule of thumb, 33 lengths per 1/16th.

I asked about stretch calls, he said yes they're the toughest at most tracks, except Santa Anita, where the chart caller is situated between the 1/16th and the 1/8th (thought that was interesting, probably the most accurate calls to be found).

The other thing I thought was interesting, he calls if from the leader back, i.e. as if he's really describing beaten lengths when the leader hits the pole, i.e. he does not wait if say the third place horse at first call is 5 lengths back, he does not wait for him to get to the pole before calling lengths back to the next horse, once he starts he keeps calling, so it is essentially beaten lengths back to the leader. This is the exact opposite of what I thought they did.

This is a snippet that covers final time and sums up the discrepency fairly well:

<<Many of your "expert" (another term I definetly use loosely) handicappers do not understand this. And, that is why it is ridiculous when somebody tries to tell you that a horse ran his final quarter (or whatever) in a specific time. They are using two totally different type of calls to determine the time...(1) the chart caller who is telling you exactly (there's that word again) how many lengths behind a horse is when the leader hits the quarter pole and (2) the photo finish camera that tells you how far he is behind when he actually hits the finish line, not when the leader crossed the line.>>

I think that sums it up pretty well, best to account for them differently.

sjk, you talked about the difference between visual reality and using a time constant instead.

<<For instance....Horse A wins the race and as he crosses the finish line, Horse B is second (maybe five lengths back) and Horse C is third (say, six lengths back).
By the time Horse B and Horse C reach the finish line, however, they easily could have changed positions. And, since the photo camera is based entirely on time elapsed, that winning margin is determined upon how long it takes those two horses to get the line. If the horse who finishes second is closing very fast, that margin may come out less than five lengths, or if they are stopping, will be more than five lengths.>>

His earlier comment about the chart caller giving a better perspective, he expanded on that, and what he was basically saying was it would be best to be consistent, i.e. either call the finish the same way as the internal calls, or vice versa.

Then he went into talking about distances to the beam, but that doesn't really fit this thread, although equally frustrating.

so yes, all pretty crazy huh? Turns out the internal calls (which aren't necessarily called at the poles, the bigger the field the further in front of the poles they start) give you beaten lengths from the leader, and the finish gives you time. Beautiful!

[kenwoodallpromos]

Sounds like the chart caller you were talking to told you he does not use the railing or uprights for lengths behind, but rather uses the various poles set every 1/16th mile around the track (1/4 pole, 1/8 pole, 1/26 poles, etc.). If that if the guage used by all or most chart callers it would be sort of a "standard", except of course it is basically wild guessing based on experience within a 330' section of track.
If they only verify at one point of the track - the finish- using a camera, that call would of course be more accurated as to the position of the horses when the leader crossed that point, which is thew only way that lengths back can be measured at any point in the race.
Chart callers are all measuring distance only. Trying to convert to time as accurately as possible requires exactly what you said- getting a ft/sec measurement of each horse at the point of call, then taking the lengths, figuring the rate of speed as adjusted for the particular horse, then projecting that horse and time to the point of call, assuming that the horse ios running on a portion of the track that is equally as fast as the portion the indiuvidual horse has already covered and would not have changed paths to another part of the track, like a slow rail; or that the horse is not about to head into the wind.
Once you obtain as close as you can get to the exact time the horse ran, you should hope the next time it runs the track speed and condition does not change!
Lengths= Since at least some chart callers use the finish canera to count uprights at 8' or 9'; and internal lengths are eyeballed at either 8', 9', or 10'; since most horses run closer to 12' per 1/5 second, especially non-leaders, I would say just use the .16 second per length and be within 20% of exact for all fractions for all races.
At least until the timers used at many tracks for races run ina divisions, like futuritys, are used for all races. LOL.
___________

This is what I meant by intentionally no set standards in racing information. just 1 example. And just 1 way to make races unpredictable and keep the payoffs from being even lower.