Is it possible that there are no real speed figs? [Archive] - Page 2 - Horse Racing Forum - PaceAdvantage.Com

steveb

03-29-2016, 07:20 PM

You can restate what you just wrote in a much more pleasant manner.

After all, we're not talking politics or religion or even race fixing here. Why the need to get all uppity?

thanks pa, but it is water of a duck's back to me.
i long ago stopped caring what people say about me, or that i can't do this, or i can't do that.

you need to remember that he needed to get his 'time only matters in jail' written and have punch to it.
pretty sure though that i have read it before, so he needs to think of something more original to denigrate.

there is NO right or wrong in racing, all there is, is an endpoint that everybody hopes to reach.
although it is difficult to figure if some people have reached that goal(not that i care either way), with others it is simple.

Dave Schwartz

03-29-2016, 07:28 PM

i don't know what has rocked your boat dave, not that i care.
i read your stuff now, and it is nothing like what i do, which i already knew it wouldn't be.

i am here because mr pace advantage has not banned me, and thus lets me contribute if i feel like it.

as for calling me out, that is not too clever, i have put lots of stuff on here over time and explanations.
and what is more i am not trying to flog anything to anyone.

it is what it is, and it's not the latest and greatest, because i have been doing it for what seems to be forever.

but if you insist i will explain it all, but first YOU put all your stuff up here for gratis, that you charge people for?
deal?

every single time a meeting/s are added to data, everything is updated.
if it was time consuming do you really think i would do that?
you are a programmer i thought.
computers are pretty quick, even when used by self taught crappy code writers like me.

I am not even sure what you are saying above. Sure, I got the obvious points - give everything away, blah-blah-blah.

The point is that you trivialized making any kind of figures and it is just not a trivial thing.

You basically said that you could build pars in a matter of minutes. I do not believe you could do that. Very simple.

Sure, there are some tracks where the planets line up and the pars just fall together. And the better my tools get the more often that happens. But there are also tracks where it is like the planets are from different solar systems.

The idea that building pars by averaging a few columns of numbers is just a dream.

Regards,
Dave Schwartz

PS: I actually started serious coding on an Apple II back in 1978. Then I did consulting on the CBM-32 (Commodore Business Machines). That would be the one that pre-dated the Commodore 64 by about 3 years.

green80

03-29-2016, 08:43 PM

thank you green80 for such a revelation.
and the higher the class, then generally it needs to run faster.

if you understood what you can learn from time, you would not be stating things like that.
speed figures are just one tiny aspect of what you could learn.

i once upon a time shunned 'time', even during the first couple of years that racing was my life, and that was because the books i had read canned time, and the reason put forth made sense.
and me being an idiot, listened to all that nonsense.
then i read a beyer book, and it got me thinking.
that was maybe 30 years ago, and i just in the past year or so stopped betting, because i no longer had the urge to make lots of money.
i would rather watch good looking chicks on the beach these days.(feathered ones!!)

If the horse with the fastest time wins more often than not, we should all be rich.

pandy

03-29-2016, 08:49 PM

If the horse with the fastest time wins more often than not, we should all be rich.

You're over simplifying it. I wouldn't say that the horse with the fastest time wins more often than not, I would say that it wins more often that a horse with a comparable speed figure that has slower times.

An interesting thing is, most horseplayers who look at time probably rely on speed figures, and don't pay much attention to the actual time at all.

steveb

03-29-2016, 10:59 PM

If the horse with the fastest time wins more often than not, we should all be rich.

it is only one factor, and i doubt too many people would be rich from using one factor, regardless of what that factor was.

but you could get many different factors just from using time in different ways, and in my models are easily the most predictive factors.
but then, time is my go, so i guess that is to be expected, but even then it's not enough, you need to consider other things too.

but you won't do yourself any favours, if you want to make a go of something, if you just dismiss it.
perhaps it is just that YOU can't get any value out of it.
that does not mean others can't.

highnote

03-29-2016, 11:03 PM

I have been posting some speed figure comparison in my thread 'What's up with the Capper?' here in PA. My AMS and Euro speed figures are being compared to BRIS speed figures. (I'm a one man part time operation. So development is slow.) But anyway, my figures are proving to be able to compete with BRIS. I believe that I have the beginnings of cracking the global model.

What I can say is that I have found a way to interpret a global distance/surface chart. I needed to have some speed figures for Euro horse appearing in North American races.

I had written code for Mordin that allowed him to make U.S. figures that were on the same scale as his European figures. It was definitely useful to have figures for Euro shippers racing in the U.S.

The big challenge was keeping up with all the figure making. He kept up with it because it was his job as a journalist to write about international racing, but it was a tedious work -- even with a computerized system.

Dave Schwartz

03-29-2016, 11:03 PM

perhaps it is just that YOU can't get any value out of it.
that does not mean others can't.

That I can completely agree with!

:ThmbUp:

steveb

03-29-2016, 11:19 PM

I am not even sure what you are saying above. Sure, I got the obvious points - give everything away, blah-blah-blah.

The point is that you trivialized making any kind of figures and it is just not a trivial thing.

You basically said that you could build pars in a matter of minutes. I do not believe you could do that. Very simple.

Sure, there are some tracks where the planets line up and the pars just fall together. And the better my tools get the more often that happens. But there are also tracks where it is like the planets are from different solar systems.

The idea that building pars by averaging a few columns of numbers is just a dream.

Regards,
Dave Schwartz

PS: I actually started serious coding on an Apple II back in 1978. Then I did consulting on the CBM-32 (Commodore Business Machines). That would be the one that pre-dated the Commodore 64 by about 3 years.

this is the problem with the written word.....you have taken offence at what i wrote, when that was never my intention.

as for what you believe or not, do you really believe i give a fig what you believe?
it matters not.
i know i can do exactly that, and if you read post 144, AND understood it, then you would know that it is certainly possible.
maybe you just need to accept that your slow way, can be done quicker!!

as for predating mine by 3 years.....so what?
i simply had an idea, and knew i would never implement it if i could not write code, which I couldn't at the time.
and do you know what, i still can't write it well.
but a man far smarter than me(and you without doubt), once asked me, after i had told him that i am not good at writing code....."does it do what you want it to do?", and when i answered "yes", he asked "What is your problem then?"
lesson learned.

03-29-2016, 11:29 PM

pandy

03-29-2016, 11:47 PM

So a horse that gets a 90 figure at Churchill with a 1:09 will beat a horse that gets a 90 at Pimlico that gets a 90 in 1:09.3? That is pretty archaic thinking, particularly when the 1:09.3 at Pimlico is actually faster all things considered.

That's an interesting comparison, a slower track vs faster track, and Churchill's layout makes for those wickedly fast but unreliable opening quarters.... but also because class comes into play as well and generally speaking, horses that race in Kentucky are classier than the ones racing in Maryland. We see this all the time with Parx shippers to New York, most of the time, the figures don't hold up.

Again, this is a problem with figures. I'm not trying to say that figures aren't useful, they are. But, figures earned over the same track are much more reliable for comparison purposes than figures that shippers have, especially if the shipper is coming from a lower grade track.

There are so many conundrums with speed figures. Another situation that arises every day at most tracks, horses dropping in class often improve their figures, and vice versa. That's why figures taken at face value lead most bettors to the poor house. It's the handicapper who is an expert at interpreting speed figures that can actually make good use of them.

highnote

03-29-2016, 11:48 PM

One person I know used the following process to make a daily variant:

He would start by normalizing all times and distances of all the races on the card to one mile.

Then he would find the differences in time per mile of all the races and denote the fastest race of the day.

Then he would project the variant for the entire card from the fastest race of the day using the top finishers in the race (sort of the way cj does it) -- assuming he felt there was no reason to believe the track surface speed changed throughout the day.

If he felt the track surface speed changed then he would make a variant for any race that he felt changed.

The logic behind using the fastest race on the card is that he felt it was probably the most formful race.. However, this method was not set in stone. He could use more than one race or use other races if he felt the fastest race was not formful. So there was an element of art in his figure-making.

steveb

03-29-2016, 11:49 PM

I am not even sure what you are saying above. Sure, I got the obvious points - give everything away, blah-blah-blah.

The point is that you trivialized making any kind of figures and it is just not a trivial thing.

You basically said that you could build pars in a matter of minutes. I do not believe you could do that. Very simple.

Sure, there are some tracks where the planets line up and the pars just fall together. And the better my tools get the more often that happens. But there are also tracks where it is like the planets are from different solar systems.

The idea that building pars by averaging a few columns of numbers is just a dream.

Regards,
Dave Schwartz

PS: I actually started serious coding on an Apple II back in 1978. Then I did consulting on the CBM-32 (Commodore Business Machines). That would be the one that pre-dated the Commodore 64 by about 3 years.

sorry, i just had another read.
i know you have a business, and have to appear cleverer than the next bloke, so i suppose it's natural you can't have some blow in doing easily what you can't yourself do
and then make some stupid assumptions.
being the polite person i am, i never commented on the worth of your stuff, because i simply don't know its worth.
all i said was it won't be the same and mine is quicker.
but now that you have hopped in and made assumptions that you would not have a clue about, then....
now it's my turn to make an assumption(although i know it's true!!).......toss it in dave, where time is concerned you are out of your depth.

highnote

03-29-2016, 11:51 PM

So a horse that gets a 90 figure at Churchill with a 1:09 will beat a horse that gets a 90 at Pimlico that gets a 90 in 1:09.3? That is pretty archaic thinking, particularly when the 1:09.3 at Pimlico is actually faster all things considered.

All else being equal except final times, if both horses raced at CD and both earned a 90, but one earned a 90 by running 1:09 and the other by running 1:09.3, if my life depended on it, I would be inclined to bet the 1:09 horse to win before I would bet on the occurrence of a dead heat or of the 1:09.3 horse winning.

Dave Schwartz

03-29-2016, 11:58 PM

sorry, i just had another read.
i know you have a business, and have to appear cleverer than the next bloke, so i suppose it's natural you can't have some blow in doing easily what you can't yourself do
and then make some stupid assumptions.
being the polite person i am, i never commented on the worth of your stuff, because i simply don't know its worth.
all i said was it won't be the same and mine is quicker.
but now that you have hopped in and made assumptions that you would not have a clue about, then....
now it's my turn to make an assumption(although i know it's true!!).......toss it in dave, where time is concerned you are out of your depth.

:lol:

I will just laugh this off and move on.

03-29-2016, 11:59 PM

All else being equal except final times, if both horses raced at CD and both earned a 90, but one earned a 90 by running 1:09 and the other by running 1:09.3, if my life depended on it, I would be inclined to bet the 1:09 horse to win before I would bet on the occurrence of a dead heat or of the 1:09.3 horse winning.

This should be very easy to test in a database.

steveb

03-30-2016, 12:05 AM

:lol:

I will just laugh this off and move on.

i would too, if i was you.
but i am 100% sure of it.

highnote

03-30-2016, 12:26 AM

This should be very easy to test in a database.

You're right. Someone should test this.

It has occurred to me that horses that have faster times, but slower or the same figures, win more than the figures say they should, but it seemed so obvious -- especially in regards to pace figures -- that I never gave it too much thought.

Many pace figure makers just use Quirin's method and add up the differences from par and divide by the number of races and call that the pace variant. Quirin claims that is close enough. And maybe it is. But pace figures sometimes have not made sense to me when I have used pace figures in conjunction with actual times.

For example, Jim "The Hat" Bradshaw described a simple method for determining if a horse was suitable to the pace of today's race. He used raw pace times.

I have made a lot of money using his simple system and whenever I am at the track without figures I use his method.

The conclusion I came to is that pace and speed figures oftentimes contain a lot of noise.

I prefer to use pace figures to find patterns over several races. It's easier to spot a pattern with a series of pace figures than with a series of raw times.

highnote

03-30-2016, 05:33 AM

to do that then you would need a pace factor which then means its not really a speed figure, because in most instances what any individual horse will do depends on the race pace, that is, what the leading horse/s are doing.

speed figures rightfully should have huge variations in my opinion that reflects the different ways races are run, the different makeup of the race, and so on.
i don't and never have believed in projections.

then you could you make additional figures that might give adjustments to the individual runners depending on how they went or were suited, not to the race as a whole

I re-read this post and it "clicked" for me. It is not an epiphany, but the way you stated it gave me a little more insight into the challenge.

The final time of a race is not only dependent on the pace, but also on the fitness of the runners. This is well-known.

Projections tend to "smooth out" the speed figures of past races. I've always felt that projected ratings are more like handicap ratings than speed figures because the figure maker is trying to say what rating he thinks the horse ran rather than what time he actually ran.

This is why Jim "The Hat" Bradshaw's simple system is so effective using only raw sectional and final times. He asks the question "how well can the horse in question run the race against the likely pace scenario?"

First, a handicapper needs to assess each horse's level of fitness. And then second, estimate how fast they are likely to run against today's likely pace given their level of fitness.

Roughly, you have two fitness scenarios:

1. Fit horse

2. Unfit horse

and roughly, three pace scenarios:

1. Slow pace

2. Average pace

3. Fast pace

A fit horse running against a slow pace should run a pretty decent final time.

A fit horse running against an average pace should run an even faster time.

A fit horse running against a fast pace should run in an excellent time.

An unfit horse running against a slow pace should run an average time.

An unfit horse running against an average pace should run an average time.

An unfit horse running against a fast pace should run a subpar time.

A very fit 5 year old $5,000 claiming horse is not going to run a mile and a quarter in 1:59.00 at Churchill Downs no matter how fit he is and no matter what the pace is -- and no matter how fast the track is.

But a very fit horse of Grade 1 or Group 1 caliber is likely to run a mile and a quarter in a very respectable time (assuming it has the genetics to be an effective 10 furlong runner).

Or how about this scenario...

A particular good horse is a deep closer.

It is in a race with an unfit lone-speed horse.

The lone-speed horse jumps out to an early lead and slows down the pace. It wins in an average time, but beat the deep closer who is in excellent shape because the pace did not collapse due to there being only one early speed horse in the race.

Now run that same race with two unfit front-runners, a field of unfit mid-pack runners, and only one fit deep closer. The deep closer should win.

Capper Al

03-30-2016, 07:09 AM

I had written code for Mordin that allowed him to make U.S. figures that were on the same scale as his European figures. It was definitely useful to have figures for Euro shippers racing in the U.S.

The big challenge was keeping up with all the figure making. He kept up with it because it was his job as a journalist to write about international racing, but it was a tedious work -- even with a computerized system.

It is a lot of work.

Capper Al

03-30-2016, 07:21 AM

classhandicapper

03-30-2016, 11:19 AM

This should be very easy to test in a database.

I have no idea what you will find, but I think it's more complex than just looking at the final times/speed figures and then testing whether the faster raw time is a good tiebreaker when the speed figures are equal. I just don't want to get too involved in this discussion because it's very complex, hard to explain, hard to measure, is dependent on how the speed figure maker makes his figures, and there are ways to back into the right answer for the wrong reason. If you have the right answer for the wrong reason in many cases, the point is moot. I may pursue this privately someday, but right now I only have anecdotal evidence and a pretty good theory.

ultracapper

03-30-2016, 11:42 AM

I wouldn't use raw times as a predictive measure at any time. Nor would I compare a 1:09 with a 1:09.3 as isolated data and think I would glean any insight. But comparing 2 horses with 5 6f races each, both receiving Beyer's regularly in the low 70s, I would lean toward that horse that had shown the ability to run faster raw times than the other. For argument sake, I would call it a very deep tie breaker. But just as Pandy had stated, I have noticed that the horse of the 2 in this example that was able, in the past, to get to the finish line faster than the other seems to have an advantage.

Capper Al

03-30-2016, 12:50 PM

This should be very easy to test in a database

Class,

I don't think this is reference to my previous post. But, just in case, my point is a Prime or Power number is not a measure of speed. I believe the factors like speed should remain separate until finally combining them into a Prime or Power number. My Prime number hits more winners than any of my single factors including speed. But I don't say that my Prime number is my speed number.

03-30-2016, 01:01 PM

I have no idea what you will find, but I think it's more complex than just looking at the final times/speed figures and then testing whether the faster raw time is a good tiebreaker when the speed figures are equal. I just don't want to get too involved in this discussion because it's very complex, hard to explain, hard to measure, is dependent on how the speed figure maker makes his figures, and there are ways to back into the right answer for the wrong reason. If you have the right answer for the wrong reason in many cases, the point is moot. I may pursue this privately someday, but right now I only have anecdotal evidence and a pretty good theory.

I don't know the results either, but the query could be constructed well enough that you would have a big enough sample size to be confident in the results.

My first query would be to test similar speed figures at the same distance, say group by every five points at 6f, and then check the win percentage and ROI broken down by final time. So for example, find all horses that ran a 90-95. Then break them down by final time. If there is any merit to this theory, the horses in the faster final time range will have better ROI and win percentage.

My personal opinion is that won't be the case. I think it will be pretty even over the range of times.

ultracapper

03-30-2016, 01:29 PM

I don't know the results either, but the query could be constructed well enough that you would have a big enough sample size to be confident in the results.

My first query would be to test similar speed figures at the same distance, say group by every five points at 6f, and then check the win percentage and ROI broken down by final time. So for example, find all horses that ran a 90-95. Then break them down by final time. If there is any merit to this theory, the horses in the faster final time range will have better ROI and win percentage.

My personal opinion is that won't be the case. I think it will be pretty even over the range of times.

That would be a very interesting study. My gut tells me the results would lean towards the "faster" horses due to my observations, but it wouldn't shock me if it leveled out simply because I have never really made any study of it at all. It's more theoretical with a good dose of "damn, I should have bet that one" experiences that have lead me to my conclusions.

At present, however, I am comfortable with my viewpoint.

classhandicapper

03-30-2016, 02:27 PM

I don't know the results either, but the query could be constructed well enough that you would have a big enough sample size to be confident in the results.

My first query would be to test similar speed figures at the same distance, say group by every five points at 6f, and then check the win percentage and ROI broken down by final time. So for example, find all horses that ran a 90-95. Then break them down by final time. If there is any merit to this theory, the horses in the faster final time range will have better ROI and win percentage.

My personal opinion is that won't be the case. I think it will be pretty even over the range of times.

Intuitively that's the correct way to test it, but I'm not sure it's controlled enough because different figures makers make their variants in different ways. Some people could be capturing any adjustments that need to be made via the track variant because they are more or less prone to breaking races out etc.. It's worth a test anyway if it's not a lot of work on your part.

03-30-2016, 02:36 PM

Intuitively that's the correct way to test it, but I'm not sure it's controlled enough because different figures makers make their variants in different ways. Some people could be capturing any adjustments that need to be made via the track variant because they are more or less prone to breaking races out etc.. It's worth a test anyway if it's not a lot of work on your part.

I didn't say I was going to do it :)

Obviously you should test it on the figures you use. Trying to apply results to another set of numbers is a mistake.

classhandicapper

03-30-2016, 02:58 PM

I didn't say I was going to do it :)

Obviously you should test it on the figures you use. Trying to apply results to another set of numbers is a mistake.

:lol:

I'm going to wait until I retire and have a lot more time for this kind of thing. I had to use a spreadsheet to keep track of what's already on my personal project list. ;)

whodoyoulike

03-30-2016, 06:38 PM

... i don't and never have believed in projections. ...

I can't imagine handicapping a race without making projections for each horse. Maybe our interpretation of projections is just different (semantics).

The horses over here run different distances and over different surfaces which makes predicting a horse's performance in a particular race IMO impossible unless you're able to project their performances.

steveb

03-30-2016, 07:13 PM

I can't imagine handicapping a race without making projections for each horse. Maybe our interpretation of projections is just different (semantics).

The horses over here run different distances and over different surfaces which makes predicting a horse's performance in a particular race IMO impossible unless you're able to project their performances.

i think you are correct, as far as semantics go.

for every race i add to the data, i have an expectation of how fast it should run, i guess that is a projection of sorts.

what i don't do after the meeting is run, is alter the speed of a race to suit what i THINK it should have been run in.

when the code is figuring the speeds of all the race on the card(on same surface-different surfaces are considered different meeting for all intents and purposes), then it will give varying weights to how much each race contributes to the 'track speed' or what you would call variant, taking the pace into account(i have optimum ways that every race should be paced, when it varies i take it into account)
but i will NEVER alter the speed of any particular race to suit.
the only times i will alter races, is if it is weather related, or occasionally the track deteriorates through wear and tear as the meeting proceeds.rarely would i do that though.
other times my computer might split the variant if there has been a severe weather event, although the computer won't actually know there has been that event, it will run tests and it seem to work just fine when it does split them.

what i would not agree with, is your last sentence, because i don't think it is peculiar to your country.
singapore races 2 surfaces same meeting often times and has 2 different turf courses
shatin race dirt and turf same meeting often.
japan race a mix of dirt and turf at their jra tracks too.
i don't have problems with them, although i guess what i would not see as a problem does not mean somebody else wouldn't....what i mean by that, is seeing the need to changes figures to fit a preconceived idea.

Capper Al

03-30-2016, 07:19 PM

:lol:

I'm going to wait until I retire and have a lot more time for this kind of thing. I had to use a spreadsheet to keep track of what's already on my personal project list. ;)

That's my retirement plan also. Have fun!

InsideTheRaces.com

03-31-2016, 03:16 AM

You can restate what you just wrote in a much more pleasant manner.

After all, we're not talking politics or religion or even race fixing here. Why the need to get all uppity?
Here's why, this guy always jumps into threads and from my perspective he's got no clue about racing in the good old USA.

steveb

03-31-2016, 03:46 AM

Here's why, this guy always jumps into threads and from my perspective he's got no clue about racing in the good old USA.

thanks for the chuckle boss.
i apologise, i did not know you were a moderator and i needed your permission to speak.
i will remember to run it by you before i post next time.

i was just talking about that to my wife today with regards to my grandson,... most three olds are bossy.

InsideTheRaces.com

03-31-2016, 03:56 AM

thanks for the chuckle boss.
i apologise, i did not know you were a moderator and i needed your permission to speak.
i will remember to run it by you before i post next time.

i was just talking about that to my wife today with regards to my grandson,... most three olds are bossy.

I'm not going to argue with you. Believe what you want to believe. It's still a free country here and you have a right to your opinion. I'm not a moderator I'm a nobody. Carry on!

steveb

03-31-2016, 05:12 AM

I'm not going to argue with you. Believe what you want to believe. It's still a free country here and you have a right to your opinion. I'm not a moderator I'm a nobody. Carry on!

no, you have got me going now.
i see you were one of the chosen nine, so you can't be a nobody now.
you must be rolling in it, after tm took you on,and turned you into a winner no doubt.
i must say i am sorry, and you were right all the time, i know zilch compared to what you must have picked up there.
are you gonna spill the beans, oh chosen one of nine? :)

jasperson

03-31-2016, 05:27 PM

Going back to the OP, the daily variant is the problem. Making pars for distances/surfaces is an easy concept, although it might be a lot of work to implement.

Some mentioned speed figures that drifted into what actually is a power or prime number by adding other handicapping factors. For me, the only acceptable addition to speed is age and form/cycle since these are projecting time. Any other manipulations make the figures a prime or power number and not speed.
If it's not giving away something you want to keep private. I would be interested in how you apply form to speed ratings. I think form is sort of nebulous factor just like class. I use scott's idea on form it is either in form or out form but I don't use it in any calculations.

highnote

03-31-2016, 06:37 PM

If it's not giving away something you want to keep private. I would be interested in how you apply form to speed ratings. I think form is sort of nebulous factor just like class. I use scott's idea on form it is either in form or out form but I don't use it in any calculations.

"Blinkers Off -- New Frontiers in Form Cycle Analysis" by Cary Fotias is the best book on using figures to evaluate form, in my opinion.

green80

03-31-2016, 06:50 PM

Quote:
perhaps it is just that YOU can't get any value out of it.
that does not mean others can't.

That I can completely agree with!

This would apply to handicapping software also.

Capper Al

03-31-2016, 08:29 PM

If it's not giving away something you want to keep private. I would be interested in how you apply form to speed ratings. I think form is sort of nebulous factor just like class. I use scott's idea on form it is either in form or out form but I don't use it in any calculations.

I'm on the road for the next 10 days, so I'm away from my books. But one book is Speed to Spare, and there is a couple of systems published by a Ken L. on form/cycle and using Beyer numbers to determine form cycle. Like I said earlier, I'm not their yet in my coding.

04-04-2016, 02:55 PM

The theory some have mentioned, as simple as I can figure, is that when speed figures are equal, the better final time should be preferred. I would also take this to mean that the "slower" the track variant, the less reliable speed figures become. Just want to see if I'm on the same page before finishing up some testing.

Hasn't been addressed either that I can tell, but what about super fast tracks that have a lot of points deducted from speed figures? Are these also supposedly less reliable, or is it just the slower times with points added?

Capper Al

04-04-2016, 03:15 PM

It must be nice to have a large database.

classhandicapper

04-04-2016, 04:51 PM

04-04-2016, 04:54 PM

I think there might be issues in both directions, but I'm not sure we can prove it using existing figures because people break out races, split days, split routes and sprints etc....

I think that would be a very small portion of the numbers so over a large enough sample it should have little to no effect. In my case I would say less the 3% of races (at least on dirt) are "broken out".

CincyHorseplayer

04-04-2016, 05:34 PM

steveb

04-04-2016, 07:34 PM

Quote:
perhaps it is just that YOU can't get any value out of it.
that does not mean others can't.

That I can completely agree with!

This would apply to handicapping software also.

i just noticed this.
may i please ask what you are inferring with your last sentence?
i already know i can get value out of my own software, and i have never used other people's, so forgive me, if i have no idea what you are talking about?

jasperson

04-04-2016, 08:07 PM

I hope you are responding through the "lens" of a nonlinear curve and non-homogeneous data.
I didn't understand a word you said.:D

highnote

04-04-2016, 08:19 PM

I didn't understand a word you said.:D

Is there such a thing as a linear curve?

classhandicapper

04-04-2016, 08:20 PM

I don't have any concrete evidence to back this up but I believe if what Pandy says is at least a somewhat common occurrence it has less to do with the actual times and more to do with the days that the surface is really slow. It has been weather compromised when drying out or wind effects. I remember talking with CJ about this on a Sunday at Tampa with some strange times a few years ago. Anyway I think those days the surface is just different than any fast days for whatever reason and those circumstances won't be replicated next time out. Nothing to do with time.

I pretty much already know that something else is amiss.

I have two years of data that alerted me to "a problem" I had never contemplated before. That's why I keep saying I think Pandy is on to something. I just can't prove it. It's not going to be easy to prove or refute with existing figures and raw times either because it's not all slow or fast tracks. I'd have to build a new model for making figures and then make them the old way and new way to do a comparison. That's a major ordeal because I don't even have a specific method for the new figures. I have a theoretical framework.

At this point I am being more of a pain the ass than helpful with my cryptic posts. I'm turning into Cratos. :lol: I'd just like to figure this out for myself. I think it's enough for me to say that I would not dismiss what Pandy is saying out of hand just because it makes no sense based on conventional wisdom. If it made sense everyone would have figured out what's going on.

Cratos

04-04-2016, 10:36 PM

I didn't understand a word you said.:D
What I am saying is that if the relationship between two variables seen as a straight line it is thought to be a linear curve.

For instance, in horseracing if distance is on the X-axis and is the predictor; and change by change by 1unit with speed on the Y-axis as the response variable change by 1 unit there exists a linear curve.

I can't think of a nature system where this occurs, but in mechanical products this is done by design.

jasperson

04-05-2016, 07:27 AM

this is all i have read above.
but why would you think i have never done it?
i started writing code for this on an old commodore 64 two lifetimes ago.

The old commodre 64 brings back memories. I started doing my speed ratings on my timex sinclare back in the 80's. I still have both of them along with the thermal printer.:):):)

jasperson

04-05-2016, 07:48 AM

Thanks for explaining

MitchS

04-11-2016, 04:36 PM

Everyone knows the formulas being used:

Parallel Speed Chart adjustment
Track variant
Daily variant

But consider how the speed figures are adjusted when it comes to the daily variant. If speed makers are trying to peg a daily variant the first thing they say is something like this: These $20,000 Claimers should be running this distance at 1:05(example). They ran a 1:10 so they are 5 over. Then they subtract 5 for the daily variant. (It's not exactly this simple, but overall this is what they do.) So the daily variant would be adjusted with a minus 5 to equalize. Here's the catch, with all this equalizing they are negating the actual speed and going by what the class of horses should of ran. So is it speed or is it class? The speed has been factored out of it.

I don't mean to be sarcastic by any means with this statement... ( And this relates to a lot of posts in this thread) but... everyone seems to forget or bring to light about the inherent properties with randomness in general. We're dealing with a herd animal that is flesh and blood. There is always, without exception an element of randomness involved. Perceived value vs Realistic value, IMHO

MitchS

04-11-2016, 05:04 PM

Was getting ready for our game night, and realised I had left something out.

Perceived value vs Realistic value vs Entertainment value :D

Capper Al

04-12-2016, 10:02 AM

Was getting ready for our game night, and realised I had left something out.

Perceived value vs Realistic value vs Entertainment value :D

I don't disagree that other factors enter into figuring a race, some known and some unknown. What this thread is about is how to measure one of these factors, speed.

pandy

04-12-2016, 10:42 AM

Pertaining to my opinion that you have to be careful not to discount real time, there were some good examples of this at Aqueduct Saturday. In the 9th, Timeform had the 1 horse Sallisaw with a last race fig of 109 for his 6 furlongs in 1:12.4 over the inner track. They had Unified at 108 based on his 1:08.4 win at Gulfstream Park, which was a tick off the track record. Now Bris had Sallisaw at 90 and Unified at 100. Obviously, my Diamond System had Unified as the clear cut fastest horse and he won pretty easily. Sallisaw never got close.

Another thing I stated, and Classhandicapper backed me up on both of these opinions, all speed figure services get numbers wrong. I noticed that some of you like to beat up on the Beyers, well, the Beyers are very good. Timeform's numbers are very good. But, they all have numbers that are just not right because it's not an exact science.

But regardless of that, real times still matter.

Timeform did have it right in the Wood. Bris had Matt King Coal as the top fig, but Timeform had Outwork at 111 and Matt King Coal at 110. Once again, my Diamond System had the fastest horse on top, Outwork. If you look at the pps, you can see that Outwork's 1:43 for 8.5 furlongs at Tampa Bay was the fastest real time that any of those horses had run. The only horse that was close was Dalmore but he ran 1:43.1 at Santa Anita, which is a faster track than Tampa Bay.

Mind Your Biscuits won the 7th, clearly the fastest time in his Saratoga race last year, and Mind Your Biscuits was a Triple fig on Time Form's figures, which really made this horse the best bet of the day by a wide margin and he destroyed the field at 3-1 odds.

04-12-2016, 12:08 PM

The only horse that was close was Dalmore but he ran 1:43.1 at Santa Anita, which is a faster track than Tampa Bay.

But isn't that the whole point of a variant, to tell you which was a faster track? How can you use that as an example yet compare Sallisaw's raw time to that of Unified. One was at Aqueduct, one at Gulfstream. Clearly one track was much faster that day.

Another note, it is possible I didn't get Sallisaw's number right, but I would never just assume that because one horse didn't run well right back on a muddy track. Three of the other five horses from that race have come back with two wins and a second in their next starts.

classhandicapper

04-12-2016, 01:51 PM

I have 3 opinions related to this.

1. Everyone gets an occasional figure wrong.

2. Even after the fact it's often tough to tell what the correct figure was because of form changes and other factors.

3. I have some evidence that raw times may matter at the margin, but I don't have the data needed to prove it. Even if I am right, it doesn't come into play very often.

pandy

04-12-2016, 02:42 PM

On another thread someone asked about track records, do they matter? Sure, sometimes you get a stupid track like when Santa Anita went back to dirt, or a wet track that's sealed and crazy fast, or Turf Paradise, but generally speaking, most horses that set track records are very fast horses. At Belmont 7.5 furlong, Commentator 1:27.2, mile Najran 1:32.1 (world record, I think), Secretariat owns two track records there, Man O War has one at Belmont, and so does In Excess.

MitchS

04-14-2016, 08:17 PM

I don't disagree that other factors enter into figuring a race, some known and some unknown. What this thread is about is how to measure one of these factors, speed.

I understand Al,

Was rushed and trying to be somewhat comical to a very deep subject that can be looked at many ways. The internet doesn't do that justice sometimes. I've created speed figs many ways over the years and quite a bit of research on the subject. Three quick things I've noticed come to light. 1. Irrespective of what track a horse has raced, fast horses win races. Raw times not converted can pay huge dividends. 2. Horses always (or almost always) cycle back to previous highs (Good races). Three, horses like certain race tracks (courses) compared to others. Some of the figs I do relate to some of these things and its worked out to be a better mouse trap for me at least.

Capper Al

04-15-2016, 10:49 AM

I understand Al,

Was rushed and trying to be somewhat comical to a very deep subject that can be looked at many ways. The internet doesn't do that justice sometimes. I've created speed figs many ways over the years and quite a bit of research on the subject. Three quick things I've noticed come to light. 1. Irrespective of what track a horse has raced, fast horses win races. Raw times not converted can pay huge dividends. 2. Horses always (or almost always) cycle back to previous highs (Good races). Three, horses like certain race tracks (courses) compared to others. Some of the figs I do relate to some of these things and its worked out to be a better mouse trap for me at least.

NP. Speed is an endless topic.

pondman

04-18-2016, 08:32 PM

What I am saying is that if the relationship between two variables seen as a straight line it is thought to be a linear curve.

For instance, in horseracing if distance is on the X-axis and is the predictor; and change by change by 1unit with speed on the Y-axis as the response variable change by 1 unit there exists a linear curve.

I can't think of a nature system where this occurs, but in mechanical products this is done by design.

There are just better ways to store data, than to use 500 year old methods.

Cratos

04-19-2016, 01:30 AM

There are just better ways to store data, than to use 500 year old methods.
You obviously do not understand the post because it had nothing to do with storage of data,, but everything to do with the explanation of the relationship between variables.

In general, many people see the world in linear relationships; nonlinear relationships which define the horse's race curve is difficult to understand by many.

Also please be reminded that the laws of motion is over 250 years old and is very valid and productive in today's calculation of objects in motion.

castaway01

04-19-2016, 10:10 AM

You obviously do not understand the post because it had nothing to do with storage of data,, but everything to do with the explanation of the relationship between variables.

In general, many people see the world in linear relationships; nonlinear relationships which define the horse's race curve is difficult to understand by many.

Also please be reminded that the laws of motion is over 250 years old and is very valid and productive in today's calculation of objects in motion.

And virtually every person who made a lot of money wagering on horse racing all the way back to Pittsburgh Phil knew absolutely nothing about the laws of motion. You're using a cruise missile to blow up a gnat.

pondman

04-19-2016, 01:52 PM

You obviously do not understand the post because it had nothing to do with storage of data,, but everything to do with the explanation of the relationship between variables.

In general, many people see the world in linear relationships; nonlinear relationships which define the horse's race curve is difficult to understand by many.

Also please be reminded that the laws of motion is over 250 years old and is very valid and productive in today's calculation of objects in motion.

A 6th grader can get the speed rating results of every race run in the US for several years. They will be able, with 6th grade set theory, see there are a few hot spots in the US, that consistently produces a margin of profit. They will also see many more areas to avoid.

But in response to your statements-- there are reason for a high rate of astronaut fatalities. Man is not as smart as he thinks he is.

Cratos

04-19-2016, 02:03 PM

And virtually every person who made a lot of money wagering on horse racing all the way back to Pittsburgh Phil knew absolutely nothing about the laws of motion. You're using a cruise missile to blow up a gnat.
And you have proof of that assertion?

Tom

04-19-2016, 02:14 PM

Reality?

castaway01

04-19-2016, 09:28 PM

And you have proof of that assertion?

Logic? Common sense? The idea that none of these guys were physics professors but all knew how to wager on horses and win?

Capper Al

04-20-2016, 10:17 AM

What I got out of this thread is that some speed figures are actually prime numbers and not speed figures, by my definition of what a speed fig should be. Also, I was amazed that Beyer speed didn't adjust for weight in the past.

Mulerider

04-20-2016, 11:42 AM

Another thing I stated, and Classhandicapper backed me up on both of these opinions, all speed figure services get numbers wrong. I noticed that some of you like to beat up on the Beyers, well, the Beyers are very good. Timeform's numbers are very good. But, they all have numbers that are just not right because it's not an exact science.

I don't have near the handicapping experience most of you have. I've only used Brisnet figures since I started, and have no experience with Beyers yet. I'm looking forward to using Timeform for the first time this weekend.

There seemed to be a pretty striking difference between Bris and Beyer concerning the horse Dazzling Gem in the Arkansas Derby. According to a friend who uses Beyer exclusively, Dazzling Gem's figs for the three races prior to the Ark. Derby were 74, 83, and 83. Brisnet's numbers for the same races were 87, 89, and 98. The third race was the Louisiana Derby; Beyer shows no improvement between the second and third races, Brisnet's indicates pretty solid progress. CJ, if you see this and it's not too much trouble to check, did you show a significant difference in the two races?

Using Bris numbers I had DG ranked fourth, which is where he finished, unfortunately, because at 18/1 I took a stab. But my system also had Cupid ranked first, and he finished tenth, so there you go. :)

Mule

raybo

04-20-2016, 11:46 AM

I don't have near the handicapping experience most of you have. I've only used Brisnet figures since I started, and have no experience with Beyers yet. I'm looking forward to using Timeform for the first time this weekend.

There seemed to be a pretty striking difference between Bris and Beyer concerning the horse Dazzling Gem in the Arkansas Derby. According to a friend who uses Beyer exclusively, Dazzling Gem's figs for the three races prior to the Ark. Derby were 74, 83, and 83. Brisnet's numbers for the same races were 87, 89, and 98. The third race was the Louisiana Derby; Beyer shows no improvement between the second and third races, Brisnet's indicates pretty solid progress. CJ, if you see this and it's not too much trouble to check, did you show a significant difference in the two races?

Using Bris numbers I had DG ranked fourth, which is where he finished, unfortunately, because at 18/1 I took a stab. But my system also had Cupid ranked first, and he finished tenth, so there you go. :)

Mule

That's about where Dazzling Gem ranked in my program, and lots of programs had Cupid ranked #1, so there you go! It's horse racing - LOL!!

04-20-2016, 01:11 PM

I don't have near the handicapping experience most of you have. I've only used Brisnet figures since I started, and have no experience with Beyers yet. I'm looking forward to using Timeform for the first time this weekend.

There seemed to be a pretty striking difference between Bris and Beyer concerning the horse Dazzling Gem in the Arkansas Derby. According to a friend who uses Beyer exclusively, Dazzling Gem's figs for the three races prior to the Ark. Derby were 74, 83, and 83. Brisnet's numbers for the same races were 87, 89, and 98. The third race was the Louisiana Derby; Beyer shows no improvement between the second and third races, Brisnet's indicates pretty solid progress. CJ, if you see this and it's not too much trouble to check, did you show a significant difference in the two races?

Using Bris numbers I had DG ranked fourth, which is where he finished, unfortunately, because at 18/1 I took a stab. But my system also had Cupid ranked first, and he finished tenth, so there you go. :)

Mule

This is what I had for the three races before the Arkansas Derby. (He ran 112 on Saturday)

pondman

04-20-2016, 01:39 PM

I don't have near the handicapping experience most of you have. I've only used Brisnet figures since I started, and have no experience with Beyers yet. I'm looking forward to using Timeform for the first time this weekend.

Mule

Spend a year looking at the numbers. You'll find there are a few good spots. And you'll find spots that you should avoid. You've got to take additional steps and know the when and the where of the game.

It's less important how they are made, as long as they are consistent. If you run 3 or 4 sets of numbers, you'll find some do well on a specific track, surface, and at a distance. While others do better elsewhere.

And then you'll run across an angle or two, that completely confounds the ratings. That's when you'll make really good money.