Links Between Sectional and Final Times

[Native Texan III]

I am interested in any discussion on any link between the shape of sectional times for each quarter of the race to the race final time. How much does one cause the other (not the obvious link of simply adding the sectional times up)?

[Cratos]

I am interested in any discussion on any link between the shape of sectional times for each quarter of the race to the race final time. How much does one cause the other (not the obvious link of simply adding the sectional times up)?
Take a look at Trakus data and calculate the slope of the race curve and the energy distribution curve for each horse.

[raybo]

I am interested in any discussion on any link between the shape of sectional times for each quarter of the race to the race final time. How much does one cause the other (not the obvious link of simply adding the sectional times up)?

While there are obviously correlations between early energy distribution and final time, finding specific concrete relationships between them, IMO, is very difficult, if not impossible. That is probably due to the fact that none of us knows what the "available" energy of horses is, before the race, nor if the horse will run to its potential in that race, and maybe if the trainer has even entered the horse to win or as a prep for a future race, etc.. Historical database research might offer some long term perspective into relationships of this kind, but I'm fairly sure you wouldn't be the first to do that. The research has probably already been done and if it is of long term value, wouldn't we know it by now? I know that %E, %M, etc., are included in several programs, but I don't know what the value of those factors actually is, over time.

[turninforhome10]

I have been experimenting with %PctE or Brohamer Pace energy. Here it is explained http://www.brisnet.com/library/software/allnews/favoriteArticles/Final%20Brohamer%20Series%20Part%203%20Percent%20E arly.pdf

What I have been trying to do, and it is still in the infant stages. I have been trying to find what I call the "breaking point". The best way I can explain it is this. Given a predicted pace scenario, at what point does the horse reach their maximum sustained energy. I think of it like this. If they were cars, which car would be geared to achieve early top speed and which would take time to wind up enough speed to switch gears without loosing torque.
My approach and I admit it is still rough, is to use a predicted pace model broken down by normalized sectional call and then I use the the PctE as early and late and normalize them. What I am looking for is the point in the race where either the speed will back up or if the late pace might be to far back to make a move.
The hardest part of this for me has been knowing how to set the weighing factors for each section. Using historical results produces much different numbers than the other way I have been setting. I have been setting the factors based on Bris Impact values for running style.
So my question to add here is this.
How to score the weighing factors to determine the section of the race where the biggest change of pace should happen i.e. the "breaking point"?

[classhandicapper]

IMO...

1. Each horse has different amount of innate speed, stamina, and overall ability.

2. Race track surfaces vary in terms of how tiring they are from day to day.

3. It's more difficult to account for things like wind and track speed when it comes to measuring fractional times than final times because they vary more around PAR.

4. There are a lot of combinations of pace on the way to the final time (very fast/ very fast, very fast/fast, very fast/average etc....) and all will have slightly different impacts.

5. Jockeys adjust to the track conditions and either get more or less aggressive depending on the nature of the surface

The combination of those 5 things and others will wreak havoc with formulas.

The general idea is that when a horses strays too far away from the set of fractions that is ideal for him (either by going too fast or slow), it will impact his final time negatively.

IMO, given these complexities, what you should be looking for is extreme pace scenarios (using pace figures) and by looking at the way races develop to see if good horses on the pace collapsed or weak horses on the pace ran better than expected. When you find strong evidence of an extreme pace, you'll know those horses are better/worse than they look on paper. Then you can either upgrade/downgrade their performances or look at previous races in the horse's PPs for clarification of their ability.

[Robert Goren]

I am interested in any discussion on any link between the shape of sectional times for each quarter of the race to the race final time. How much does one cause the other (not the obvious link of simply adding the sectional times up)?There have been books, actually a lot of books written on this. There are several webs sites on it too. Only God knows how many threads there have been on it. For the record, while I believe it has some importance, but it is over rated. I think how much the front runner is pushed and the quality of the front runner is far very important. Unfortunately that is not always reflected in the sectional times. Then there is the issue of how accurate these sectional times are.

[raybo]

There have been books, actually a lot of books written on this. There are several webs sites on it too. Only God knows how many threads there have been on it. For the record, while I believe it has some importance, but it is over rated. I think how much the front runner is pushed and the quality of the front runner is far very important. Unfortunately that is not always reflected in the sectional times. Then there is the issue of how accurate these sectional times are.

I agree, I believe one's time is more wisely used by attempting to predict the possible pace scenarios in races and concentrate on those horses who, by benefit of their preferred running style and early speed potential, will be advantaged, or disadvantaged, by those scenarios. And, I also agree that the subject has been argued exhaustively over the years and to my knowledge nobody has become wealthy using that research/method. I suppose some of the folks over at P&C will disagree though.

[sjk]

I think it is one of the most important tools that I use. A key reason why speed ratings in and of themselves are not predictive is because horses are aided or compromised by pace. I think trying to measure this and using it as an adjustment to the speed rating gets you to a much more predictive figure.

[Native Texan III]

Thanks for the replies.

It was more of a generalised question than for specific horses at specific tracks. Some horses in the pack will be disadvantaged by the pace set / course characteristics but they could cause a pace shape that could allow others to hit an optimum spot and achieve an all time best final time. A best time they may never achieve again.

For example is the so called even pace time the truth to make the fastest overall time? Obviously the start and finish cannot be at even pace but how much over the top does an over fast start make to final times (the lead horse and the others in the race) and in a slow early sectionals race horses run closing sectionals faster, but perhaps not as fast as the energy they have left but as fast as they are physically able to run.

All that effects the final time.

[Nosed]

For me it's as still the old adage "Pace makes the Race" ( old time handicappers, Tom Ainslie for one). Figuring out the probable half and 3/4 times depending on the length of the race. Not sure if that's what you're looking for, but it might be of some use.

[Native Texan III]

For me it's as still the old adage "Pace makes the Race" ( old time handicappers, Tom Ainslie for one). Figuring out the probable half and 3/4 times depending on the length of the race. Not sure if that's what you're looking for, but it might be of some use.

Nosed,

Thank you.
Of course, you are right.

Pace makes the race but for which horse. slight variations up or down can favour one or disadvantage it. If a horse is self pace disadvantaged, then others gain not from their own pace but from the faults of others. Dirt will be different to turf.

I suppose I am really asking whether the final time can be better used to understand what has caused the way the cards have fallen in today's race.

If this has been discussed elsewhere I would be pleased to have a few links from PA readers on the topic and that are recommended.

[Tom]

I suppose I am really asking whether the final time can be better used to understand what has caused the way the cards have fallen in today's race.

Yes, the relationship between pace call time and final time can be very important. I will post an example later, from my current research project.

[reckless]

One way to help solve this puzzle, so to speak, is to try to understand the differences and demands of each individual race track that you play or follow.

Knowing the early pace-final time demands of a particular race track could point a player to more contenders than to simply try and determine 'who's the speed?' or 'who's the best closer?'.

There is a correlation to both questions, of course, but matching the horses that fit the energy demands of the track itself could prove more profitable than simply pitting horse against horse.

The link provided earlier by turningforhome should be read over and over and over as %Early is a great way to determine the individual demands of each racetrack. Model 10-20 races per cycle for each distance and you'll be on your way! :)

Could luck.

[Tom]

I wanted to have a tool to use to look at results charts, and to look at pace lines in the PPs that would help explain the pace shape.
I took the pace multiplier idea Randy Giles wrote about and expanded on it a little bit.

The math is simple - using race times, divide the pace time by the final time.
In sprint, use the 4 furlong time, in route, use both and come with two numbers.
I decided to use +/- 1 standard deviation and +/- 1/2 standard deviations to identify average pace, fast and slow pace, and very fast and very slow pace.

Using Belmont, 7 furlongs on the dirt, I came up with a basic multiplier and a multiplier for the various degrees of fast and slow. The variable I use to confirm theses is beaten lengths.

I first used o beaten lengths, 0.1 - 1.0, 1.1 - 2.0, 2.1 - 3.0, 3.1 - 4.0 and over 4.0. After doing a few distances, it made sense to narrow that down to 0-1.0, 1.1 - 4.0, and over 4.0. 0-1.0 are the front runners, 1.1 - 4.0 are the middle runners, and over 4.0 are the closers.

Here is the summary for Belmont 7.0 on the dirt, showing pace time zones and the beaten lengths that occurred in each pace shape as a percentage of all the races in that shape. There were 670 races total, broken down by pace shape -

All.............670
Fast............97
V Fast.........85
Average.....286
Slow...........83
V Slow.......119

[Tom]

Races that go Fast by 4 or more are more favorable to closers than others, so how has this worked out this year so far?

Seven races have been F4 or faster, and of those, 3 went to horse back at least 4 lengths, and one was 3.75 back.

[raybo]

Pretty cool, Tom! Are you using raw times or are you incorporating variants somehow?

Also, in routes, you say to come with 2 numbers, what do you mean by that?

[Tom]

All of this is raw times - I use it immediately to look at the charts.
For routes, I do the same thing for both the 4 furlong and 6 furlong calls.
Eventually, I used this to make a pace variant, but it was mainly for a tool to mark up charts and explain outcomes.

[Nosed]

To Native Texan III
Ok, first I know I'm not nearly as scientific as our esteemed colleagues in my handicapping. I'm simply talking about, in a 6f race let's say, comparing the half and finish times of the favorite (usually based on his highest speed rating at that distance) with the rest of the field. It's not perfect but it usually at least weeds out the "also rans".
Some might include the quarter times, but I don't. Tom Ainslie said years ago "Pace makes the Race".
Horses are athlete's, made of flesh and bone, so I don't know if one can predict exactly how they're going to perform every time even in the best of conditions.

[Exotic1]

All of this is raw times - I use it immediately to look at the charts.
For routes, I do the same thing for both the 4 furlong and 6 furlong calls.
Eventually, I used this to make a pace variant, but it was mainly for a tool to mark up charts and explain outcomes.

I have to study this on a dark day. Not cloudy, just a Monday or Tuesday.

Thanks for posting, looks interesting.

[traynor]

All of this is raw times - I use it immediately to look at the charts.
For routes, I do the same thing for both the 4 furlong and 6 furlong calls.
Eventually, I used this to make a pace variant, but it was mainly for a tool to mark up charts and explain outcomes.

Interesting work. If I were doing it for my own use, my initial inclination would be to factor out races with "impetuous entries with little chance of winning" (using anything from your own ratings to ML or odds to determine the chances of that impetuous entry winning).

To use Castaneda's schema of the shadow of a thing revealing more than the thing itself, the "shadow" would be the reality that a race with an "unlikely to win" impetuous entry is likely to have both a faster-than-usual 2C, and--the more impetuous that entry--a winner that seems to have been further off the pace than normal at the 2C. That winner may not have been a closer (pace definition) at all. It could every bit as easily have been an EP entry whose rider chose not to chase the impetuous entry, knowing full well it would fold in the stretch.