EP0395773A1

EP0395773A1 - Electronic speed rating calculator and method

Info

Publication number: EP0395773A1
Application number: EP89107924A
Authority: EP
Inventors: Robert S. Sinn
Original assignee: ESRAC COMPUTER CORP
Current assignee: ESRAC COMPUTER CORP
Priority date: 1989-05-02
Filing date: 1989-05-02
Publication date: 1990-11-07

Abstract

Apparatus and method are set forth for calculating a comparative speed rating for an entrant in a race such as a horse race so that the entrants' performance can be compared with the performance of other entrants. The speed rating is determined in accordance with a formula which is based on a relationship between speed and distance over a particular distance for a particular class of entrant taking into consideration the track and its conditions; the class of track, the race type allowances and the age of the horse and time of year.

Description

BACKGROUND OF THE INVENTION

The present invention relates to methods and systems for calculating speed rating for paticipants in a speed contest or race. More particularly, the present invention relates to methods, systems and formula for determining a speed value or speed rating for participants, such as race horses, in a speed contest, such as a horse race, for example.
There are many methods for calculating a speed rating for race horses. A speed rating of a race horse is a comparative rating of the speed of the particular race, taking into account the distance, in furlongs, of the race.
An animal's ability to race is a function of this innate ability modified by its physical condition at the time of the race. That is, an animal in peak physical condition can run over a given distance at a certain average speed, taking into account that the particular animal is, at the time, running the distance at its ultimate capacity and no further amount of training can improve its performance. The innate ability to race for the individuals of any species has a bell shaped curve, the same as all other physical and performance characterstics, such as height or intelligence. The individuals of a species tend toward a normal, that is, the peak of the bell shaped curve, with exceptional individuals of the species out at the tails on either end of the bell.
For many years, horse racing enthusiasts have been seeking a method to evaluate the innate running capacity of horses, at various distances and at different race tracks, different conditions, different races, and takes horses of different sex and age. This is difficult for several reasons:

a) Horses race at many different distances at different tracks and there is great difficulty in relating the performance in one race at one track at one distance to another performance in another race at another track at another distance.
b) Although past performance of all horses in a race are published in a publication called the Daily Racing Form, this does not take into the class of track of other factors.
c) The same horse will run at different speeds on different tracks. This difference is a function of the track structure, ie., the length of the track and the track condition. There are both long term variations. Long term variations are a function of the track structure, while short term variations are a function of weather, ground conditions, amount of scraping, etc.

The racing class of a horse is related to his gposition on the bell shaped performance curve. A horse of high class, a horse whose position on the curve is out on the high side of the bell, will beat a horse of average class, a horse whose position on the curve is at the peak of the bell at any typical racing distance. Higher class horses appear to have more ability and tend to perform better than lower class horses at all racing distance. Higher class horses appear to have more ability and tend to perform better than lower class horses at all racing distances.
The problem is how to rate a horse's ability such that:

a) Horses of the same class average the same rating at all distances.
b) Horses of different class have different ratings on an ascendant scale with performance.
c) Horses taken individually on the average have the same speed rating at all distances.

Some of the above problems have found solution in the teachings of U.S. Patent No. 4.133.031 and a pending United States Application Serial No. 789,123 whereas the speed of a race horse is calculated by a method that employs a relationship between speed and distance for horses of the same class, a relationship between horses of different classes and a relationship between speed and distance for a particular horse, and where taking into consideration those relationships a specific equation is employed to determine the winner. The equations taught are as follows:

For 9 furlongs or less:
For more than 9 furlongs:
- Where f is the length of the race in furlongs.
- Where t is the time of the winning horse, in seconds.

It has been found that this equation is substantially accurate and effective for deriving the speed rating (SR). However, this equation lacks certain factors, type of race, price, speed, sex, age and class of track. The present system seeks to take these into consideration by adding track variants according to the formula:

Where TV is the track variant.
Where T is time.
Where D is date.
Where C is category.
Where N is number of races in that category.
Where i is the running index of races in that category.
Where CRR is connected race rating.
Where AL is algorithm.
Where CD is connections used on type of race.
Where CRR = SR (Winner) + Sex Corrective + SBC
Where
CD = OCL 1 x Price
BCL .8 x Price
MCL .5 x Price
SAL 2 x Price
SHC 2 x Price

SUMMARY OF THE INVENTION

The present invention provides takes into consideration the equations for solving the speed rating (SR) of a horse. The equations are:
and
and adds to them factors for type of race, space, applicable, purse, sex, age and class of track according to the formula
Accordingly, the method of the present invention includes the steps of classifying the entrant relative to other entrants in the same or similar sporting events. The speed rating of the entrant is calculated according to a mamthematical formula which includes the pertinent factors for determining a speed rating for each entrant independent of variations such as track and classification of the entrant. Thus, by the method of the present invention, a speed rating is calculated which is general in its application in that the same speed rating applies for the entrant regardless of the distance of the track on which the race is run.
The apparatus of the present invention includes means for data entry such as keyboard similar to keyboards commonly used by 4-function arithmetic calculators, a random access memory for storing input data and other intermediate and output data, and arithmetic unit for calcuting a speed rating given the input data, a display device for displaying selected data and a read only storage program control for controlling the sequence of steps to be executed in the calculation of a speed rating.
These and other objects, features, and advantages of the present invention, together with the operation of the invention, will be understood by reference to the following detailed description taken together with the following drawings.

DETAILED DESCRIPTION OF THE DRAWINGS

Fig. 1 illustrates the generator of the speed rating system.
Fig. 2 is a graph illustrating the relationship between Race Rating and Corrected Dollars for an idealyzed medium fast track.
Fig. 3 is a chart showing an example of the invention.
Fig. 4 is a graph of Race Rating versus Corrected Dollars.

PREFERRED EMBODIMENT OF THE INVENTION

For many years horse racing enthusiasts have been seeking a method to evaluate a horse's innate running capacity at various distances. This is difficult for several reasons:

1. Horses race at many different distances and there is great difficulty in relating performances at the different distances. For example, if you know the racing ability of horse B at 1 1/2 miles and the racing ability of horse A at 1 1/2 miles and the racing ability of horse A at 1 mile, who would be the fastest at 1 1/4 mile?
2. The same horse will run at different speed on different tracks. This difference is caused by the track structure and the track condition. There are long term variations (track structure) and short term variation (weather) and etc.

A horse's racing class is related to this position on the bell shaped performance curve. A horse of high class (out on the high side tail of the bell) at any typical racing distance. Higher class horses tend to perform better than lower class horses at all racing distances.
The problem is how to rate a horse's racing ability such that:

1. Horses of the same class average the same rating at all distances.
2. Horses of different classes have different ratings on an ascendant scale with performance.
3. Horses taken individually on the average have the same speed rating at all distances.

The present speed rating systems do not meet the criteria as stated above. In most systems such as the Daily Racing Form one point is subtracted from 100 for each fifth of a second the horse's performance was higher than the track record at that distance. This system does not meet the criteria stated above for the following reasons:

1. One fifth of a second at a distance is much less than one fifth of a second at a sprint.
2. The track records at different distances could have been set by horses of different classes. The track record is a function of the horse that set the record.
3. The rule of thumb of one fifth second equals one length is not accurate.

In all the present systems of speed ratings, the focus is on the time of the race. The basic discovery of the linear relationship of speed and distance for horses of the same class now allows for a system which fulfills the original requirements of the speed rating system. The slope of the line relating North American Records was found to be about exactly the same as the slope of the line relating horses of all different classes of horses. This slope is approximately (0,07) length/second furlong below 9 furlongs and approximately (0,0465) length/second above nine furlongs.
Fig. 1 illustrates the generator of the speed rating system. The ordinate is the average speed over the distance. The abcissa is the distance in furlongs. The desire is to make a scale of 0-100 where the vast majority of all performances will fall within that scale. The rating of 100 is chosen as follows:
An optimum speed at nine furlongs is five lengths per second (this is approximately the North American Record. This is assigned a speed rating of 100, the minimum speed at nine furlongs is assigned a speed rating of 0. The 100 speed rating line is then drawn for the ordinate 5 with a slope of (-0.0465). Each 0.08 length/sec. at nine furlongs reduces the speed rating by ten until a speed rating of 0 is achieved. This is set forth in the aforementioned patent application and patent.
In evaluating the racing class of each race, the algorithm takes in consideration:

1. Type of race
2. Price, if applicable
3. Purse
4. Sex
5. Age with monthly correctives
6. Class of track

Obviously, horses are flesh and blood and not machines and therefore, the actual races do not exactly match the predicted "speed rating" fro the mathematical algorithm. However, the average fit is extremely close and individual races fit with a standard deviation of +/-5. The difference between the predicted value from the track algorithm and the speed rating of the winner is the race variant. The race variants are then averaged into four categories - "Dirt Sprint", "Dirt Route", "Turf Sprint" and "Turf Route". The separation into dirt and turf is obvious-they are different tracks of different types. The separation into sprint and route is necessary because the clubhouse turn is a standard part of a route trip, but never a part of a sprint trip (bullrings excepted), and the clubhouse turn can have quite different characteristics from the rest of the track.
These averages in four categories are then the correctives (track variants) which are added to or subtracted from the raw speed nubmer in order to correctly answer the question - How fast did the horse run?
The factors which affect the track variant can be divided into two categories, primary and secondary. Primary factors are relatively constant factors such as soil composition and track configuration. Secondary factors are changeable from day to day and even on occasion form race to race. These are factors such as moisture content, cushion depth, tire tracks, wind, etc. It is difficult for the computer to differentiate between statistical clusters and real physical conditions which could cause the track variant to vary race to race.
In fact, tracks typically do not vary much race to race on the same day, but do vary considerably day to day. There are obvious conditions, however, when the track variant does have a significant trend race to race during the day such as cause by rain and snow storms etc.
The track variant which is used to modify the basic formula is as follows:
This formula is then used to generate the graph of Fig. 2 which has Race Rating or speed rating of horses - race rating or the ordinate and corrected dollars on the absicca.
We then take into consideration the track class as follows (all tracks are split in First Class, Second Class and Third Class - The abbreviations are the standard abbreviations for the names of the tracks - AQU is Aqueduct in New York.

1. CORRECTED DOLLARS

1. TRACK CLASS

Then we take into consideration the race types as follows:

2. RACE TYPES

and the allowances:
and the universal correctives including sex, age, time of years, and stall in which the horse was bred as follows: 3. UNIVERSAL CORRECTIVES
All races fall into four categories Sprint - less than 8 furlongs or a route 8 furlongs or more and either dirt or turf.
Also the track is either dirt or turf.
To see how fast a horse ran on a certain day a Hewlett-Packard computer is used with the program set out hereafter to compute the speed rating of the horse in that race on that day. That is the first step, then one corrects that for the track variant. In using the track variant one subtracts the track variant so if the horse ran a certain Speed Rating on a dirt sprint we would subtract the track variant to obtain the corrected speed rating. The program in D-base is as follows:

Now the corrected speed rating of a horse at a certain track at a certain race number at a certain date and a certain race number.

SPEED RATING =

CSR (H,T,D,#) = SR (H,T,D,#) - TV (T,D,C)
H = Horse
T = Track
D = Date
# = Race #
C = Category of Race
Dirt Sprint
Dirt Course
Turf Sprint
Turf Course
These are the Speed Rating Formulas for less than 9 furlongs and greater than 9 furlongs.
minus the track variant.
N = # Races in that category
i = Running index race # in that category
The corrected race ratings are :
Algorithm number is as stated in the graph Fig. 4.
CRR = SR (Winner) + Sex Corrective + State Bred Corrective
CD =
OCL 1 x Price
BCL 0,8 x Price
MCL 0,5 x Price
SAL 2 x Price
SHC 2 x Price
Example:
Race 1 State Bred Filly
SR = 70
Filly + 4 (State Bred) + 6 Corrected Race Rating = ₇0 + 4 + 6 =80
N = # Races in that category
i = Running index race # in that category

Fig. 3 is a typical chart illustrating the calculations of this invention. We refer to this chart to illustrate a typical application of the track variant. This particular race is at Churchill Downs on May 20, 1987. The first entry is 8 furlongs. Type is OCL (open claiming). The second entry is listed as maiden claiming (MCL). A1 C means an allowance race. If it is a claiming race, it means that the horse can be claimed for a preset price. If it is not a claiming price - there is no price listed (see entries 7 and 8). Every race also has a purse (in thousands of dollars). The claiming price and purse are not the same. In the first case the purse is $7,400, whereas claiming price is $7,500. In the second race the claiming price is $25,000 and the purse is $9,500. The next column lists the sex of the horse; age 2, 3 or older. The next column lists the number of horses in the race - A1 after the age means it is older than the year listed for ex., the first horse is older than 3. The next two columns list Pace Time and the Race Time.

PACE TIME

If the race is less than a mile, the pace is the first half mile. If the race is a mile or over, pace is the first 3/4 mile. Pace is only two distances - half mile or 3/4 mile. Race time is the time of the winner. The next column lists the condition of the track. F means dirt and T means turf-grass. The second letter F or S is fast or slow. It really means whether the track is wet or dry. In an extremely slow track we will see fast. It does whatever it says about the condition of the track has nothing to do with how fast or slow the track is. It has only to do with how wet the track is. So the track may be the slowest track in the world, but it is called fast. It is dry and slow if it is wet.
The next column is corrected dollars. This is an attempt to evaluate the class of all equivalent races at the claiming price. In other words if the race is a claiming race, an open claiming race such as the first listing, the price as the corrected dollars - $7,500 purse will be 7,500 corrected dollars. Maiden claiming races are half the price, because the class of a maiden claiming race is equivalent to half of an open claiming race. So for the second entry the price is $25,000 and the corrected dollars are $12,500. The algorithm includes the aforementioned formulas for generating corrective dollars from all the races. If they are claiming races, open claiming corresponds on a one to one basis, maiden claiming on a 50% basis and then beaten claiming is 80%.
In the allowance races, such as 7 and 8, the starter allowance is multiplied by 2. So if it is 10,000, the starter allowance will be 20,000.00. In the allowance races there is no price. So for those, we use the table alone. All the tracks are put into the first, second and third category (class category).
A2 is a non-winner of two races avenue. A3 is non-winner of three races. A1 C is a non-winner of a race other than a maiden or claiming race. A2C is a non-winner of a race other than a maiden or claiming race. A2C is a non-winner of two such races, A3C three such races and so on. This whole allowance classification is non-winners of a dollar amount for a certain time.

PACE RATING (PR):

That is the speed rating formula applied to the pace time on the pace system. We take the speed rating formula and take out F. The pace is always 1/2 mile or 3/4 mile.

SPRINT or ROUTE:

Route is defined as a race of 8 furlongs or over. Thus, 8 furlongs or greater is going to be 6 furlongs at the pace distance. The pace time is always based upon the first 3/4 mile if it is a route race and it is always the first 1/2 mile if it is a sprint race. A sprint is less than eight furlongs and a route is eight furlongs or greater. So a mile or over is a route. With either a 1/2 mile which is 4 furlongs or 3/4 mile, which is six furlongs, we then use the speed rating formula to get the pace rating formula.

To obtain the Pace Variant we take the Pace Rating then we add universal correctives. As stated above universal correctives are sex, age, time of year and where it is bred. For example, if it is a female we add 4, if it is a male, we do not add anything. If the horse is 2 or 3 years or old, you add the amount stated on the table. Also it depends which month we are in, for example. In June for a 2 year old we add 8, 3 year old we add 2. Now, referring to the table - the third entry, it is a 2 year old and the race is in May. So for a 2 year old in May according to the universal correctives we add 8, since the sex of the horse is male. The amount added would be zero. Then computing the PR from the formula we get 51. We then add 8 to get 59. Now we refer to the graph Fig. 4. We look up at the corrected dollars-8.750 and the Race Rating is 67. 67 is what the graph says, 59 is what the race was actually run at. The difference is -8 which is listed as PV. So it is 59 - 67 = -8. This gives us the difference between the projected speed and actual speed. It is the difference between this algorithm from the graph and pace rating with universal correctives.
The graph is derived by looking at hundreds and hundreds of races, and horses at all different prices. What we are doing is comparing this horse with the average horse on a medium fast track for all the male horses. In other words if we have the price and we can tell how fast the race should be run pursuant to this formula. In fact the race in the above example was run slower shows that the track is slow. You could take a great horse, such as Secretariat, and he would run slower on a slow track.
AVERAGE PACE VARIANCE (APV):

To obtain this we take the pace variant in the four categories, viz., route, sprint, dirt, turf. We take an average for each category and place the average in for each race it corresponds to. The break point used for the formulas is different from the definition of sprint or route. So, we use the formula and we compute the number. Some races run fast in the beginning, and some races do not. So this shows you whether they are running fast in the beginning or running fast at the end. We take 57 then add the universal correctives and get -8. If it is an older male, there is no universal correctives. There are no age correctives because it is an older horse. 3-1 means it is older. There are correctives after 3-1. Now let us take the 2 year old (No. 3) we get 43 using the formula. It is a male so there is not any sex correctives, but there is an age corrective. This is May; May for a 2 year old is 8. So we add eight to make 51. The difference is -16. Now, we average them for the 2, 3, 4, 5, 6 and 7 races because they are all the same type of races and the average is -6. Then we put into those races that correspond to it. Now we go to CRR.

Corrected Race Rating (CRR):

We take the race rating and subtract the average race variant. So for the first race 57 - (-4) = 61. This is the normalized rating of the race. Because now we are using the average. So this is the best indication of how fast comparatively the race is really run. Now we go to the next column. This is the corrected race variant. In other words, the race variant -8, but the average race variant was -4 which means that this race was run 4 points slower than what it should have been run. And this race 7 is 10 points faster and race 3 is 10 points slower. So this tells us how fast each race is run versus what it should have been run. This is the normalized condition of the track. This tells you how fast or slow each race was run versus the norm for the day. There is no condition of the track used in this. In other words, this race was 10 points slower after being corrected for the average race variant. This race is still 10 points slower than what it should have been. There are many reasons for it; it could have been not enough competition. It could have been a great horse, but no competition.

While the present invention has been described with reference to preferred embodiments thereof, it is understood by those skilled in the art that various changes in form and application of the electronic speed rating calculator and method may be made without departing from the spirit or scope of the invention.

Claims

1. An automatic speed rating method for determining the speed rating of an entrant in a speed contest over a predetermined distance comprising, in combination:

determining the world class record for the speed of horses at particular distances

determining the length of the race in furlongs (f)

inputting the above into the relationship

modifying such relationships by the track variants for determining the speed rating of the entrant.

2. The automatic speed rating method according to claim 1 wherein the entrants are horses and the race is a horse race over substantially nine furlongs.

3. The automatic speed rating method according to claim 1 wherein for races of 9 furlongs or less A = .077 and B = 4.92 and for races of more than 9 furlongs A = .0465 and B = 4.686.

4. The automatic speed rating method according to claim 3 wherein the track variant is determined in accordance with the formula: