Last week we talked about the origin of the SRF figures which are actually the Delta Racing Figures . We also talked about the handicappers of the past who spent countless hours and went through many trials and errors to develop a system that had never been developed before. They were the true pioneers of the speed figures,. By working on ideas and theories through many nights of discussions they came up with something that had never been developed before (years before the Beyers ever came around) and gave them an edge in handicapping.
It would be another dozen years or so before I would start to comprehend the meaning of the figures. When I was fifteen years old in 1973 I began to do the figures in the Racing Form for my dad and start to ask questions to my dad on what they meant. (By the way my mom did not want me fooling with the horses, maybe she was right J) I wanted to know why we were adding or subtracting points. I wanted to know what the difference was between sprints and routes. I wanted to know what a Plus 4 was and why was that important.
What I came to find out is that once you learn and understand the theory behind it and what the numbers represent you find that you begin to look at races in a different way. Only a few handicappers can understand the language. You begin to see the intricacies of a horse race. You begin to understand what class and speed is all about.
One of the first things I noticed was how horses with the same time would have different speed figures, sometimes the difference was small but other times it was large. More than that what intrigued me most was the time a horse from another track (
HollywoodPark) with a much faster raw time came to the Fair Grounds and face horses with raw times that were much slower. Yet despite this the horse at the Fair Grounds came up with a speed figure faster than the horse from
HollywoodPark. I could not believe that at the time. I figured the
California horse was faster, he has to be. However in the end it would be the Fair Grounds horse that won. I could not believe that, but the speed figure of the Fair Grounds horse said he was faster.Wow, as a young man that intrigued me and set me on the path to want to understand this better and I wanted to know what was behind those numbers and how they worked.
I will try to explain that here, I will try not to get to technical and keep it as simple as possible it may at first seem a little confusing. You will have to do as I did
many years ago and that is open up your mind and think differently. Some of what I say may not make sense at first or you may doubt or question some things and that is just normal. I questioned many things when I first started to learn the system, but that is how you learn. In the years to come this led me to come to a full understanding of the figures. At that point in time once I understood I along with my dad began to refine the system and expand it to route and turf speed figures, develop internal fraction criteria(race within the race) Closing fraction criteria for various distances and develop a pace system based on the track variant. All this took a few years to develop, test and implement to our handicapping needs. This when combined is a handicapping arsenal.
The Variant
The first step in establishing speed figures is to obtain an accurate variant. The process of creating a variant is to establish a Class base. This was begun years ago as we briefly discussed it in last weeks article. The first Class base chart was simple by todays standards. The numbers of classes in the 1960s and 70s was far fewer which represented the nature of racing back then. Through the years the number of race conditions and qualifications has changed. There are more races written today that take into account the number races won with varying restrictions. However even so the most important thing into structuring class is to establish key elements and then from there you can begin to breakdown subsequent classes and give a rating for each class.
The most important factor into establishing a variant is to ensure it is a tight variant. What this means is that based on class the individual race variants at a particular distance should reflect each other. If you are looking at eight 6f races with varying classes all of the race results or individual race variants should be clustered. This is a strong indication that your class chart is accurate. If the numbers tend to be all over the place then you have to examine the reason including the validity of the class chart itself. In essence the individual race variants for a particular distance should fall within a low standarddeviation when computed. This will display any outliers. If a race falls too far outside of the curve it is not used in averaging the variant as it is not a true indicator of the track variant.
Standard Deviation explanation:shows how much variation or dispersion exists from the average (mean, or expected value). A low standard deviation indicates that the data points tend to be very close to the mean, whereas high standard deviation indicates that the data points are spread out over a large range of values. Without getting too complicated in the mathematics this is how it is done ( I know many of you may not quite understand)
Consider if you had eight 6f races with the following eight values:
2, 4, 4, 4, 5, 5, 7, 9 = 4040 / 8 = 5
These eight data points have the mean (average) of 5:
To calculate the standard deviation, first compute the difference of each data point from the mean, and square the result of each.
Next compute the average of these values, and take the square root:
This quantity is the race standard deviation it is equal to the square root of the variance.
In the race data points the 9 is borderline and would have to be further assessed to validity in most cases it would be included . If that number was say a 12 or higher it would be tossed. This is where skill is needed in determination of the track variant.
CLASS CHART:
From what I have studied most class charts begin with establishment of a class figure for the $10-15000 thoroughbred range from there a scale can be determined up and down the class ladder. Years ago it was determined that the Delta figures class rating for $10,000 claimers was a 93. The range back then was the top echelon handicap horses had a rating of 98 while the bottom class $2000 claimers were ranked at an 87. As you can see that is not that great of a range of figures. What you cannot see though is the compounded nature of the figure. In other words the number is earned through the running of each portion or fraction of the race. Individual fractions have a direct impact on the final time, the final time is used then determine to the race variant.
Class charts have a range that varies based on the author. Some scales find the upper GR-1 runner with a 120class down to the lowest level in the 70s.
In the Delta/SRF figures the range is for 4yo&up C&G Gr-1 is a 102 while the $2000 claimer is an 85class. This scale is further broken down based on wins, sex and age, starter Alw, Maidens. For NW3L a point (or length) is deducted, example: a 4yo colt ALW class is a 98, a non winner of 3 races would be a 97. A NW2L races would be a 95 as three points are deducted. A maiden special weight is based from ALW conditions. For maidens five points are deducted from the parent class. Example MSW 3yo is a 93class minus five points making the race a 88 class.
The SRF chart also takes into account maturity progression throughout the year. The 4yo&up class does not change and remains the same throughout the year. However 2 and 3 year olds (colts & fillies) the class rating will change every two months. Example a three year old colt ALW in January will have a 93class rating by the time December comes around he will have matured to a 98 class rating and in most cases can challenge the four year olds or be equal to them. This is true and established throughout the Class scale. In all there are 19 classes established for each classification. There are six major classifications:4yoC&G, 4yo f&m, 3yo C&G, 3yo fillies, 2yo C&G, 2yo fillies. These classes can be further broken down into sub classes. These sub classes include NW2, NW3, Maiden, StarterAlw. In all horses can be broken down into 282 classes.
Finally with all this establish a variant can be created. The way this is done is pretty simple, data from each distance is gathered. A variant is created for various sprint distances, each distance calculates its own daily variant. In routes all routes are combined and average. However depending on the racetrack there are a few exceptions such as one turn mile races. Also races of 10f and up are usually separated from other route races. For the turf they are calculated in the same manner as the main dirt track.
Calculating a Class based Variant
Example of calculating a variant. Below are three 6f races run at GP last Saturday Jan 21st. All 6f races happen to be maiden races on that day.
Race #1 3yo maidens. As we said MSW are based from ALW conditions. In this case 3yo ALW are a 93 class in January. Five (5) points are deducted from that per maiden deduction making race #1 a 88 class race. The 88 is then calculated against the raw time of the race 1:11.13 or an 85
For ease a base number is used in replace of the track record time 1:08 is the 6f base. 1:08 = 100 for every fifth of a second a point or length is deducted. In this case the race was run in 1:11.13 (1:11 flat) which equates to a 85 raw figure. The race was three seconds slower than the 1:08 base which equals 15 lengths/points. You then take the Class minus the raw time. (Note: the track record can be used in replace of the base number however once calculated the variant may be slightly different but the exact same speed figure will be obtained in the end)
88class minus 85 (raw time) = 3 race variant
Race #2 is the same condition as race #1, Races was run in 1:10.89 (1:10 4/5) or 86
88class minus 86 (raw time)= 2 race variant
Race #3 4yo&up Maiden (4yo ALW=98class minus the maiden deduction(5pts) makes the race a 93class) Race time 1:10.10 (1:10 flat) or 90
93 class minus 90 raw time = 3 race variant
You then add up all individual 6f race variants:3 + 2 +3 = 8
The total 8 is divided by 3 (total # races) = 2.666 or a 3variant.
The 3 variant is then added back to the raw time of each race.
Race #1 85 raw time + 3points = 88 SRF speed figure
Race #2 86 raw time + 3points = 89 SRF speed figure
Race #3 90 raw time + 3points = 93 SRF speed figure
1/21/12 GP
Race #1 12:35
6 Furlongs Dirt Maiden Special Weight Purse $40.50 Open 3 Year Olds.
In the examples above each race pretty much calculated back to their class. This not always the case. I used them as examples because of its simplicity.
Next we will look at some Turf route races from TAM on 1/26/12
TAM 1/26/2012 - Race Entries & Results
Race #1 12:25
1 1/16 Mile Turf Allowance Purse $21.20 Open 4 Year Olds And Up.
Track: Firm Scratched: Tizoff 100 Base 1:33 =100 CLASS 9393 minus 92= 1 var
The race variants for the TAM Turf route are:3 + (-)1 + (-)5 + 1=-2
-2 divided by total # of races 4 =-0.5We do not round down unless it is a .6 or greater so the variant for the day would be a 0 (zero)
Race #1 95 raw time + 0points = 95 SRF speed figure
Race #2 86 raw time + 0points = 86 SRF speed figure
Race #3 93 raw time + 0points = 93 SRF speed figure
Race #5 92 raw time + 0points = 92 SRF speed figure
Based on this the variant for the day was a 0 (zero) so no points are added or subtracted to the raw figure. Race #3 came in as a strong race with horses finishing 5 points/lengths above their class. The first two finishers need to be watched especially the 2nd place finisher who will once again face maiden company in his next start.
What you look for is races where horses perform better than their class. A horse or race that is 4 points or more above its class can be considered a Key race. The winner is known as a Plus 4. When this occurs in the claiming ranks a horse has the opportunity to move up in class.
The above procedure is performed as we said for various distances run each day including turf figures. Once this is done each horse is then given a speed figure depending on their individual time (or beaten lengths) each horse in the race receives a speed figure based on the variant.
Some may ask what are the advantages of a class based variant. The answer is simple it is a more accurate variant. In the DRF/HRP variant derived only from raw time against a the track record class is not considered so the true variance could be skewed. As I showed you with class the standard deviation can display outliers. Another advantage is that the figures can be used from track to track so horses from different tracks can be compared easily.
I just wanted to give you a basic lesson today on the Classes used and basic idea of how the variant is calculated. Next week we will look at the figures themselves and how they are used and we will discuss their meaning. We will also look inside races , the differences between sprints, routes, off tracks, turf, etc.
