File Specification for the STK2002 comma delimited data files.
Each record is terminated with new line characters hex 0d0a (LineFeed).
The first record in the file is the minimum cost of one exacta bet,
usually 2 meaning two dollars. This is followed by one record for
each horse in the following format:
2 File header is minimum cost of one exacta bet (note no comma)
(1) "Horse name", (Text string) Array Element
(2) Class, Avg speed par of past races ( 0 )
(3) Raw speed, (Numbers are integers) ( 1 )
(4) Early pace, ( 2 )
(5) Mid race pace, ( 3 )
(6) late pace, ( 4 )
(7) Jockey's point score, ( 5 )
(8) Trainer's point score, ( 6 )
(9) Dollars per start / 100, ( 7 )
(10) Runline score, ( 8 )
(11) Data supplier's speed, ( 9 )
(12) DRF style speed, ( 10 )
(13) Adjusted speed, ( 11 )
(14) Off odds when results are present ( 12 )
(15) Runline used when handicapped ( 13 )
(16) Handicappers addition to Runline score ( 14 )
(17) Race distance in yards, ( 15 )
(18) Winners time in seconds, ( 16 )
(19) This horse's lengths back at finish, ( 17 )
(20) Odds calculated at runtime, ( 18 )
(21) Point score / 1000 calculated at runtime ( 19 )
Total point score is calculated at runtime by multiplying
each item by its assigned importance value and adding all
the results of the multiplication. Since it is calculated
at runtime, its value in the data file itself has no impact
and is not significant.
There is one control record per race. It is marked by "##" in the
horse names field and is always the last record in the race. The
format of this record is as follows: (All data except the string
marker are integer numbers in text format.)
(1) "##", a text string marker for the record. Array Elements
(2) Race number, ( 0 )
(3) Number of records in the race, ( 1 )
(4) Post position of the winning horse, ( 2 )
(5) Win payoff of winner * 10, ( 3 )
(6) Place payoff of winner * 10, ( 4 )
(7) Show payoff of winner * 10, ( 5 )
(8) Post position of the second horse, ( 6 )
(9) Place payoff of the second horse * 10, ( 7 )
(10) Show payoff of the second horse * 10, ( 8 )
(11) Post position of the third horse, ( 9 )
(12) Show payoff of the third horse * 10, ( 10 )
(13) Exacta payoff * 10, ( 11 )
(13) Race Distance in yards, ( 12 )
(15) Trifecta payoff /10 , ( 13 )
(16) Surface conditions of this race, ( 14 )
(0) = Fast
(1) = Wet Fast
(2) = Good
(3) = Sloppy
(4) = Muddy
(5) = Ice
(10) = Firm
(11) = Yielding
(17) Race Class of This race, ( 15 )
(1) = Maiden Claiming
(2) = Maiden Special Weight
(3) = Claiming
(4) = Starter Handicap
(5) = Starter Allowance
(6) = Allowance
(7) = Nongraded Stake Handicap
(8) = Grade III Stake Handicap
(9) = Grace II Stake Handicap
(10)= Grade I Stake Handicap
(18) Exotic payoff / 10, ( 16 )
(19) Exotic payoff / 10, ( 17 )
(20) Exotic payoff / 10, ( 18 )
(21) Exotic payoff / 10, ( 19 )
______________________________________________________________
A template for manual data entry may be created easily from
a text file of horse names the following way.
Edit the horse names file with any text editor. After the last
horse name in each race, add the following four characters:
"##"
Create a blank data row starting with a comma followed by
20 zeros all followed by a comma, for example,
,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
Use edit, cut, to get the zeros in the edit paste box. Then place
the cursor at the end of each horse's name and edit, paste the
zeros in. Paste the zeros after the control character also.
Replace the first zero in the control record with the race number.
Replace the second zero in the control record with the number
of records in the race. (counting all the horses and the
control record) Start the file with the cost of one exacta bet.
Example:
2
"RUFFIEN",0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
"TURNBOW BULL",0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
"JACKPOLL JONNY",0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
"A HORSE CRAZY",0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
"HORSEY JOHN",0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
"##",1,6,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
"CIGAR",0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
"LOOK LIKE ME",0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
"MY LAST DIME",0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
"JO BULL JOHN",0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
"FIVE HORSE",0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
"SIX HORSE",0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
"##",2,7,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
Save the file with a .doc extension, example: myrace.doc
This creates a race day file that may be loaded with stk99.
Each of the data elements can be changed in stk99 by placing
the cursor over the element and pushing the tab key, or the
insert key, or the number 0 key.
A useful function for this kind of template is to generate
a consensus rating for handicappers. Just use one of the
variable columns for each handicapper. For each handicapper's
top pick, place a 5 in that horse's row at that handicapper's
column. Put a 3 for the second top pick, and a 2 for third
etc.
After collecting a few days of racing data run the stk99
optimization function. The variable weight assigned to each
handicapper will change in accord with that handicapper's
success rate. The A, B, and C picks in future races will reflect
the importance assigned to the handicappers. You might reserve
a column for your own picks and see how you compare with the
great guru's of the art.