JCapper Program Features
What is JCapper?
JCapper is a database driven Handicapping Software package. Its simple interface makes true database handicapping possible for the non-programmer.
JCapper's database tool set enables the thoroughbred horseplayer to create a database using the player's own Bris or TSN data and results
files and then run queries against that database to determine the precise factors that are working (or not working) at specific tracks, distances,
surfaces, and race types. JCapper's modeling tool set empowers the player to create User Defined Models so that JCapper can do what it was designed
to do best- quickly point the player to those horses each day that fit the exact User Defined Model Definitions determined by the player's own research.
JCapper is a handicapping software package that will:
Build a database using your own Bris or TSN data and results files as the source.
Let the non-programming horseplayer perform his or her own research by running queries against that database.
Let the non-programming horseplayer create and edit User Defined Models as spot plays.
Let the non-programming horseplayer download, unzip, and load past performance data files for today's races.
Let the non-programming horseplayer download scratches and races taken off the turf right from the internet.
Analyze today's loaded past performance files at the click of a button. JCapper will very quickly point out those horses that qualify as potential spot
plays based on the User Defined Model Definitions created by the individual player.
User Defined Models or UDMs
JCapper's key strength is enabling the user to set up his or her own unique wagering models. User Defind Models or UDMs allow the horseplayer to handle any handicapping situation imaginable. Users can create UDMs for specific tracks, distances, surfaces, and classes. UDMs can be created to exploit the patterns of specific trainers, riders, and owners. Users can set definitions for rank, numeric value, and gap for any of the individual factors supported by JCapper. UDM definitions can be created using each of the factors in combination with each other.
A well thought out UDM is a thing of beauty and a true reflection of each player's preferred individual handicapping style and can be as as generic or as specific as the player desires.
I have developed two Out Of the Box UDMs that ship with JCapper. Both were proven
to be profitable based on a $2.00 win bet during the benchmark testing period. For more
information on JCapper Benchmark Testing, click
Are the Out Of The Box UDMs as strong as they could be? No. Should users be satisfied
with them? Again, no. While profitable, they should really be viewed as starting points.
They are inclucded for two reasons: First, to allow new users to hit the ground running
and make winning selections from the first time JCapper is started. And second, to provide
an example to users of how UDMs might be constructed- so that they can see the types of
things that go into them before setting off to construct UDMs of their own.
Users who get the most out of JCapper are the ones who undertake the journey of performing their own research and creating their own UDMs to exploit what they discover.
The amount of power and flexibility included in UDMs is enormous. Nothing else quite like it exists anywhere in any other handicapping software on the market - at any price.
The Data Window
Another of JCapper's key strengths is the Data Window. This is the one piece of software that actually lets the user see the data. Using the Data Window, the user can very quickly see the win rate, average mutuel, and return on investment broken out by any of the factors supported by JCapper.
Q. What separates the JCapper Data Window from the data tools included with other handicapping programs?
A. First, the user can bring up the Profile Table Interface and see information broken out by any and all combinations of the handicapping factors
supported by JCapper. This truly breaks new ground. This kind of power and flexibility has never been offered in any other handicapping program- ever.
Second, users can use the Data Window in concert with the Profile Table Interface to create and modify UDMs on the fly. As the user performs research and
makes new discoveries, he or she can capture the essence of his or her work and apply it to future races by saving what was discovered as part of a UDM
definition. Again, this type of power and functionality has never been offered in a handicapping program before.
Third, other handicapping database tools return one dimensional or single query results with each separate query. Running queries this way can quickly eat up of a lot of your time.
The JCapper Data Window returns multidimensional results with each single query. This feature is a great timesaver.
Let's say you want to test out a factor- let's call it factor A. Ultimately you want to see results for factor A broken out by rank and various odds ranges. With the data tools of other programs you have to run a single query for each rank and odds range. Let's say you want ranks 1 through 12 and you want to see 12 different odds ranges. You'd have to run 144 (12 times 12) separate queries just to get the information you wanted.
With the JCapper Data Window, you get all the information you want to see by running a single query. What comes back is a section broken out by rank for Factor A for ranks 1 through 20 and a second section with an odds breakout showing 20 separate odds ranges for Factor A.
Again, the kind of power and flexibility you'll find here has never been offered in a handicapping program before- ever.
A Unique Set Of Numbers
The past few years have seen an enormous proliferation of online players who download past performances and use handicapping software as part of their selection process. The game has changed. Unless you have numbers that are unique you stand very little real chance of winning in the long run.
JCapper offers a powerful array of insightful numbers and handicapping factors. Many of these factors are unique and can be found nowhere else.
A true edge can be obtained over the crowd using these numbers. It will take some work on your part. JCapper provides the tools. The rest is up to you.
But it can be done.
For a glossary of all the factors supported within JCapper click
The Profile Marker
The Profile Marker compares the past performance record of every horse against the profile table definitions for each active UDM. Whenever a horse matches a UDM definition it becomes marked as a potential play. Marked horses will appear highlighted on all HTML Reports for easy identification. In addition, all marked horses are written to a special text file which can be viewed with the JCapper Report Viewer or with Notepad if you like. This provides the user with an organized list of all potential UDM plays for each day.
Black Box Capability?
Is this true Black Box capability? If not, it's very close. The Profile Marker does all the work. It identifies all potential UDM plays using the exact criteria the user has stored in the Profile Table.
A true Black Box would make selections without the user ever knowing why. In JCapper, users create their own User Defined Models and thus gain complete control over the program's selection process. So JCapper is not a true black box in that sense. But once the user has set up one or more of their own winning UDMs the Profile Marker makes the rest completely automatic. Simply put: This is the closest thing to a commercially available true Black Box that I know of.
Scratches and Changes
Without a doubt, one of the most annoying things that horseplayers deal with on a daily basis are scratches and late changes.
JCapper provides the user with three separate modules to handle this problem quickly and smoothly.
Scratch Bot is a module that presents the user with two links for each loaded race card file:
A live tote link. When this link is clicked, the odds for the current race at
the selected track are parsed. Any horse with non-numeric odds is processed as
A scratches link. When this link is clicked, one or more web pages for
the selected track are parsed. Horses named as scratches on the web page being parsed are
processed as scratches. If races taken off the turf information is listed on the web page being parsed, the surface change is also
processed by Scratch Bot.
Scratch Bot automatically processes each scratch or surface change and then notifies
the user that a change has taken place. If you play from home or anywhere with a
live internet connection, you can run Scratch Bot throughout the day to get all
your scratches. To the best of my knowledge, no other program (at any price) offers
anything like this.
Scratch Parser is a module that lets the user import and process scratches directly from
the Bris website at the click of a button. You need a Bris account for this service, but Bris
currently offers scratches to their members at no charge. This is a great feature for those who
want to get a head start on their handicapping before heading out for the track or otb where
there is no live internet connection or for those players who just want to make their picks and
then go do something else for the day.
Manual Changes Module is for anyone playing from a location where there is no live
internet connection. The JCapper Manual Changes Module provides the player an intuitive quick
and easy interface for entering scratches and surface/distance changes. Horses can be scratched
and unscratched with a single mouse click. Scratched horses are clearly shown in strkeThru red.
Horses not scratched are clearly shown in blue.
Based on user feedback, people simply love the simple organized way that JCapper handles scratches.
Profile Table Interface
JCapper ships with an Access database. The Profile Table in the JCapper database is where UDM definitions are stored. Every supported factor in JCapper has a separate field in the Profile Table for activation/deactivation, min and max numerical value, min and max rank (where applicable) and min and max gap (again, where applicable.) The Profile Table currently has over 250 fields. To facilitate UDM creation and maintenance, JCapper provides its users with several different interfaces to the Profile Table.
UDM Wizard. This module makes the process of creating and maintaining UDMs a breeze. Users can create UDMs or choose to edit existing UDMs. UDMs can be activated and deactivated. After bringing up a UDM, users can select a factor and then set min and max values for numeric value, rank, and gap. The interface is intuitive and very similar to a form based web page.
Complete View All Profiles. This view gives the user access to all fields for all UDMs stored in the Profile Table. The interface uses a Microsoft Data Grid Control which very closely mimics opening up the table in Microsoft Access.
Complete View Active Profiles. This view gives the user access to all fields, but only for the active UDMs stored in the Profile Table. The interface uses a Microsoft Data Grid Control and very closely mimics opening up the table in Microsoft Access.
Speed and Pace View. This view gives the user access to fields that relate to the speed and pace ability of a horse. Only active UDMs stored in the Profile Table are accesible with this view. The interface uses a Microsoft Data Grid Control and very closely mimics opening up the table in Microsoft Access.
Connections View. This view gives the user access to fields that relate to the connections of a horse. Users will find this view helpful in constructing trainer specific and rider specific UDMs. Only active UDMs stored in the Profile Table are accesible with this view. The interface uses a Microsoft Data Grid Control and very closely mimics opening up the table in Microsoft Access.
Filter View. This view gives the user access to fields that relate to preset and dynamic filters. Users will find this view very helpful when filtering horses out of a UDM. Only active UDMs stored in the Profile Table are accesible with this view. The interface uses a Microsoft Data Grid Control and very closely mimics opening up the table in Microsoft Access.
Misc View. This view gives the user access to fields that do not relate to speed and pace, connections, or filters. Users will find this view very helpful when defining UDM parameters for factors like class rating, form rating, recent activity, rail position, and a horse's age. Only active UDMs stored in the Profile Table are accesible with this view. The interface uses a Microsoft Data Grid Control and very closely mimics opening up the table in Microsoft Access.
Racetype View. This view gives the user access to fields that relate to track, class of race, surface, and distance. Users will find this view very helpful during intitial UDM creation. Only active UDMs stored in the Profile Table are accesible with this view. The interface uses a Microsoft Data Grid Control and very closely mimics opening up the table in Microsoft Access.
The algorithms within JCapper were designed and tested using the $1.00 DRF Single Format Data Files and 25 cent Exotic Results Data Files available
from Bris at http://www.brisnet.com. Bris file users are not locked into a data subscription plan and
may download as many or as few data and results files as they see fit.
Many users have also reported oustanding results using TSN Procaps Advantage data files available by monthly subscription from http://www.tsnhorse.com/
Ease of Operation
JCapper is very easy to use. Below is a quick start guide showing how easy it is to get started. All you need to do is just follow these easy steps.
Daily Quick Start/Operating Instructions
What follows below is a Daily Quick Start guide to using JCapper each day to Load Race Cards, Process Scratches, Calculate Races,
and View Reports. The Daily Quick Start presented here is intended as an overview of the processes a player would go through
each day in using JCapper to arrive at potential UDM plays.
Further detailed information and instructions for the use of each module can be found in the appropriate section of
the User's Guide.
Load Race Cards. Click the Load Race Cards button. You'll need to first copy and unzip one or more DRF Single Format Data files onto a folder of your choosing. Then, just use the JCapper Race Card Loader to load the desired file(s) into JCapper. This is an easy process. You can navigate through your system's folder and file structure just like in Windows Explorer. There's even a button that automatically searches, finds, and loads all data files in your default folder for a given race date.
Scratches and Changes. Use Scratch Bot, Scratch Parser, or the Manual Changes Module to quickly and easily process any scratches or changes.
Calculate Races. Click the Calculate Races button and JCapper goes to work. It performs basic number crunching, making thousands of calculations, for every horse, in all loaded data files. A comma delimited export file is prepared. Some of our users like to import this into a spreadsheet like Excel for their own purposes. The Profile Marker is launched. The past performance record of every horse in all loaded data files is compared against active UDM definitions in the Profile Table. Horses that fit the definitions for any active UDM become marked so that they appear highlighted on all HTML Reports. Marked horses are also written to a separate text file that can be viewed with the JCapper report viewer or opened in Notepad if you like. The HTML Report Builder is launched. An HTML Report is prepared. This report can be viewed using the JCapper Report Viewer or the user's browser of choice. Numbers are formatted in an easy to follow format. All potential UDM plays are highlighted for easy identification.
All of this happens automatically and it also happens fast. With seven or eight loaded race cards and 25-40 active UDMs, the entire process from start to finish takes no more than 90 seconds on most machines. At the end, the user simply sees a message that says "Races Calculated."
View Reports. Click the Report Viewer button. The JCapper Report Viewer lets the user toggle back and forth between the HTML Report and the Text Report. There's also a print button anytime the user wants to send the Text Report to a printer.
HTML Report. The JCapper HTML File Report shows the significant factors with numbers and rankings for every horse in an easy to follow format. All potential UDM plays are highlighted for easy identification. There are hyperlinks so that the user can freely navigate from track to track. There's also an internal print button anytime the user wants to send the report to a printer.
Text Report. The JCapper Text File Report shows all UDM Plays for the day neatly organized in one place. Here you'll find a header for each race that contains the Track, Race Number, Surface and Distance, and the class and RaceVolatility for the race. Then, any potential UDM plays in the race are laid out for you. Here you'll find post position, name of horse, name of rider, name of trainer, the name of the UDM that the horse qualifies for, morning line odds, and finally, the BettingInstructions from the UDM definition are shown in brackets.
Seeing this report leaves very little doubt as to what horse(s) should be played or at what odds. It's right there in a very easy to follow format.
Building a Database. To get the most out of JCapper, users will at some point want to build a database. Below is a simple 'hands on' discussion of how best to accomplish this process.
Prerequisite: Matching DRF and XRD files in the same folder. Before being able to build a database, JCapper users will first need to have some DRF and XRD files with matching dates sitting in the same folder. The folder used as a data location can be be any folder of the user's choosing. We recommend a simple folder structure as follows: Drive\FileYear. One practice users should avoid is having DRF and XRD files for the same track for different years residing in the same folder. This is because Bris does not have the year as part of their filename structure. So the DRF file for Saratoga on Aug 29, 2004 and Aug 29, 2005 would have identical file names. When the user tries to copy the second file into the same folder Windows will prompt with a message stating that the file already exists- Overwrite Y or N?
File Sharing Between Handicapping Programs. JCapper users who have more than one handicapping program installed on their system can store DRF and XRD files in any folder structure required by another handicapping program. This allows for true file sharing between programs and eliminates the need for users to have multiple copies of DRF and XRD files residing in separate folders.
JCapper Database Builder. This module is launched by clicking the Build Database button. Here, the user will find very explicit instructions for building a database right on the screen. First, the user selects a target folder simply by clicking the Folder Icon and using the folder selection tool to point to the folder where DRF and XRD files are sitting. This folder navigation process is remarkably similar to folder navigation using Windows Explorer. The name of the folder being pointed to is clearly shown right on the Build Database button. After pointing to the desired folder, the user simply clicks the Build Database button and program logic does the rest.
At this point, JCapper goes to work. DRF and XRD files are matched up according to track and date. XRD files not yet processed are imported into the database. An exception report is prepared which shows any XRD files that might be missing. This all happens automatically and it happens fast. Importing 50 XRD files into the database usually takes less than four minutes from start to finish on most systems. At the end, the user sees a simple message that says Database File Built.
Validation Data. To avoid overfitting when constructing UDMs, it is recommended that users Build Databases in separate Data Development and Validation Data folders. As a best practice, we recommend that the Data Development Folder contain approximately three DRF and XRD files for every one DRF and XRD file in the Validation Folder. DRF and XRD files for a given date should exist in one folder or the other- not both. Users who construct track specific UDMs should have a selection of separate DRF and XRD files for different dates for the track being studied in the Data Development Folder as well as in the Validation Data folder.
Sample Size. There are no hard and fast rules for how many races constitute a valid sample. It really depends on the factors being measured inside each individual UDM. In general, UDMs should work with causal or logistical factors instead of factors that make no sense. For example, rail position can sometimes be used as a factor to exploit inside or outside position biases found at certain tracks and certain distances. AVGE1, CPace, and PctE can often be used to exploit speed biases on speed favoring tracks (or closing biases on speed tiring tracks.) These would be examples of logistical factors.
Conversely, it is also entirely possible to base UDMs on factors that make no logistical sense at all. An example of this might be creating a series of UDMs to only identify horses with odd numbered post positions or requiring that an owner's name begin with an "H" character. Naturally, the more true influence a factor included in a UDM has on the outcome of a race, the smaller the sample size the creator of the UDM can get away with. For that reason, some UDMs will have obvious value after just a few days of racing while others may need to be run over much larger sample sizes before showing their true merit (or lack of it.)
After creating both Data Development and Validation Data databases with the Database Builder, UDM development is best done working with the Data Window and Profile Table Interface on data found in the Data Development Folder. Then, before risking real money on horses that fit UDM definitions in actual play, the UDM is further tested on data it hasn't seen yet: data residing in the Validation Data folder. The statistical theory of using Validation Data is a valid one and can be generalized in these terms: UDMs that work extraordinarily well on Data Development data but fall apart when confronted with Validation Data very likely performed well on the Data Development data becasue of overfitting. UDMs that work well on Data Development data and that continue to perform well on fresh Validation Data are far more likely to hold up for the user going forward than UDMs not proven on validation data.
Excluding sets of poor horses from consideration is every bit as important as displaying strong horses for consideration. JCapper users have the ability to filter horses out of consideration. By identifying and eliminating sets of horses that have been proven histiorically to be poor bets, what the player is left with is an improved set of horses from a betting perspective.
Most handicapping programs on the market today (and players) give this very important idea very little consideration. Even very small incremental gains in win rate and roi can have a drastic impact on bankroll size (for the better) as the gains are applied over a period of time. Using the Data Window and Profile Table Interface, JCapper users are able to see the data and apply relevant filters to their UDMs.
JCapper supports two types of Filters:
Preset Filters. Currently, there are 400 separate Preset Filters that ship with JCapper. The Filter Tool in the UDM Wizard makes
adding Preset Filters to and removing Preset Filters from a UDM Definition a breeze. For a complete list of Preset Filters click
Dynamic Filters. JCapper users also have the ability to create their own Dynamic Filters.
Dynamic Filters offer the user the ability to define the factors, context, behavior, and numeric cutoffs of each Filter as the
FilterCode itself is written. This gives the user the power and flexibility to design and write Dynamic Filter Codes for specific
situations indicated by the player's own Data Window Research.
The Dynamic Filter Tool in the UDM Wizard makes creating and editing Dynamic Filters and adding them to and removing them from a UDM Definition
a snap. For a complete tutorial on creating and applying Dynamic Filters as part of the UDM Definition click
The JCapper Help Documents Module contains a handy index to all JCapper Help Documents. Each of the modules in JCapper has (or soon will have) a
corresponding help document. The intention is (and I hope that I am succeeding to some degree in this effort) to explain in plain English how to
operate each part of the program, what each part of the program was designed to do, and why. Many of the Help Documents are illustrated with program
screenshots. Some of the Help Documents are in HTML format, and some are in Word Doc format. To view a Help Document, simply click on the
appropriate link in the index. Help Documents that are HTML format will be rendered in your web browser. Help Documents that are Doc format
will be rendered by Microsoft Word if you have that program installed on your machine. If you are not running Microsoft Word, then Help Documents
that are in Doc format will be rendered in Wordpad. As new Help Documents become available they will be posted on this site. To download the most
current JCapper Help Documents click here.
The JCapper File Downloader is a separate module. There are two screens. The first screen acts as a Web Browser and connects directly to the
download pages of the Bris and TSN websites. Now users can download and unzip data and results files without ever leaving
JCapper!!! Downloaded files are automatically extracted right onto your JCapper Default Data Folder. The second screen offers additional functionality
for directory browsing, file unzipping, and loading of race card files. The user interface is intuitive with instructions for use provided right on
screen. The JCapper File Downloader truly makes the process of downloading, unzipping, and loading of race card files as easy and painless as possible.
Copyright © 2005 JCapper Software