JCapper Message Board

JCapper 101
-- track weight

Home	Register Log In

track weight

rkskipster
4/22/2008
3:27:33 PM

Jeff I was wondering if you are going to be able to post you posts on yhe track weight report from the old board? Or is there post on the new board that I havn't seen ? thanks in advance rick

jeff
4/22/2008
6:52:44 PM

I do plan on recovering the posts from the old message board and setting them up in an archive where you guys can at least search and view those posts. But that's at least several weeks away right now.

JCapper2008 does have a Track Weight Report available in the Data Window. And work is currently underway to add track weight (and Pace-Fit) for a horse's prior races to what you see on race day for today's race.

Individual players may or may not agree with the "label" the report assigns for track weight. And individual players may or may not agree with my own thoughts about the importance played by Track Weight in determining race aoutcomes. But if you're looking to get a handle on how individual surfaces and distances at specific tracks have been playing, the numbers presented by the Track Weight Report make a very good starting point.

-jp

.

jager fury
4/29/2008
4:46:45 PM

How do you print a track weight report?

Thanks,

jeff
4/29/2008
6:00:16 PM

The prerequisite is that you must have rebuilt a database from scratch after installing JCapper2008. After that, just select Track Weight Report from the factors drop down in the Data Window. Be sure to set factor constraints for Track, Surface, and Distance before hitting the ALL button. You can also see a Track Weight Report for any of your UDMs.

The report itself looks something like this:

code:

     Data Window Settings:
     999 Divisor  Odds Cap: None
     Dirt (All*)  SPRINTS  
     From Index File: C:\2008\Q2_2008\pL_profile.txt
     Track: KEE 

     Data Summary         Win     Place      Show
     Mutuel Totals     998.60   1079.60    970.50
     Bet             -1164.00  -1164.00  -1164.00
     Gain             -165.40    -84.40   -193.50

     Wins                  68       132       198
     Plays                582       582       582
     PCT                .1168     .2268     .3402

     ROI               0.8579    0.9275    0.8338
     Avg Mut            14.69      8.18      4.90

code:

                                Actual         Expected     Actual/Expected       Win
     Track Weight Analysis     Winners          Winners         Ratio             Roi
     CPace 123                   25.00            34.47        0.7253          1.3464 (CPace 1)
     CompoundLate 123            26.00            28.79        0.9030          0.3575 (CompoundLate 1)



                                Actual         Expected     Actual/Expected
     Run Style Analysis        Winners          Winners         Ratio
     E                           15.00            13.38        1.1207
     EP                          18.00            20.39        0.8827
     P                           11.00            13.50        0.8147
     S                            8.00             7.64        1.0470
     NA                          15.00            11.08        1.3538



     Approximate Track Weight:    5.00  (Speed Tiring)

~Edited by: jeff on: 4/29/2008 at: 6:00:16 PM~

Hdcper
4/30/2008
2:11:51 AM

Jeff,

I am interested in what the track weight report is really telling us, in relationship to an individual udm. Below, I have attached the output of the track weight report for all tracks against one of my Udms. I realize I could run it for individual tracks too, but wanted to see what the output would be for all tracks as a whole.

This specific Udm basically pinpoints horses that should get an uncontested lead (a lone front runner) and it seems strange to me that the output would reflect a somewhat speed tiring track (4.71).






     
     Data Window Settings:
     Divisor = 999      Odds Cap: None
     Filters Applied: 

     Surface: (ALL*)  Distance: (All*) (From Index File: C:\JcapperDatabases\AADATABASE2008\pL_profile.txt)


     Data Summary         Win     Place      Show
     Mutuel Totals     822.80    721.10    643.40
     Bet              -694.00   -694.00   -694.00
     Gain              128.80     27.10    -50.60

     Wins                  93       158       194
     Plays                347       347       347
     PCT                .2680     .4553     .5591

     ROI               1.1856    1.0390    0.9271
     Avg Mut             8.85      4.56      3.32


                                Actual         Expected     Actual/Expected       Win
     Track Weight Analysis     Winners          Winners         Ratio             Roi
     CPace 123                   93.00           109.02        0.8531          1.1856 (CPace 1)
     CompoundLate 123            13.00            12.23        1.0626          0.0000 (CompoundLate 1)

                                Actual         Expected     Actual/Expected
     Run Style Analysis        Winners          Winners         Ratio
     E                           66.00            81.35        0.8113
     EP                          13.00            16.46        0.7900
     P                            3.00             3.58        0.8372
     S                            0.00             0.57        0.0000
     NA                          11.00             7.05        1.5592

     Approximate Track Weight:    4.71  (Somewhat Speed Tiring)

Thanks,

Bill

jeff
4/30/2008
1:20:09 PM

If there's one thing I could emphasize about track weight reports, it's this:

Whenever you look at a track weight report, it is the NUMBERS and not the "label" that you should be paying attention to. Everyone (at least at first) seems to want to focus on the label part and not the numbers. And that's just backwards.

By "label" I'm referring to: Approximate Track Weight: 4.71 (Somewhat Speed Tiring)

In order to create "labels" to identify degree of track weight, I first created a benchmark using the ALL button across a large data sample. That benchmark track weight report looks like this:

code:

     Data Window Settings:
     999 Divisor  Odds Cap: None
     Surface: (ALL*)  Distance: (All*) 
     From Index File: C:\2008\pl_Complete_History.txt


     Data Summary         Win     Place      Show
     Mutuel Totals  346793.00 345030.10 342391.30
     Bet           -454006.00-454006.00-454006.00
     Gain          -107213.00-108975.90-111614.70

     Wins               27800     55324     81523
     Plays             227003    227003    227003
     PCT                .1225     .2437     .3591

     ROI               0.7639    0.7600    0.7542
     Avg Mut            12.47      6.24      4.20

code:

                                Actual         Expected     Actual/Expected       Win
     Track Weight Analysis     Winners          Winners         Ratio             Roi
     CPace 123                14360.00         14882.61        0.9649          0.9025 (CPace 1)
     CompoundLate 123         14297.00         12833.25        1.1141          0.8442 (CompoundLate 1)

code:

                                Actual         Expected     Actual/Expected
     Run Style Analysis        Winners          Winners         Ratio
     E                         6590.00          7084.95        0.9301
     EP                        8335.00          8136.10        1.0244
     P                         5541.00          5344.68        1.0367
     S                         4363.00          4318.25        1.0104
     NA                        2818.00          2714.16        1.0383

     Approximate Track Weight:    2.28  (Somewhat Speed Friendly)

Understand that the "label" part on each track weight report reflects a comparison of that report to the benchmark - nothing more. Bill, the horses included on your report make the surface(s) they raced on appear more speed tiring than the surface(s) raced on used for setting the benchmark.

The number of actual winners for CPace123 and CompoundLate123 is simply a count of the number of winning horses that happen to rank in the top 3 for either category. The number of expected winners for each category is arrived at by summing JPRMLProb for every top 3 horse within each category. A comparison between the two provides (IMHO) some meaningful information. For example, if the number of winners is less than expected, yet the roi is ++ then it is very likely that the profit is being driven by just a small pct of the overall plays... which often reflects the reality of the game.

Your report has the ratio of actual winners / expected at .8531 for early and 1.0626 for late. Compare that to the benchmark numbers of .9649 and 1.1141. The pct gap between early and late is wider on your report than the benchmark gap. The algorithm creating the label notices that. And that's the only meaning that the label has.

It is the NUMBERS on the report that are important... and evaluating those numbers within the context of factor constraints that are driving what is and isn't included on the report... THAT is really important too.

Actually, the numbers on your report are almost exactly what I would expect to see for the type of UDM you describe. The reason you were able to obtain a + roi on CPace horses is that your UDM most likely seeks out those going to post at good odds... meaning that you are trading win rate for value. Your report reflects the realities of the game, nothing more. The early end of the spectrum almost always offers fewer winners than the late end of the spectrum. But value, when it is there to be found, almost always comes from the early end of the spectrum as opposed to late.

-jp

.

~Edited by: jeff on: 4/30/2008 at: 1:20:09 PM~

JCapper Message Board

JCapper 101 -- track weight

JCapper 101
-- track weight