By |
Jeff.......UDM question |
JustRalph 6/11/2010 6:32:27 PM | jeff
where do I acces the below items for use in a udm ? maybe an example would help ?
I am trying to i.d. horses that hang, fade in the stretch.....thought these would help....
len2call number single beaten lengths at second call last start pos2call number integer position at second call last start lenstrcall number single beaten lengths at stretch call last start posstrcall number integer position at stretch call last start lenfincall number single beaten lengths at finish call last start posfincall number integer position at finish call last start
|
JustRalph 6/11/2010 7:12:55 PM | just in case I haven't told you lately
I like to do a google search for
"jcapper (insert whatever I am looking for)"
such as
"jcapper weighted figure" and I get tons of shit back that helps me out every day. I wanted you to know I appreciate all the stuff you put online........Thanks.......
|
jeff 6/11/2010 11:57:08 PM | --quote:"jeff
where do I acces the below items for use in a udm ? maybe an example would help ?
I am trying to i.d. horses that hang, fade in the stretch.....thought these would help....
len2call number single beaten lengths at second call last start pos2call number integer position at second call last start lenstrcall number single beaten lengths at stretch call last start posstrcall number integer position at stretch call last start lenfincall number single beaten lengths at finish call last start posfincall number integer position at finish call last start" --end quote
Data Sample (using HDW Data) spanning the last 90 days or so showing a quick peek at PScore >= 90 horses in routes, led or within 2L at second call last out... faded from there to be beaten by X lengths by the time the horse staggered past the finish line:
query start: 6/11/2010 8:39:37 PM query end: 6/11/2010 8:39:39 PM elapsed time: 2 seconds ` Data Window Settings: 999 Divisor Odds Cap: None ` SQL: SELECT * FROM STARTERHISTORY WHERE PSCORE >= 90 AND DIST >= 1760 AND (POS2CALL=1 OR (POS2CALL > 1 AND LEN2CALL <= 2)) AND POSFINCALL > 1 ` ` Data Summary Win Place Show Mutuel Totals 2205.90 2098.40 2028.10 Bet -2258.00 -2258.00 -2258.00 Gain -52.10 -159.60 -229.90 ` Wins 226 400 529 Plays 1129 1129 1129 PCT .2002 .3543 .4686 ` ROI 0.9769 0.9293 0.8982 Avg Mut 9.76 5.25 3.83 ` ` By: Finish Call Lengths Last Out ` >=Min < Max Gain Bet Roi Wins Plays Pct Impact -999.00 -10.00 0.00 0.00 0.0000 0 0 .0000 0.0000 -10.00 -9.00 0.00 0.00 0.0000 0 0 .0000 0.0000 -9.00 -8.00 0.00 0.00 0.0000 0 0 .0000 0.0000 -8.00 -7.00 0.00 0.00 0.0000 0 0 .0000 0.0000 -7.00 -6.00 0.00 0.00 0.0000 0 0 .0000 0.0000 -6.00 -5.00 0.00 0.00 0.0000 0 0 .0000 0.0000 -5.00 -4.00 0.00 0.00 0.0000 0 0 .0000 0.0000 -4.00 -3.00 0.00 0.00 0.0000 0 0 .0000 0.0000 -3.00 -2.00 0.00 0.00 0.0000 0 0 .0000 0.0000 -2.00 -1.00 0.00 0.00 0.0000 0 0 .0000 0.0000 -1.00 0.00 0.00 0.00 0.0000 0 0 .0000 0.0000 0.00 1.00 -79.80 188.00 0.5755 20 94 .2128 1.0629 1.00 2.00 -20.80 172.00 0.8791 20 86 .2326 1.1618 2.00 3.00 40.50 176.00 1.2301 24 88 .2727 1.3624 3.00 4.00 -9.90 160.00 0.9381 15 80 .1875 0.9367 4.00 5.00 -0.40 178.00 0.9978 19 89 .2135 1.0665 5.00 6.00 -29.60 144.00 0.7944 12 72 .1667 0.8326 6.00 7.00 -45.90 176.00 0.7392 15 88 .1705 0.8515 7.00 8.00 -47.40 154.00 0.6922 14 77 .1818 0.9083 8.00 999999.00 141.20 910.00 1.1552 87 455 .1912 0.9552
The above sql expression illustrates how to use positional calls/beaten lengths in a UDM... feel free to experiment and use the other positional calls/beaten lengths in a similar manner.
Note the position of the parenthesis characters... Using them in the above manner allows us to use OR conditionals so that everything fits nicely into a single UDM... Without the OR conditional we'd need to use multiple UDMs to flag the same set of horses.
-jp
.
|
JustRalph 6/12/2010 1:33:25 AM | I don't use Sql UDM's
|
jeff 6/12/2010 10:00:18 AM | Same concept expressed as a UDM this time in playlistfile mode using a 90 day HDW data sample from a different time period:
query start: 6/12/2010 6:43:58 AM query end: 6/12/2010 6:45:14 AM elapsed time: 76 seconds ` Data Window Settings: Divisor = 999 Odds Cap: None PScore_HDW: MinVal=90 Filters Applied: 2CALLLENMAX02-!ALL!MIN8- ` Surface: (ALL*) ROUTES From Index File: C:\2010\Feb2010\pl_profile.txt
` Data Summary Win Place Show Mutuel Totals 821.90 772.00 703.30 Bet -764.00 -764.00 -764.00 Gain 57.90 8.00 -60.70 ` Wins 73 133 174 Plays 382 382 382 PCT .1911 .3482 .4555 ` ROI 1.0758 1.0105 0.9205 Avg Mut 11.26 5.80 4.04
2CALLLENMAX02 is a preset filter code requiring that the horse have been within 2L at the second call in its most recent start.
!ALL!MIN8 is a dynamic filter code requiring that horses (from all classes and ages) selected by the UDM were beaten by at least 8L at the finish call in the most recent start.
-jp
.
~Edited by: jeff on: 6/12/2010 at: 10:00:18 AM~
|
jeff 6/12/2010 9:58:55 AM | --quote:"I don't use Sql UDM's" --end quote
Note the difference in elapsed query times:
SQL Mode: 2 seconds
Playlist File Mode: 76 seconds
-jp
,
|
JustRalph 6/12/2010 12:31:17 PM | !ALL!MIN8
Thanks Jeff, I know it's much faster, I am just lazy........ I actually used to use sql back in the day.... and forgot most of what I knew about it then..... it was a for a Support Job I had.
I will change eventually I guess.
Is the number 8 changeable in the example above? or do I have to use the 8 ?
|
jeff 6/12/2010 12:38:38 PM | The interface in the Dynamic Filter Tool allows you to specify an integer value for beaten lengths.
If you understand the individual elements of the filter code itself you can also do it via manual edit of the filter code in the FilterCodes field right on the Modify Screen in the UDM Wizard.
Changing !ALL!MIN8 to !ALL!MIN7 would change the dynamic filter code to require that horses (from all classes and ages) selected by the UDM were beaten by at least 7L at the finish call in the most recent start instead of 8L.
-jp
.
|
JimG 6/12/2010 1:49:26 PM | Hi Ralph,
Not to get off track on the subject but the best part of using sql mode is being able to create your own html report, sort it like you want, and use different fonts for certain things if you like.
Jim
|
JustRalph 6/12/2010 8:54:26 PM | Thanks Guys........ appreciate it.
Jim, I swear some day I will convert 90 udms to sql........and yes I would love to have the ability to edit the html report. But for now, I am looking up at the mountain......and wondering how long it will take.........
|
JimG 6/12/2010 10:50:48 PM | Ralph,
90 active udm's. Wow. I have 7-9 I use depending on what surface the race is. My udm's are so simple. They are designed to hit me over the head so I do not miss something on the html report and allow certain factors to converge.
If I had 90 active udm's I doubt I would change over either. But since I handicap from my customized html report, sql mode is a must.
Take care.
Jim
|
JustRalph 6/14/2010 9:05:19 AM | maybe I am a freak, but I actually do use all 90 as active......sometimes I have one hell of a long HTML report....but I never print them anyway. The UDM's are color coded and I add HTML code to something I really want to stick out. Normally just bolding the UDM.
I have actually already developed one from this thread
It seems to be doing what I wanted it to. It has also turned me onto some other interesting info on horses that are stretching out.......across a few different databases....... it is amazing to me how certain things are universal in the sport. Jcapper shows some of those universal truths very well. Many of them go against often repeated items of info that JCapper reveals to be utter nonsense. I like that part of the app too.
I know more about evaluating a race now than ever. And I really really enjoy it. But finding good races to bet is the problem.
|
GlennF 6/14/2010 9:27:14 AM | Others may disagree but in my opinion even greater than the increase in speed in SQL are the things that you can do with SQL that can't be done in playlist mode. This is especially true if you also use HDW files. Of my SQL UDMS I have zero that are straight ports of (even successful) playlist UDMS and my recent ROI justifies the complete switchover to SQL and HDW.
|