|
JCapper Message Board General Discussion
|
| Home |
Register
Log In |
| By | Breeders Cup 2019 | |
| jeff 11/1/2019 1:29:54 PM | I compiled a separate Breeders Cup only database. It contains the Breeders Cup cards for calendar years 2009, 2010, 2011, 2012, 2013, 2014, 2015, 2016, 2018, and 2018. (Fri-Sat only for the Breeders Cup host track each of those ten years.) To make that work, I created a separate BC_Hist-HDW folder. From there, I created one sub folder for each calendar year. Each of the yearly sub folders contain the .JCP, .XRD, and HDW F*.txt files for the Fri-Sat Breeders Cup cards of that calendar year. I copied a fresh blank JCapper2.mdb file onto my c:\JCapper\Exe folder and added the desired SQL Factor Setup. After that I built the first yearly folder (2009) in Mode3. After that, I built each of the other yearly folders (2010 to 2018) one at a time in Mode 4. At that point, the StarterHistory table in the JCapper2.mdb file on my c:\JCapper\Exe folder contained Breeders Cup history for the 10 most recent years. Screenshot showing my folder structure:  After performing the above steps I did some tuning of a Breeders Cup only pace-based UDM I've been using. Some notes about the UDM:
As you can see, the UDM has done well (winning 9 out of the last 10 years.) That said, following the UDM blindly is not without risk. The database covers the last 10 years of the Breeders Cup. At first glance, that might sound impressive. However, the database contains a total of 2334 starters in just under 220 races. In terms of horse racing data, that's a small sample. I mentioned that the UDM is pace-based. That means it completely whiffs on foreign shippers (because they do not have speed or pace figs in their past performance records.) The year that the UDM lost was 2017 and it went 1 for 16 at DMR. Data at DMR that year leading up to Breeders Cup Fri and Sat suggests an environment that was hardly speed favoring. Link to some Data Window output here: Impact_Study-DMR-2017-Pre-BC.txt7 The above data sample shows CPace underperforming historical norms. It also shows rank 3 for EarlyConsensus outperforming rank 1. At the same time ranks 1 and 2 for HDW Distance Pedigree Rating outperformed historical norms. (To me, this suggests the environment wasn't speed favoring or normal.) Here's a link to a thread on Paceadvantage about this year's version of the Santa Anita surface for the Breeders Cup being deeper and more tiring than normal: http://www.paceadvantage.com/forum/showthread.php?t=155032 CJ, whose work I greatly respect, commented you could have timed Thursday's races with a calendar.  Here's a link to a text file showing some Data Window output for the current Santa Anita meet: Impact_Study-SAX-2019-Pre-BC.txt Unlike the DMR sample from 2017, meet to date data for the current 2019 Santa Anita meet has seen both CPace and HDW Distance Pedigree Rating overperforming historical norms. Has the track maintenance crew at Santa Anita deadened the dirt surface? Will early speed be compromised in this year's version of the Breeders Cup? Or will it be ok? I guess we'll know in a few hours. -jp . ~Edited by: jeff on: 11/1/2019 at: 1:29:54 PM~
| |
| jeff 11/1/2019 1:09:33 PM | Right now as I type this, it's about 12 mtp for R2 and the above linked to UDM flagged the following horses for the rest of today's BC card at Santa Anita: R2 #9 HEATHERS GREY R3 #1 HOMEHOME R4 #2 CALI DUDE R4 #8 GREAT ULYSSES R5 #9 FOUR WHEEL DRIVE R7 #4 BRITISH IDIOM R7 #7 WICKED WHISPER R10 #7 SECRET COURIER R10 #9 RICKS DREAM Good luck to all, -jp .
| |
| cglove1 11/1/2019 6:35:59 PM | Jeff, This is more than I could have expected. Appreciate your time. Thank you
| |
| jeff 11/3/2019 2:15:36 PM | The Equibase Scratches and Changes XML was full of holes yesterday. I first noticed it when I tried to play the early P5 at AQU only to have my ticket fail because there were horses on it that were scratched from the tote that hadn't been reported as scratches in the Equibase XML. (Ugh.) Fyi, I posted about it over at Paceadvantage in the Death by 1000 Paper Cuts thread: here. --quote: "Right now as I type this it's about 12:30 eastern time on Breeders Cup Saturday -- A lot of tracks have early post times today. Race 1 at AQU has already been run. Race 1 at FLX has already been run. Finger Lakes reported their early scratches and changes to Equibase about 8 minutes after race 1 had been run. With about 1 mtp for race 2, AQU has yet to report their early scratches and changes to Equibase. How does this happen?" --end quote. And: --quote: "Received an email a few minutes ago from someone fairly high up on the ladder at Equibase -- One of the Jockey Club networks is down. The network itself (among other things) is designed to act as a fail safe by parsing the tote to pick up scratches that haven't been reported to Equibase by the tracks. Right now as I type this, scratches and changes in the XML are those that have been reported to Equibase by the tracks. Missing from the XML are scratches and changes not reported by the tracks that (under normal circumstances) would be picked up by the back up system that parses the tote. Hope I managed to type most of that in a way that makes sense. Of course none of this would be happening if ALL of the tracks were making a good faith effort to report their scratches and changes directly to Equibase in a timely manner. Now you know what I know." --end quote. Things were spotty at best until about 7 pm eastern. Every time a race came up I thought might be playable, I found myself looking at track video and manually scratching one or two horses that hadn't been reported as scratches in the XML. From there I'd run a single race Calc Races and (quickly) look at HTML Two in the Report Viewer. It wasn't until that point I finally knew whether or not I was looking at a playable race. I ended up doing that for almost 100 races yesterday. In a nutshell, that's why I didn't post UDM selections for Saturday's Breeders Cup card like I did for Friday. I realize this is after the fact, but this is how my pace-based Breeders Cup UDM fared over the two days: Impact_Study_BC-PaceUDM2019.txt Hardly outstanding. But not terrible either. Imo, the grounds crew did have the Santa Anita dirt surface deeper and more tiring than normal. Here's a link to a screenshot showing the final quarter times as compiled by Trakus for R6: SAX R6 Trakus Imo, when the (supposed) best dirt milers in the world get a mile on a fast dirt surface in 1:36.58 on a clear 81 degree day (with minimal wind) and 9 runners in a field of 10 get the final quarter in 26.00 seconds or slower: There's something unusual going on. Fyi, after Equibase got The Jockey Club's network back up - I'm told they had to manually stitch together and re-key all of the scratches from all of the races that had already been run. Apparently, their day was even more of a grind than mine was. It seemed to me like they finally got caught up sometime around 7 pm Eastern. I'm happy to report the Equibase scratches and changes system appears to be running smoothly today. -jp .
| |
| jeff 11/3/2019 7:15:21 PM | Breeders Cup Betting Challenge final leaderboard: https://www.breederscup.com/bcbc/leaderboard Congrats to Blake Courtney ($39,400 10th place.) Nice work! -jp .
|
| Reply |
|
Copyright © 2018 JCapper Software back to the JCapper Message Board www.JCapper.com |