Smartgambler
Pro-Punter

Go Back   OZmium Sports Betting and Horse Racing Forums > Public Forums > General Topics
User Name
Password
Register FAQ Search Today's Posts Mark all topics as read

To advertise on these
forums, e-mail us.

Reply
 
Thread Tools Search this Thread Display Modes
  #11  
Old 3rd March 2018, 03:05 PM
walkermac walkermac is offline
Member
 
Join Date: Nov 2013
Posts: 1,037
Default

Mixed results.

Last night's favourite What Lizahead never looked likely. It was overtaken in the betting by The Kroc, who struggled into 3rd. Those two were both ruled out by me on account of them having the worst CDs in the field for the given race distance.

While those ruled out by Dosage did well, the Dosage-based picks didn't fare so great. Best of them was Big Blows who finished in 4th after going via the carpark.

The winner was Melba's Maestro - at $20 - who never got a look-in in the selections on account of having a small Classic figure (the second smallest in the field). He does looks slightly more appealing when using the Australian/NZ chefs-de-race though (and What Lizahead and The Kroc look just as unsuited).
Reply With Quote
  #12  
Old 15th March 2018, 01:33 PM
walkermac walkermac is offline
Member
 
Join Date: Nov 2013
Posts: 1,037
Default

Quote:
Originally Posted by walkermac
He does looks slightly more appealing when using the Australian/NZ chefs-de-race though (and What Lizahead and The Kroc look just as unsuited).
Alright. I did it. Here are some results using the list of Australian/NZ chefs-de-race.



Similar to using the official list, there does seem to be a tiny advantage to have a higher CD in races 1200m or less. (The 95% confidence interval for the mean is show in the next graph).


Eagle-eyed viewers may have noticed something else: the CD using the list of Australian/NZ chefs-de-race is more descriptive than that derived from the official list. The range over 900m to 2400m is 50% greater than that using the official chefs. If you're using dosage in this part of the world, it looks like it may be a smart idea to include local chefs. Whodathunk?!




In the race above I successfully excluded the two favourites on account of them being outside of the 90-95% confidence interval of a normal distribution with the given sample mean. (I haven't actually confirmed that the CDs are normally distributed yet, but the Central Limit Theorem would imply that they are). Here's what the graph looks like for a 95% interval. In theory, were the list of chefs-de-race maintained and there was little change to the distribution of horses/abilities entered in each race distance, runners outside of the range should have win odds $20 or greater. ....though whether that was true for any one particular race, rather than across all races, is a different story

Reply With Quote
  #13  
Old 16th March 2018, 12:38 AM
walkermac walkermac is offline
Member
 
Join Date: Nov 2013
Posts: 1,037
Default

Using the Australian and New Zealand chefs-de-race also makes some other figures more sensible. Here is the Dosage Index (the ratio of speed to stamina):




Most of the totals of each category vary over race distance, quite unlike what was noted when using the official list of chefs.




Dosage Total is still not a great indicator of class though (if 'class' is winning races with a large prize pool):

Reply With Quote
  #14  
Old 19th November 2018, 02:01 PM
thorns thorns is offline
Member
 
Join Date: Aug 2007
Posts: 731
Default

Been doing some grave digging and resurrecting this thread.
Have had some success (beginners luck?) finding some losers using what you have outlined in the above posts.

Walkermac, is there a way to find the dosage of a unraced/unnamed horse?
Just for interest sake I want to run a small project on some horses which are being sold at the RTR sales in NZ end of this month, would be interested in noting down a few well bred horses with good dosage profiles to try follow along on paper.
Reply With Quote
  #15  
Old 20th November 2018, 01:38 AM
walkermac walkermac is offline
Member
 
Join Date: Nov 2013
Posts: 1,037
Default

Quote:
Originally Posted by thorns
Walkermac, is there a way to find the dosage of a unraced/unnamed horse?
Just for interest sake I want to run a small project on some horses which are being sold at the RTR sales in NZ end of this month, would be interested in noting down a few well bred horses with good dosage profiles to try follow along on paper.
There are a couple of ways. The simplest way is to do a hypo(thetical) mating at pedigreequery.com. This function is only available to members though ($10/month; $50/6 months). The other drawback is that it only uses the official chefs-de-race, which might not be as illuminating as using the AU/NZ chefs.

You can actually make your own spreadsheet pretty easily, though it would take a couple of minutes to do the data entry for each horse you wish to examine (actually, if you've got some skills you could just scrape their webpage and fill the data automatically).

Get the list of both official and AU/NZ chefs from http://dosageprofile.com/anz-chefs. Convert the textual chef attributes to numerical ones (i.e. a Brilliant chef's row has a 1 in its Brilliant column and a 0 to the rest; a Brilliant/Intermediate chef has a 0.5 in its Brilliant column, a 0.5 in its Intermediate column and a 0 in the rest). Should take 10-15 minutes.

We do this so we can determine the total of any particular aptitude of any sire using something like:
Code:
IFERROR( VLOOKUP(CellReferenceOfSire, sheetWithChefDetails!A:F, numberOfAttributeColumnWeWant, FALSE), 0)

Next set up a data entry sheet that looks the same as a 4 generation pedigree. Off to the side someplace, for each generation and each attribute, use the above VLOOKUP on each relevant sire to keep a tally. For generation 4 the formula will be ridiculously long (you'd have to sum eight VLOOKUPS - one for each sire in that generation - each of which would be wrapped in an IFERROR in case the sire isn't a chef at all).

Recall that parents are twice as important than grandparents who are twice as important than, etc. The way it practically works is that the 4th generation total for each aptitude is multiplied by 2, the 3rd gen by 4, the 2nd gen by 8 and the 1st gen (parents) by 16. These are summed together to get the horse under examination's total for this aptitude category.

Quote:
Originally Posted by Wikipedia
This results in a Dosage Profile consisting of five separate figures, listed in order of Brilliant-Intermediate-Classic-Solid-Professional. Secretariat, the 1973 Triple Crown winner, for example, had a Dosage Profile of 20-14-7-9-0. To arrive at the Dosage Index, the first two figures plus one-half the value of the third figure are added together, and then divided by one-half of the third figure plus the sum of the last two figures. In this case, it would be 37.5 (20 + 14+ 3.5) divided by 12.5 (3.5 + 9 + 0), giving Secretariat a Dosage Index of exactly 3.00 (the figure almost always being expressed with two places to the right of the decimal point and rounded to the nearest .01).

A second mathematical value, called the Center of Distribution, can also be computed from the Dosage Profile. To determine this value, the number of Brilliant points in the profile is doubled, and added to the number of Intermediate points; from this is then subtracted the number of Solid points and twice the number of Professional points. The result is then divided by the total number of points in the entire profile, including the Classic points. In Secretariat's case, this would work out as 54 (40 + 14) minus 9 (9 + 0) divided by 50 (20 + 14 + 7 + 9 + 0), yielding a Center of Distribution of 0.90 (the figure nearly always being rounded to the nearest 100th of a point, as with the Dosage Index).
It'll be easy to make a mistake so test your sheet first by comparing it with pedigreequery's output for a horse featuring no sires from the AU/NZ list.
Reply With Quote
  #16  
Old 21st November 2018, 09:16 AM
thorns thorns is offline
Member
 
Join Date: Aug 2007
Posts: 731
Default

Thanks for that Walkermac. Appreciate the research and postings you have made.

On another note, most of the tables that were in the post seem to have disappeared for me, is this the same for everyone else?
Reply With Quote
  #17  
Old 21st November 2018, 10:05 AM
walkermac walkermac is offline
Member
 
Join Date: Nov 2013
Posts: 1,037
Default

Quote:
Originally Posted by thorns
On another note, most of the tables that were in the post seem to have disappeared for me, is this the same for everyone else?
Yeah, it looks like the image hosting site tinypic is currently down. I can't see anything about them having shut down, so hopefully it's just a temporary thing.
Reply With Quote
Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is On
Forum Jump



All times are GMT +10. The time now is 06:47 PM.


Powered by: vBulletin Version 3.0.3
Copyright ©2000 - 2018, Jelsoft Enterprises Ltd.
2008 OZmium Pty. Ltd. All rights reserved . ACN 091184655