Introducing self

[Buchan]

Hello people.

I have just built up the courage to make my first post.

I have been reading this board for a long time, and there are obviously some very clever people here, so have been hesitant to dip my toe in until now as I don't want to look too silly.

I am really interested in the following topic I found here.....

http://www.paceadvantage.com/forum/showthread.php?t=35264

That topic about Harville discount has me confused, although that's not hard to do.

I can figure how to use lambda to calculate the probabilities, but for the life of me, I cant understand how to arrive at the lambda number in the first place.
Could some lovely dear please explain in simple terms how to calculate lambda from the following .

class, publicP, actualP
1, .467099 , .366492
2, .342353 , .328877
3, .273493 , .241379
4, .223371 , .230252
5, .173147 , .17455
6, .122878 , .124636
7, .072397 , .079312
8, .036324 , .042836
9, .017084 , .018597
10, .007561 , .078

Thank you very much

Kristy Buchan

[andymays]

Welcome Kristy! :ThmbUp:

I'm not much on the numbers end of things so I won't weigh in. You will find several people on the Board are experts in that area!

[Buchan]

Welcom Kritsy! :ThmbUp:

I'm not much on the numbers end of things so I won't weigh in. You will find several people on the Board are experts in that area!


Thank you so much for the warm welcome andymays.

I have to go to bed now as it's past midnight in Australia, but hope when I awaken in the morning, I shall be able to reply to any posts If I'm lucky enough to have received any.

Kristy

[andymays]

Thank you so much for the warm welcome andymays.

I have to go to bed now as it's past midnight in Australia, but hope when I awaken in the morning, I shall be able to reply to any posts If I'm lucky enough to have received any.

Kristy


I'm in California near Del Mar. Kayla Stra rides out here. Are you familiar with Kayla?

http://en.wikipedia.org/wiki/Kayla_Stra

Kayla Stra (born December 3, 1984 in Adelaide, South Australia) is a jockey in Thoroughbred horse racing who raced in Australia until coming to compete in the United States at racetracks in California. As a baby, her family moved to Fountain Valley, California where they remained until she was five years old. After returning to her homeland, she left school at age thirteen and went out on her own. For a time she worked on a farm then became involved in Australian flat racing.

Stra became a leading Adelaide apprentice jockey who posted 220 winners in Australia. Her biggest Australian win was in the 2005 City of Marion Stakes with gelding Navy Shaker.

Racing in California, on November 9, 2007 Kayla Stra won her first American race aboard Flying Bearcat at Hollywood Park Racetrack.

Stra is one of the jockeys featured in Animal Planet's 2009 reality documentary, Jockeys

[GameTheory]

Hello, welcome.

I think the lambada values in the discounted version were arrived at by those devising that method as the best fit for what they were trying to accomplish. They are "suggested values" that gave good results -- I don't think it is something to be calculated on the fly each time.

Someone can correct me if I'm wrong?

(Creating a thread called "Harville formula" may raise the radar on the right people if if don't get a response in this thread.)

[Buchan]

I'm in California near Del Mar. Kayla Stra rides out here. Are you familiar with Kayla?

http://en.wikipedia.org/wiki/Kayla_Stra

Kayla Stra (born December 3, 1984 in Adelaide, South Australia) is a jockey in Thoroughbred horse racing who raced in Australia until coming to compete in the United States at racetracks in California. As a baby, her family moved to Fountain Valley, California where they remained until she was five years old. After returning to her homeland, she left school at age thirteen and went out on her own. For a time she worked on a farm then became involved in Australian flat racing.

Stra became a leading Adelaide apprentice jockey who posted 220 winners in Australia. Her biggest Australian win was in the 2005 City of Marion Stakes with gelding Navy Shaker.

Racing in California, on November 9, 2007 Kayla Stra won her first American race aboard Flying Bearcat at Hollywood Park Racetrack.

Stra is one of the jockeys featured in Animal Planet's 2009 reality documentary, Jockeys

Hi andymays,

I was not aware of her.
I concentrate on my own state's racing(Victoria), as we have so much racing here it's too hard to follow everywhere.
South Australia(capital Adelaide) is next state west, but I have never had a bet there.

Kristy

[andymays]

Hi andymays,

I was not aware of her.
I concentrate on my own state's racing(Vicoria), as we have so much racing here it's too hard to follow everywhere.
South Australia(capital Adelaide) is next state west, but I have never had a bet there.

Kristy


I believe she appears in the show "Jockeys" on the Animal Planet station here in the United States.

http://animal.discovery.com/tv/jockeys/

You can click on Kayla Stra and find out more about her!

[Buchan]

Hello, welcome.

I think the lambada values in the discounted version were arrived at by those devising that method as the best fit for what they were trying to accomplish. They are "suggested values" that gave good results -- I don't think it is something to be calculated on the fly each time.

Someone can correct me if I'm wrong?

(Creating a thread called "Harville formula" may raise the radar on the right people if if don't get a response in this thread.)

Thank you very much Game Theory, for your welcome and your suggestion, which I will follow up after I finish work today.

I should have chosen a better subject heading.
I should have realised it was a silly heading if I wanted help on something to do with probability.

Thanks
Kristy

[Overlay]

I have heard and can understand the reasoning as to why the horse most likely to finish second in a race may not be the horse that has the second-highest chance of winning, and so on down the line. But it has always seemed to me that if you incorporate a properly weighted and balanced mix of a sufficient number of fundamental handicapping factors that rank a field from top to bottom (including elements that would have a bearing on the unique dynamics of how the race will be run, such as early speed and pace), and apply them from a probability standpoint to a whole field of horses (rather than focusing solely on finding the most likely winner by a go/no-go process of elimination), the resulting composite figure for each horse would be a valid measure for calculating the likelihood of exotic combinations using standard probability formulas, without a need for further modification.

[GameTheory]

I have heard and can understand the reasoning as to why the horse most likely to finish second in a race may not be the horse that has the second-highest chance of winning, and so on down the line. But it has always seemed to me that if you incorporate a properly weighted and balanced mix of a sufficient number of fundamental handicapping factors that rank a field from top to bottom (including elements that would have a bearing on the unique dynamics of how the race will be run, such as early speed and pace), and apply them from a probability standpoint to a whole field of horses (rather than focusing solely on finding the most likely winner by a go/no-go process of elimination), the resulting composite figure for each horse would be a valid measure for calculating the likelihood of exotic combinations using standard probability formulas, without a need for further modification.
But all those probabilities are derived from a study of who wins (generally), rather than getting a correct ranking. When you make a model of winners, you will get different results than if you make a model for who-comes-in-front-of-whom, even using the same factors. (Think of the lone speed horse -- he is either going to run away and win or finish up the track.) And the proof of this phenomenon is of course in the actual data that led to the invention of the discounted version in the first place -- it has been shown over & over that horses don't come in second at the proper rate according to their win probability, even if those win probabilities can be shown to be accurate...