PDA

View Full Version : Whos #1: The Science of Rating And Ranking

jfdinneen
03-01-2012, 03:10 AM
I would like to recommend Whos #1: The Science Of Rating And Ranking (http://www.amazon.com/Whos-Science-Rating-Ranking-ebook/dp/B0076LOSHC/ref=tmm_kin_title_0?ie=UTF8&m=APZETQFSWU1YM) for your handicapping library.

John
Dave Schwartz
03-01-2012, 08:52 AM
Ordered. Thank you, John.
TrifectaMike
03-05-2012, 12:26 PM
John and Dave,

I haven't read the book. However, I'm sure it should be a good read. I'm familiar with Dr. Amy Langville both as a researcher and a thesis advisor in the area of Ranking Algorithms. She primarily approaches the subject from a Linear Algebra perspective ( get yourself a good Linear Algebra textbook).

What you will not find in the Ranking Algorithms and techniques is how to convert rankings to probabilities.

For example:

We have N horses ranked by some criteria. How can we assign a probability to the N ranked horses as a function of N and rank order? This is not a simple exercise, and I doubt it'll be addressed in the book.

There are theoretical distributions of distributions which can be used to approximate the probabilities. In fact, these distributions are very useful in horse racing.

For example, if you gave me the ranked order by odds ( not the actual odds) of N horses, I could give you a very good estimate of the individiual odds, without any handicapping.

Mike (Dr Beav)

P.S. Dave, in your New Pace can the contenders be ranked. If they can, I can provide to you a chart which I think you'll like very much.
Dave Schwartz
03-05-2012, 01:05 PM
Oh, alright, if you insist that I must take advantage of what might become a huge breakthrough by learning something new, go ahead and send it to me.

:lol:

Mike, any time you have anything you'd like to send along, please don't hesitate to do so. Email is best.


Dave
jfdinneen
03-06-2012, 05:41 PM
Mike,

With respect to the rank order of favorites on Betfair, I use Benfords Law to generate an initial distribution of win probabilities. Perhaps you could extrapolate this "ranking to probability conversion rule" to other sets of data.

John
Dave Schwartz
03-06-2012, 05:59 PM
Looks like I will not be able to get to this until Thursday or Friday.

Dang.
TrifectaMike
03-06-2012, 06:32 PM
Mike,

With respect to the rank order of favorites on Betfair, I use Benfords Law to generate an initial distribution of win probabilities. Perhaps you could extrapolate this "ranking to probability conversion rule" to other sets of data.

John

John,

Can you PM your email address to me?

Mike (Dr Beav)
Handiman
03-06-2012, 07:27 PM
Just got my book in the mail this afternoon. Going to spend tonight reading it and see if there is anything that will help me with Handifast rankings.

Handi:)
BIG49010
03-08-2012, 12:47 PM
Just got my book in the mail this afternoon. Going to spend tonight reading it and see if there is anything that will help me with Handifast rankings.

Handi:)

I hope you will give us a review, I have been working along these lines with a project of mine.

Thanks
Handiman
03-13-2012, 03:29 AM
I had to put down the book and go back and refresh my Algebra skills. Been out of school for a long time and my math skills have crumbled a bit. If you are not a math person, you will be totally lost in this book.

It is a very good read in a very Math Geeky way. I am only about half way through it. But taking my time as I have a lot of other things going on.

I would advise anyone who really wants to delve into the subject to go ahead and get the book. But it's not an easy read.

Handi:)