speed figure comparision

[osophy_junkie]

Hello,
I'm in the process of creating my own competition comparision figures. I do not have access to Beyer or "the sheets" figures. Does anyone with access to these know the percentage of winners the lowest number in a set of entries produces.

regards,
Ed

[garyoz]

The Sheets and Thorograph are far more complicated than taking the best last race number, as I believe that you are suggesting. A starting point might be to read Len Ragozin's book, The Odds Must Be Crazy.

A study was done several years ago (probably 1996?), that compared the last best number of several speed fig systems. I think it may have been done by Nunamaker (?). I think I remember seeing it mentioned on this board. You might want to do a search.

[Bill Cullen]

Originally posted by garyoz
The Sheets and Thorograph are far more complicated than taking the best last race number, as I believe that you are suggesting. A starting point might be to read Len Ragozin's book, The Odds Must Be Crazy.

A study was done several years ago (probably 1996?), that compared the last best number of several speed fig systems. I think it may have been done by Nunamaker (?). I think I remember seeing it mentioned on this board. You might want to do a search.

I know in Nunamaker's book "Modern Impact Values" that he implies in several places that Beyers can be substituted for Brisnet speed figures but I don't recall ever seeing a comparative study between Beyers and Brisnet figures that validates that assumption.

Bill C

[Valuist]

When you talk about a "best number", are you referring to a horses last race? Its career? In the past year? What if that last race was at a completely different distance or surface? I think the best way to compare Beyer, Brisnet, T-Graph and Ragozin would be to compare a horse's best number, within the past year, at a similar distance (no more than 1 furlong more or less) and over a similar surface/track condition.

[garyoz]

Deep within my library of racing "literature" I was able to find the study. It is Speed Ratings Comparison compiled by Sport Stat and published in 1994. The cost of the publication was $45. At a total of 27 pages, the study examined the ratings for Bris, Ragozin, Thorograph, DRF, Kuck speed & pace Ratings, Pugliese Pace Ratings, Trackmaster Power Ratings, Meadow's Win Ratings, RPM Power ratings, and TIS speed ratings. The methodology used the last race and the best one of last three.

I don't want to step on the copyright, although the data are 10 years old. Ragozin did the best of the speed ratings.

Needless to stay this is pretty old stuff. Maybe the study is of historical significance? Amazing the old stuff I still have, including the original Nunnamker's Impact Value Study.

[Niko]

I've tried to contact Jim Bayle and buy this book (see other posts regarding Jim Bayle). He's no longer to be found. Because of that fact only; can you re-cap some of the significant findings regarding the results. I know it's old but I'm particularily interested to see how ITS, Bris, Beyer and Ragozin did in comparison to win percent and ROI.

[garyoz]

Happy to provide a recap if there's no copyright issue. I don't know the story on Bayle. I'm jammed the next few days on some writing, but will get to by the end of week.

BTW, ITS was using Beyer figs until 1998. Remember this study was from 1994. The TIS refers to Thoroughbred Information System--Robert Velasquez. I think they are no longer around.

[osophy_junkie]

I use a projected number that is generated based on several factors for my testing. It isn't the only part of my system and I'm trying to gage it's usefullness by comparing it's winning ability to other speed ratings. I don't have access to these ratings so I can't back test my system on them and compare results.

I was hoping winning percentage produced by choosing the best projected number out of a set of race entries would be a decent benchmark.

regards,
Ed

[bobphilo]

I know in Nunamaker's book "Modern Impact Values" that he implies in several places that Beyers can be substituted for Brisnet speed figures but I don't recall ever seeing a comparative study between Beyers and Brisnet figures that validates that assumption.

Bill C
Here's a result of various studies that compares equivalents for Beyers, Brisnet, Timeform, Equibase and other figures.
The file is too large to attach so I'm sending the link. Hope it helps
http://www.chef-de-race.com/pfs/comparative_speed_figs.htm

Bob