|
|
02-25-2004, 08:26 AM
|
#46
|
Registered User
Join Date: Nov 2003
Location: Ohio
Posts: 1,307
|
Tom wrote:
I have a database of many factors.
What would you suggest I do to start a regression study?
Look at single factors, combinations?
I could use Excel to plot finsh positions or beaten lengths of horses with certain factors by rank, look for postivie correlation?
##
Tom, check the above discussions on regression analysis. For a number of reasons this statistical approach is problematic in horse racing. In sum, unless you carefully control for multicolinearity (high correlation of variables) the results are suspect, certainly their Beta values would be. Also, you need to have probit or logistic curve as your dependent variable. Finally if you go on a fishing expedition throwing independent variables into the models you run the risk of interpretability problems. Remember that correlation is not causation. Correlation relationships will ultimately regress to the the mean and would make a poor basis for wagering.
Trying to save you some time and maybe ($'s)
|
|
|
02-26-2004, 12:30 AM
|
#47
|
Registered User
Join Date: Feb 2004
Location: Windermere, FL
Posts: 151
|
Hi Game Theory,
Thank You. That was a great post and informative post! Your illustration of gap variation was great. I whipped up a new way (for me) to look at various ratings based on various gap sizes. It is too early to tell how well it will do overall, but on simple tests, it has a pretty good impact on winning percentage. Unfortunately, it doesn't appear to do anything for ROI, but I haven't thrown it into my "grinder" yet. I have great hopes that it will wind up making my odds-line significantly more accurate.
Thanks again!
|
|
|
03-24-2004, 04:17 PM
|
#48
|
Registered User
Join Date: Feb 2002
Location: asia
Posts: 48
|
Why did Benter choose Hong Kong?
because i went there!!
__________________
"An inexhaustible good nature is one of the most precious gifts of heaven, spreading itself like oil over the troubled sea of thought, and keeping the mind smooth and equable in the roughest weather." Washington Irving
|
|
|
03-24-2004, 04:37 PM
|
#49
|
Join Date: Mar 2001
Location: Reno, NV
Posts: 16,908
|
Entropy,
And that would make you Allan?
Regards,
Dave Schwartz
|
|
|
03-24-2004, 04:44 PM
|
#50
|
Registered User
Join Date: Feb 2002
Location: asia
Posts: 48
|
|
|
|
03-24-2004, 05:17 PM
|
#51
|
Registered User
Join Date: Mar 2002
Location: Hamden, CT
Posts: 420
|
Modeling
Game Theory, excellent post. Good insights.
All The Best,
Jaguar
|
|
|
03-24-2004, 06:21 PM
|
#52
|
Registered User
Join Date: Feb 2004
Location: Paragould, Arkansas
Posts: 198
|
Allan
Since we are so interested in the computer teams in Asia and for that matter North America, Why don't you enlighten us on what it has been like. Tell us about the early days and everything else you want to tell us. I will tell you all of the people that post here find the computer teams to be fascinating. If it takes a novel to post it, please do it.
|
|
|
03-24-2004, 06:30 PM
|
#53
|
Registered User
Join Date: Feb 2002
Location: asia
Posts: 48
|
arkansasman
you know the URL. and as i said only one l.
|
|
|
03-24-2004, 06:53 PM
|
#54
|
Registered User
Join Date: Feb 2004
Location: Paragould, Arkansas
Posts: 198
|
Allan
Ok. LOL What is your understanding of the reason why Benter left Hong Kong. It would not surprise me if he is betting somewhere else.
|
|
|
03-24-2004, 07:10 PM
|
#55
|
Comfortably Numb
Join Date: Sep 2001
Location: Lexington, Ky
Posts: 6,174
|
Sorry for butting in , but he is trying to say "only one ell" i.e. Alan
|
|
|
03-24-2004, 07:12 PM
|
#56
|
Registered User
Join Date: Feb 2004
Location: Paragould, Arkansas
Posts: 198
|
Bill
Thanks
|
|
|
03-24-2004, 07:18 PM
|
#57
|
Registered User
Join Date: Feb 2002
Location: asia
Posts: 48
|
why Benter left Hong Kong?
maybe partly connected with the reason i left. don't ask me why --- do a search!!!
|
|
|
03-24-2004, 08:31 PM
|
#58
|
Veteran
Join Date: Mar 2002
Location: new york city
Posts: 1,424
|
Guys...
Benter .... dont u guys learn anything i say ur worse than inner
city black & assorted non-white thugs who got 0 assets ...
that guy knows 0000 ... racing is NOT a math game .... accept
that .... but math can be used to turn a profit ... when at which
point .. well thats the trick ... neural nets .... lol ... maybe i should
stop saying what i do (since i get NO clear response) and stay at
this other board .... (maybe i just CANT forgive u know-it-alls for
not helping in that MAJOR board contest last year .. Boris was the
Captain LOL)
|
|
|
03-24-2004, 08:56 PM
|
#59
|
Join Date: Mar 2001
Location: Reno, NV
Posts: 16,908
|
Alan,
lol - I consider myself a student of the game but never thought to Google "Benter." You suggested a search and I did one. Amazing what you find when you search for Benter. Thanks for the idea.
Welcome aboard.
Regards,
Dave Schwartz
|
|
|
03-24-2004, 09:19 PM
|
#60
|
Registered User
Join Date: Mar 2002
Location: Hamden, CT
Posts: 420
|
Math Game
Derek, why would you say racing is not a math game?
In fact, that's exactly what racing is- finding patterns and betting overlays- a spot game which we analyze using mathematical tools.
Horses are athletes that demonstrate discernible patterns in their past performances, while trainers are manipulating their charges to win purses- in measurable, repeated behaviors.
We handicappers love the horses and the romance of the game, but we also know that racing is very much a math game.
All The Best,
Jaguar
|
|
|
|
|
Thread Tools |
|
Rate This Thread |
|
Posting Rules
|
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
HTML code is Off
|
|
|
|
|