I am a retired University lecturer in Computer Science who now spends most of my time backing horses via automated machine learning methods directly into Betfair
I am a retired University lecturer in Computer Science who now spends most of my time backing horses via automated machine learning methods directly into Betfair
Hi, I am an Italian guy who is studying the world of betting in the most scientific way possible and I would like to ask you for help.
Starting from elements of statistics, in J. Buchdal, etc. I studied a lot, really.
Now I follow several strategies, some have been implemented in a private software (made by a developer, unfortunately I am not a coder).
Tonight, almost by accident, I came across SmarterSig.
Tonight I will start studying and understanding something more, but I wanted to ask you first of all: is it just a theoretical discourse or have you created a downloadable software to then exploit the theory?
I’m reading the site, but I’m not sure what it is.
TY
Mattia
Hi Mattia
Have you watched the demo videoof the software, there is a lin to it on the MySportsAI page of smartersig
Hello,
I’m working on my own model for horse racing and after reading through this blog, I believe we are relatively aligned.
I’m pursuing a graduate degree in computer science specialized in machine learning.
I’d love to discuss some ideas with you.
I’m assuming you can see the email that made this post.
If not, let me know how we can get in touch in a less public forum.
Very Respectfully,
Fred
Hi Fred Good to hear you are modelling. You can contact me via http://www.smartersig.com the email address is on the contact tab
Can your software categorize where particular tipsters are profitable or wherever they lose most.
You would have to have access to data that specifies the selections of a set of tipsters or indeed one tipster and then accompany this with features that you think are relevant to tipster performance. So for example
raceId,tipster,Rtype,Rdist,Horse,BFSP,Finpos
1,Newsboy,Hcp,5,Mill Reef,5.4,0
1,Goodfellow,,Hcp,5,Nijinsky,2.3,1
1,Marlborough,Hcps,5,Golden Fleece,6.2,0
You could then one hot encode the tipster field. However this tipster data would have to be set up by yourself