Xeres wrote: ↑Sun Jan 30, 2022 9:38 am
I ran my own ML algorithms in horse racing for a few years knowing next to nothing about the sport, its not that hard finding consistent value if you have a good model, the hard part is being able to scale the bets to make it worth the effort.
If they really had something that worked the thing to do is find an investor and then scale it, or sell the IP.
I agree it continues to be possible.
I've recently focussed on my own algorithm (of sorts) and have become more confident on its' performance going forward. I use the term algorithm loosley, as it means different things to different people. eg: Harry Findlay has a small group of people he listens to, if it tallys with his own opinion, he'll increase the % on the bet. For others it means a computer coding solution to create a list of rules to find a solution/recommendation.
For my own purposes, when ready, I can scale it comfortably 10x (maybe more) without using any on-line or whatsapp bookmaker. An investor or group of investors would move things along quicker, but IMO, it's good to grow/compound from a smaller bank and then work with capital from clear profit(s). You're also not beholden to others or get questioned when the losing runs occur.