Hi, I came across your project and found it quite interesting because I used to scape odds for my ex-company.
Your requirement is big because it integrates with many bookmarker.
My last company need to do in 6 months with 3 full time headcount, which may cost around 80k, just to integrate with 2 partner.
The challenges here is
1. different bookmarker has different logic to scrape and they keep improving to prevent robot. So to keep our program robust, we need to handle a lot of scenarios and exception, with strong infra (for 1 bookmarker, may need 2-3 servers, with ability to generate dynamic outgoing IP so that bookmarker will not detect us as robot)...
2. How to sync data between bookmarkers, means for matches, leagues, teams, how to know they are the same among these bookmarkers (cannot just base on name).
3. and many many more
And another alternative, instead of having own odds database, we can just scrape a website that has odds comparison feature, rewrite look and feels using proxy server (apache), then it may serve this purpose.
The amount I put here is just for illustration, but if you want to scape all sites, the cost may be infinite and need to redo in the end. So better to start with a few, 2-3 or scape another site that already compare these data.
Thanks.