I'm using a json file at this stage as it was a quick knee jerk project that i pulled together as an experiment after reading this thread. if it bears fruit, then for sure I'll throw it into a DB.
Yes, I have an API key, but I don't use any BF related info linked into this just yet. However, for analysing things further down the line, then I will link into the BF odds. One thing that is useful with oddschecker odds (and more so with OC bookies [when they come back]) is that you can compare OC odds to BF odds and see if there's any balancing of the book going on that you can take advantage of in BF.
As i said, still not 100% certain of this is a goer or not, but it's ultra cheap to throw up a quick scrape and json feed which I can analyse with a pre-existing stucture, so the only pain point is the hour or so up front in looking at the elements required for the scrape. TBH -i had imagined that oddschecker would chuck everything uot in json format (i.e a hidden api) and transform it in the UI via react or jquery - that would have of course made things far simpler... however!!
runnerjp - as alluded to in one of my prior posts, this is EXACTLY how is see this approach being useful. monitor it from 09:00->13:00 and see which candidates have stayed in the top slots and how their %age vs price has varied. use this to look at low odds lay based candidates. I only have 4 days of data (i've saved all the json from each day), so will look in 4-5 weeks time to see if there's anything to be had in this approach.
jimibt - Sounds like you are going by the exact approach I am. I have seen your name on the BA forum. Are you pulling from the site every 30min? - I was thinking of doing this race by race and monitoring the price movements as a race on the whole but I dont have a BF api key and dont think they will like my scraping their site!
)
