search

Seb943 / scrapeOP

A python package for scraping oddsportal.com
211 stars 76 forks source link

Update to oddsportal website means that scraper now fails #8

Open CanProjects opened 1 year ago

pferreira8 commented 1 year ago

Sadly dealing with this issue too. I've used this code for going on 2 years now in my nba model pipeline. I'm not the developer of this project but I will try and identify the root issue.

ApuestasPPH commented 1 year ago

I miss the old Odds Portal website does this code still work with the New Site?

andrewting19 commented 1 year ago

https://codereview.stackexchange.com/questions/283181/code-scraping-oddsportal-using-beautifulsoup

cardchase commented 1 year ago

https://codereview.stackexchange.com/questions/283181/code-scraping-oddsportal-using-beautifulsoup

Yep, I posted the question. This code works better as the new site is an infinite scrolling site so the more you scroll, the more odds you get.

andrewting19 commented 1 year ago

Thanks for the code! Do you know of any way to scrape the historical odds on the results pages?

cardchase commented 1 year ago

Thanks for the code! Do you know of any way to scrape the historical odds on the results pages?

Replace the page url with the page you'd like to scrape. Usually the date works fine from 1 jan 23 onwards.

andrewting19 commented 1 year ago

Unfortunately doesn't work for links like this one (https://www.oddsportal.com/esports/usa/league-of-legends-lcs/results/). I ended up writing my own janky code to do it lol

CanProjects commented 1 year ago

Unfortunately doesn't work for links like this one (https://www.oddsportal.com/esports/usa/league-of-legends-lcs/results/). I ended up writing my own janky code to do it lol

I have my own league of legends pipeline (its actually why I asked this question but i ended up writing the scraper code myself and scraping it) but I have been unable to obtain a superior logloss to the regular betting odds, have you been able to fit a meaningful model?

andrewting19 commented 1 year ago

Haven't tried moneyline yet. I'm using the odds as a feature for modeling kills instead.

SSaltyyy commented 1 year ago

Unfortunately doesn't work for links like this one (https://www.oddsportal.com/esports/usa/league-of-legends-lcs/results/). I ended up writing my own janky code to do it lol

Could you post it? Or if someone could update the original scrapeOP...

SSaltyyy commented 1 year ago

Thanks for the code! Do you know of any way to scrape the historical odds on the results pages?

Replace the page url with the page you'd like to scrape. Usually the date works fine from 1 jan 23 onwards.

Result pages are without infinite scrolling so this doesn't work.

oconbball4 commented 5 months ago

https://codereview.stackexchange.com/questions/283181/code-scraping-oddsportal-using-beautifulsoup

Yep, I posted the question. This code works better as the new site is an infinite scrolling site so the more you scroll, the more odds you get.

How do I implement this code? Looking through it not sure where to change the settings on what sport. Trying to pull league of legends historical data.