-
Feedback form
-
Behind-the-scenes data
-
Brief description
-
Calculated 5-man roster
-
Work in progress
Inspiration
There are easily over a dozen CS:GO fantasy and betting websites that report dealing out 100's of thousands of dollars a week. I briefly tried my hand at $0.50 leagues to only get destroyed, despite being an avid fan. I knew there was a better way of getting a head. Combined with my passion for gaming, and my partner Raj's passion for economics, we set foot to create a statistical gaming tool.
What it does
CS:GO Forecast is a web platform dedicated to evaluating the performance of eSports professionals and their careers playing Valve's blockbuster PC game, Counter Strike : Global Offensive. Historical and live statistics are pulled directly from the source of top professionals as they play. This data is aggregated, analyzed, and published as a roster of a 5-person fantasy draft.
How we built it
We accumulated Steam profiles of over a dozen of the top CS:GO professionals and put Steam's Web App API to use. Using PHP, we data mined as much statistics Steam published, which ended up being more than we needed. From the PHP script, the data is logged into a MySQL database to then be cross examined and presented in a user-friendly manner.
Challenges we ran into
We spent 3 hours trying to get a local Apache server running via xampp before getting another machine. Multiple expected roadblocks in learning a new platform such as MySQL - a lot of trial and error.
Accomplishments that we're proud of
Using PHP and MySQL to push data back onto a web template. Manipulating a large amount of data.
What we learned
First time working with MySQL - never realized how valuable it can be for web apps with large data sets. PHP may not have been the best language to use, but it certainly broadened my web dev perspective. Existing documentation on the 3rd party API was extremely limited, we spent a fair amount of time combining our previous programming knowledge to accomplish our goal.
What's next for Counter Strike : Global Offensive Forecast
Additional analysis on most recent games: predict trends of new and upcoming players. Pulling unused data from multiple profiles can be slightly taxing on computation time - we would like to continue streamlining the algorithms used. Hope to eventually push the app into production and release it to the public.
Log in or sign up for Devpost to join the conversation.