The Dota 2 Landscape in R
summary
Downloads a curated CSV of pro matches from DAP Toto, queries the match API to pull JSON, and uses R JSON flattening and list unnesting to tidy nested match records for analysis. Finds the 2018 prize pool was 25.5 million prize pool and shows that winning teams hold >50% kill share until ~45 minutes, after which kill share weakens as a predictor. In games with a 0–10 kill-gap the team kill participation metric strongly predicts victory.
AI-written summary · about
chapters (12)
- 0:00Introduction
- 0:25What is Dota 2
- 1:09Why choose Dota 2
- 1:51Recipe
- 2:30API
- 2:41Data wrangling
- 3:05Data structure
- 3:29Terminology
- 4:02Results
- 4:47Kill Share
- 5:23Kill Gap
- 6:02Conclusion
related talks
35mDecoding bias and narrative in competitive video games
41m"R" You Ready for Some Football? Hacking Fantasy Sports with Open Source Software
49mBehind the Streams: In Depth Look At ML&AI Powered Analysis of FPS Gameplay
47mData Sketches: a year of exotic data visualizations
19mTop datasets for top DevRel results
14mModeling Individual and Team Behavior through Spatio-temporal Analysis
56mRecreating the State of the Player
21mUsing Open Data to Predict Market Movements