The largest analysis ever attempted on Dota's least understood item

We've all seen the memes about Rapier's "85% win rate", but how much does this item actually help or hurt the teams that buy it? I analyzed 77,300 ranked replays, looked at purchase rates and trends in each skill bracket, broke down rapier games by hero and built 100 machine learn models to figure out once and for all: when does Divine Rapier even help in Dota? Ain't got time for all that? 6:55 - Initial Findings 8:35 - Rank tier differences/similarities 11:28 - Damage output by hero 12:43 - Beginner's guide to understanding logistic regressions 16:29 - Regression analysis on purchase timings & conditions Please like and subscribe, I've really enjoyed doing video game analysis again, and I'd love to be able to do much more. I'll do my best to respond to comments if I have a direct answer to a question. Want a game other than Dota 2? I'm open to it, let me know. Have a specific Dota 2 question? Leave me a comment and I might answer in a future video! #dota2 #dota #gaming #statistics #dohashtagsevendoanything