The Curious Case of the Chinese Win Rate ft. TaToH



I noticed something very strange in the stats about how often Chinese win, and thought I’d investigate it a bit further. In the end I decided to call in Tatoh for an interview to hear a pro player’s perspective about it, as well.

Tatoh’s Youtube Channel:

AoEStats:
2000+ elo rankings:

Patreon:

Background music from Epidemic Sound:

Game: Age of Empires 2 HD with Forgotten Empires, African Kingdoms, and Rise the Rajas expansions

Nguồn: https://indiancinema-analysis.com/

Xem thêm bài viết: https://indiancinema-analysis.com/category/giai-tri

Related Post

35 Replies to “The Curious Case of the Chinese Win Rate ft. TaToH”

  1. With a sample size that low, isn't it possible that there is one or two really good players that only play Chinese and skew the stats? Or is the faction randomized?

  2. p value doesnt mean anything, spirit. It assumes that your model is the only possible state in the universe. If your p value is low it doesnt really tell you anything. Google p value wrong if you're interested, theres a good article by towards data science abt it. Awesome video!

  3. Vietnamese will bulldoze the Chinese. Chook sucks. Imperial skirm(best trash unit in the game), elephants and rattan archer.

  4. In the 2000+ ELO, they have horrible match up against Spanish (33%), Incas (35%) and Mayan (40%), I'm not convinced they are top tier even among the elite.

  5. Tatoh suele hacer caminos de torres de bombardas hacia el enemigo cuando juega en equipo, seguro es por eso que les gusta tanto las chinas. Es raro escucharlo en ingles.

  6. I'm probably very bad at this game compared to others, but I practically only play chinese. I only play against AI and just mess around every once in a while though. I am hooked on upgrades and trash units though so might look into other civs that are good at that to mix things up a bit.

  7. Was anyone surprised at how Persians went from the mid tier all the way to bottom 5? Monks are universal in top tier matchups if the enemy civ doesn't have monks.

  8. imagine making statistical tests and spreadsheets about a game made 20 years ago. That's some real dedication and a testament to the game's quality.

  9. I'm superlate but I also found very interesting how Franks have a 65% win rate in 2000+ elo games for 60+ min games (the highest win rate they get in 2000+ elo games), while in "all elo ratings games" only have a a 53% win rate in 60+ min games which is the lowest for them.

  10. This is insane. You brought me straight back to Energy Data Analysis – a master level statistics course for my business school!

Leave a Reply

Your email address will not be published. Required fields are marked *