If JxmyHighRoller Doesn’t Make This Video, Then I Will…

119,405
0
Published 2024-04-16
If JxmyHighroller doesn’t make this video, then I will. In today’s video I present the most compelling statistics on what makes an NBA championship team. From expanding on theJK’s work I have managed to craft a model that has managed to predict the 8 out of the last 10 NBA champions correctly, so lets see if we can get this right this time. Shoutout to JxmyHighroller for the inspiration.

Article:
allenjake440.medium.com/2024-nba-champion-machine-…

Socials:
Instagram: www.instagram.com/_jake.allen_/
TikTok: www.tiktok.com/@allenjake440
Medium: allenjake440.medium.com/

Timecodes:
00:00 – Intro
02:25 - Champion Rules
04:51 - True Contenders
06:41 - "Buckets"
07:11- Prediction

*All the videos, songs, images, and graphics used in the video belong to their respective owners and I or this channel does not claim any right over them.

*Copyright Disclaimer under section 107 of the Copyright Act of 1976, allowance is made for “fair use” for purposes such as criticism, comment, news reporting, teaching, scholarship, education, and research. Fair use is a use permitted by copyright statute that might otherwise be infringing.

All Comments (21)
  • @lemonjames6585
    You would think that with this quality, this channel would have over 100k subs, great job!
  • @Aquaneos1193
    I completely agree, it's going to be Denver vs Boston and I favor Denver
  • @satangcorgi
    this is crazy quality for a video. im an aspiring data scientist and you might just have inspired my last run to finish this program predicting the 2025 nba champions, insane content. biggest props 🐐
  • @bboyggez
    For the past decade, the NBA champion was more or less clear-cut. But this year, I have no freaking idea
  • @ThatOneDude7
    The biggest thing i noticed watching the 2023 playoffs that seemingly nobody else did, the Nuggets don't make mistakes. They don't have bad games. Every other team had bad games, notably the Celtics and the 76ers, but the Nuggets didn't have a single one. There is efficiency, and then there's the Nuggets machine efficiency.
  • @ededman3656
    love love the outro music bro great work!!!!
  • @whenurefree
    You should try PCA for dimension reduction and LASSO as good secondary model after you reduce and select your dimensions with PCA. Ridge is awesome for highly correlated data, as I assume the dataset used would exhibit high correlation. Probably why Ridge performed the best. I think that combination might bring down your MSE quite a bit, but that is total speculation as it would be cool to see what you come up with, if you try it! Also, you NEED to (no you don’t) throw your MSE’s in next time! I’d be curious of all the model parameters and performance.
  • @mauriratm6145
    Loved the video and quality of reasearch! one small complain maybe i am not the only one, but the voice recording sounds with excesive low frequencies making it hard to listen at times, feels like deacreasing those a bit or raising the highs would be much clearer. Hope you can consider it and if im correct fix it cause im subscribing and waiting for more videos. Cheers!
  • Fire 🔥 keep it up! Super high quality video and you’ll get lots of subscribers one day. I clicked because I love jxmyhxrollers video
  • Making a Machine Learning model for this is toughhhhh, and ik from being a computer engineering major. well done sir!
  • @PerspectiveLove
    it will be the Nuggets. Every Eventually NBA champion has a series in which it goes to 6 or 7 games of adversity that they overcome and use that momentum to win the chip.
  • @colefairfield
    Great video! If you improve mic quality im sure that this channel will blow up
  • @eliasmi6495
    wow what a great video pls more with this statistics about the brackets
  • @frostbite7959
    I’m subscribing. This high quality video with this low amount of subscribers is crazy. Great video keeps it up. …… Go OKC!
  • I read your article, as well as your article on how you fed the algorithms and AI with data (and which sets of data were particularily predictive). Maybe (hopefully) in the offseason you can run a series with the following: - use the same process for each decade of nba championships - for the years that the data DID NOT predict the winner (or get the final 4 teams particularity correct) you can examine why those seasons/teams beat the data. - the human story that could be captivating is how a team can beat the data/norm. - this is how you can beat that jimmy guy who you thinks makes better utube videos than you! Thanks! This was cool!
  • @jurismuris
    Very nice job, mate! Trully great content
  • @JakeyFoxy_Z
    Great video with great editing, I thought you had 100k!
  • @richteffekt
    Great stuff; I subbed. Can I ask what the idea was to run the bookies' cut before applying the last performance based cut?
  • @rocimi7321
    That was quality content I fucking love numbers when it comes to basketball good stuff brother ☝