Hello folks! Applications for baseball analytics internship are now open. You'll get to tackle interesting baseball problems with data science and work with a terrific group of analysts:
@JoeSetyon
hi, baseball analyst here. having a strong data science/statistics/math background is really important for working in baseball R&D, but it's easier to teach someone baseball than data science, so teams hire for the former and teach the latter. there's really nothing wrong here
hi folks! the seattle mariners are looking to hire an analytics intern for 2023. you'll get to work with a bunch of really smart and talented analysts, and also me. if you have any questions about this role, feel free to DM me!
At 17 years old, Will Still decided to pursue coaching after immersing himself in the video game Football Manager.
Now, at 30, he's a real-life manager in the top division in France — and unbeaten in 12 games.
@DougGreenberg
on Still's unlikely story ⤵️
morning folks! happy to announce that the
#nflverse
has added a couple new fields to the load_participation() function in nflreadr.
for 2016-on, for our play by play data, we've included:
* NGS-tracked air yards
* Time to throw
* QB pressured flag
* WR route type
and...
Hello folks, applications for our baseball projects internship are now open! you'll get to work with coaches, staff, and analysts to build cool apps and reports that directly support our baseball operations department!
speaking of data viz -- incredible work by the capitals broadcast graphics team. the color scheme representing shootout results for the caps is white on red. for calgary it is... white on white?
hi folks -- i put together a blog about how to use {stacks} in {tidymodels} to boost model performance. it can feel so easy it's almost like hacking the kaggle leaderboards!
you will forgive me for doing an unironic post at 1 in the morning hopefully. I just have a couple things to say while I'm still champagne drunk, hopefully they make sense in the morning
Really like this! I compiled my own:
(Likely missing a couple of positions I applied for but don't have a record of in my email... but I think this illustrates how competitive entry level roles in sports can be)
🌍Predicting the Olympics Medal Table: Amelco are sponsoring this year's Prediction Competition ahead of the RSS International Conference in September. Open to the public, get involved here:
▶️Competition Link:
▶️Webinar Launch:
hi folks! our baseball systems team is looking to add a full-stack engineer. expected pay of $110k-$125k with some great benefits, and you get to work with baseball data!
hi folks! we are looking to hire interns for 2024. you'll get to work with some cool data on some fun problems with some great people, and hopefully win a few ballgames along the way. if you have any questions about either of these roles, feel free to DM me!
Hi folks, pleased to announce that we have a
@JuliaLanguage
version of nflreadr in development, called "NFLData.jl". You can read nflverse data into Julia as DataFrames. All of the load_* functions from nflreadr are already implemented, as well as caching
Seems the well has run dry on NFL participation data. We previously were able to scrape participation data from every game until the NFL limited that data to just primetime games. Unfortunately, as of last night, we're unable to obtain any participation data.
#nflverse
John Edwards grew up as a big baseball fan. Now, he's a quantitative analyst for the Seattle
@Mariners
— and his LMC degree helped him get there. See how in his alumni profile! ➡️ ⚾ 🥜
finished the notebook! ended up having to change the plotting package I was using because of some weirdness with text layering but still quite satisfied with with the results. please give it an upvote on kaggle if you liked it!
#BigDataBowl
#JuliaLang
okay! after a lot of iterating (thanks to the discord folks and especially
@nguyenank_
and
@rossdrucker9
!), feeling pretty good about the setup. next will need to come up with some intuitive coloring rules and actually write up the code
#BigDataBowl
okay! after a lot of iterating (thanks to the discord folks and especially
@nguyenank_
and
@rossdrucker9
!), feeling pretty good about the setup. next will need to come up with some intuitive coloring rules and actually write up the code
#BigDataBowl
hello our baseball projects team is hiring!
assistant:
intern:
i started out as a projects intern and found it a really rewarding experience with a great group of people. feel free to dm me with questions!
successful 5k in the books! team baseball ops raised over a thousand dollars for the Washington Coalition against Domestic Violence. Thanks to everyone who donated, can't wait for next year!
hi folks,
i'm participating the refuse to abuse 5k again this year and raising funds for the washington coalition against domestic violence! anything you can contribute helps support victims and take steps towards ending domestic violence:
Using generative AI technology, ESPN now offers game recap stories for
@PremierLacrosse
&
@NWSL
- incremental coverage for both leagues
This initiative has been collaboratively developed by ESPN & Disney teams with
@Accenture
More:
hey folks, {nflreadr} now has updated nfl injury report data for the 2021 season with nflreadr::load_injuries. this will be updated daily throughout the season. feel free to play around with it!
#nflverse
the phrase of "AI is threatening its ability to meet key environmental goals" is offloading the agency of Google onto the nebulous concept of "AI". It's not that AI is threatening the goals, Google is deliberately prioritizing AI development over our planet's long term health
New from me: Google's emissions shot up *almost 50%* since 2019 due to its massive focus on AI. This makes it the second huge tech company (see Microsoft in May) that has said AI is threatening its ability to meet key environmental goals. Who'll be next?
.
@SaberSeminar
is a wonderful place with wonderful people in it and I leave each conference feeling better about the state of our industry than I did coming in
super excited for this release, especially for the advent of full participation data! now you can see what players were on the field for what plays, and do things like calculate snap counts directly from nflreadr data!
a h/t to trashman in the NFLverse discord for flagging that the NGS added these fields, and of course, a huge shout-out to the NGS team for putting these fields together (we are merely standing on their shoulders at this point)
our player development internship is also live! come support one of the best farm systems in minor league baseball and work with our terrific coaches and staff!
We're a couple weeks away from the start of the CFB season which means it's time to throw some appreciation
@CFB_Data
's way for providing the most incredible repository of college football data anywhere
Thank you to everyone who stopped by my stream on Thursday! Despite the data issues... I was able to get out a submission for the contest!
Here's how I have the top 10 teams in the medal table shaking out!
I'm going to try to crash the party across the pond and put together a submission for this contest tomorrow! I'll be live-coding on twitch at 1 PM PST at , hope to see you all there!
𝗝𝗢𝗜𝗡 𝗨𝗦!
We’re looking for GT Students who are interesting in joining us as a manager or data analytics expert!
Email tgibson
@athletics
.gatech.edu for more information!
fun 2022 regular season in the books!
69% straight up (T-52nd overall on ESPN)
57% AT(ESPN)S (top 1000, admittedly not that impressive given ESPN's spreads are really bad)
#5
overall in
@LeeSharpeNFL
's prediction contest (1 of only 7 to beat the market!)
it's been a priv 🫡
We have no plans on removing the existing participation data, and we'll keep an eye out for anywhere else that it might pop up and in what forms. I'm extremely disappointed because there was a lot of awesome work done with the data, but the call is the NFL's, not ours.
#nflverse
I tried out this tool and wrote about my experience. It seems like a remarkably useless slop generator, perfect for excacerbating Kaggle's existing notebook spam problem:
Today we are launching Data Science Agent, a Google Labs experiment.
◆ Helps with complex tasks like planning, and error correction.
◆ Helps with data science tasks like predictive modeling.
◆ Outputs an AI-generated Colab notebook based on your prompt.
Try it at
I have noticed that although I have 300 views on my tweet, I am not receiving 300 likes. I'm not sure if this is being done intentionally or if these "followers" are forgetting to click 'like'. Either way, I've had enough. I have compiled a spreadsheet of individuals who have
thanks to everyone who stopped by my stream! i managed to embarass myself multiple times in front of strangers on the internet, this was the biggest messup:
my
#rstats
solution to day 7 of
#adventofcode
. part 1 super simple, look for the absolute difference between each possible position and calc fuel used for each. part 2, you can find the sum of all integers between two numbers w/ a pretty simple formula: