PySport supports the Sports Analytics community, mainly with open-source and sharing knowledge.
Joined July 2012
- Tweets994
- Following230
- Followers7.5K
- Likes318
PySport retweeted
La Analytics Cup 2.0, de @SkillCorner y @PySportOrg, incluye por primera vez una rama de baloncesto. El dataset público trae:
· Tracking XY de 10 partidos de la Liga Endesa 25/26
· Eventos derivados del tracking: bloqueos directos, penetraciones, closeouts, distancia del defensor en cada tiro…
· Agregados de tiros, penetraciones y bloqueos de 293 partidos de la temporada
PySport retweeted
Lo más chulo en lo que hemos podido trabajar. Mucho por hacer, pero mola muchísimo como primer toque de la idea
Agradecer a @SkillCorner por la disposición a los datos y a @PySportOrg por organizar algo tan guay
Utilizando el Open Data de @SkillCorner para visualizar cuál es el impacto de superar o no el show a la probabilidad de encontrar un tiro de calidad.
Dejamos un ejemplo de cada situación en este tweet.
Interesantísimo el material para preparar la #AnalyticsCup con @PySportOrg.
PySport retweeted
Utilizando el Open Data de @SkillCorner para visualizar cuál es el impacto de superar o no el show a la probabilidad de encontrar un tiro de calidad.
Dejamos un ejemplo de cada situación en este tweet.
Interesantísimo el material para preparar la #AnalyticsCup con @PySportOrg.
PySport retweeted
𝐀𝐧𝐚𝐥𝐲𝐭𝐢𝐜𝐬 𝐂𝐮𝐩 𝟐.𝟎 𝐢𝐬 𝐡𝐞𝐫𝐞. 🇪🇺🇺🇸
Following the success of the first edition, @PySportOrg and @SkillCorner are inviting the analytics community to work with open-source sports data.
PySport retweeted
awesome analysis by luke, using basketball data from skillcorner. analytics cup is open and covers basketball and football - go try stuff, have fun!
Used @skillcorner's open ACB tracking to ask one question: can you measure "advantage" as a number, every 0.2 s?
Here's a drive collapsing the defense. The bright space is court the offense reaches first and would score from. Watch where it goes.
Write-up + interactive replay👇
PySport retweeted
Super exciting pose data available 🤩🤩
𝐀𝐧𝐚𝐥𝐲𝐭𝐢𝐜𝐬 𝐂𝐮𝐩 𝟐.𝟎 𝐢𝐬 𝐡𝐞𝐫𝐞. 🇪🇺🇺🇸
Following the success of the first edition, @PySportOrg and @SkillCorner are inviting the analytics community to work with open-source sports data.
PySport retweeted
🚀 𝗕𝗶𝗴 𝗻𝗲𝘄𝘀 𝗳𝗼𝗿 𝘁𝗵𝗲 𝗳𝗼𝗼𝘁𝗯𝗮𝗹𝗹 𝗮𝗻𝗮𝗹𝘆𝘁𝗶𝗰𝘀 𝗰𝗼𝗺𝗺𝘂𝗻𝗶𝘁𝘆!
We at @impect_official are thrilled to announce the release of our 𝗼𝗽𝗲𝗻 𝗱𝗮𝘁𝗮 – the 𝗺𝗼𝘀𝘁 𝗰𝗼𝗻𝘁𝗲𝘅𝘁𝘂𝗮𝗹𝗶𝘇𝗲𝗱 𝗲𝘃𝗲𝗻𝘁 𝗱𝗮𝘁𝗮𝘀𝗲𝘁 𝗶𝗻 𝗳𝗼𝗼𝘁𝗯𝗮𝗹𝗹, covering the 𝟮𝟬𝟮𝟯/𝟮𝟰 𝗕𝘂𝗻𝗱𝗲𝘀𝗹𝗶𝗴𝗮 𝘀𝗲𝗮𝘀𝗼𝗻, including Bayer 04 Leverkusen’s historic invincible campaign.
Together with our friends at @PySportOrg, this dataset is now available through 𝗞𝗹𝗼𝗽𝗽𝘆 and our 𝗱𝗲𝗱𝗶𝗰𝗮𝘁𝗲𝗱 𝗚𝗶𝘁𝗛𝘂𝗯 𝗿𝗲𝗽𝗼𝘀𝗶𝘁𝗼𝗿𝘆.
This release marks an exciting step forward for the open-source football analytics ecosystem. We can’t wait to see what the community creates with this data!
A massive thank you to @unravelsports, @mr_le_fox, Koen de Raad, Dries Deprest and the entire @PySportOrg team for their collaboration and for making the Kloppy integration possible.
🚧 Let’s build together!
(Links in the threads 👇)
PySport retweeted
@SkillCorner data will remain open and accessible outside of the Analytics Cup via a dedicated GitHub repository.
𝐎𝐩𝐞𝐧 𝐃𝐚𝐭𝐚 github.com/SkillCorner/opend…
PySport retweeted
Last week we announced the first ever 𝐒𝐤𝐢𝐥𝐥𝐂𝐨𝐫𝐧𝐞𝐫 𝐗 𝐏𝐲𝐒𝐩𝐨𝐫𝐭 𝐀𝐧𝐚𝐥𝐲𝐭𝐢𝐜𝐬 𝐂𝐮𝐩.
If you have not registered yet, do so now to not miss out on a chance to showcase your work at the live event in Paris!
𝐒𝐢𝐠𝐧 𝐔𝐩 pysport.org/analytics-cup
PySport retweeted
𝐀𝐧𝐚𝐥𝐲𝐭𝐢𝐜𝐬 𝐂𝐮𝐩 𝐀𝐧𝐧𝐨𝐮𝐧𝐜𝐞𝐦𝐞𝐧𝐭
Due to the overwhelming interest we have made the decision to close registrations on 𝐖𝐞𝐝𝐧𝐞𝐬𝐝𝐚𝐲 𝟐𝟐𝐧𝐝 𝐎𝐜𝐭𝐨𝐛𝐞𝐫 (𝐌𝐢𝐝𝐧𝐢𝐠𝐡𝐭 𝐂𝐄𝐓).
𝐒𝐢𝐠𝐧 𝐔𝐩 pysport.org/analytics-cup
PySport retweeted
Last week we announced the first ever 𝐒𝐤𝐢𝐥𝐥𝐂𝐨𝐫𝐧𝐞𝐫 𝐗 @PySportOrg 𝐀𝐧𝐚𝐥𝐲𝐭𝐢𝐜𝐬 𝐂𝐮𝐩.
Due to the overwhelming interest we have made the decision to close registrations on 𝐖𝐞𝐝𝐧𝐞𝐬𝐝𝐚𝐲 𝟐𝟐𝐧𝐝 𝐎𝐜𝐭𝐨𝐛𝐞𝐫 (𝐌𝐢𝐝𝐧𝐢𝐠𝐡𝐭 𝐂𝐄𝐓). If you have not registered yet, do so now to not miss out on a chance to showcase your work at the live event in Paris!
𝐒𝐢𝐠𝐧 𝐔𝐩 pysport.org/analytics-cup
SkillCorner data will remain open and accessible outside of the Analytics Cup via a dedicated GitHub repository.
𝐎𝐩𝐞𝐧 𝐃𝐚𝐭𝐚 github.com/SkillCorner/opend…
PySport retweeted
🚨🚨 MAJOR UPDATE RELEASED 🚨🚨
@nflseedR #rstats package 📦 v2 is now live on CRAN.
Updates include a new generation of
-- high efficient code to calculate NFL standings and apply tiebreakers (almost max depth),
-- high efficient code to simulate tens of thousands NFL seasons.
.@nflseedR 2.0 coming soon
ALT Travis Kelce is yelling "Coach, they did it. nflseedR 2.0 is ready" at Andy Reid
PySport retweeted
Continuing the tradition, I have compiled an end-of-year overview of the soccer analytics content that I liked the most in the past year. The 2024 edition includes pointers to over 160 resources, including research papers, blog posts, articles, and books.
janvanhaaren.be/posts/soccer…
PySport retweeted
Made some progress in the nfl season simulator efficiency redesign.
Out of curiosity: if you simulate the remainder of the 2024 regular season (incl. calculation of playoff seeds) 10k times. How long does it take?
Let me know in the comments 👇
Last year, I started replacing the NFL standings calculation in @nflseedR with high efficient code in data.table.
The implementation of tiebreakers has caused a few headaches and my plan was actually to write the season simulation in data.table as well, but the development came to a stagnation.
That's why I decided to merge the standings code into the nflseedR dev branch today.
It is designed to manage thousands of seasons with one call. The performance boost is brutal!
Calculating standings (incl. all tiebreakers) of 10k seasons used to take almost 3 minutes on my laptop. Now, it's less than 30 seconds!
data.table rewrite of the season simulation has no eta, but it will come.
Check out the new standings function here
nflseedr.com/reference/nfl_s…
PySport retweeted
Do you want to work on a soccer analytics project and present in on stage in front of industry professionals and academic experts?
In partnership with @SkillCorner, we are excited to put out a call for proposals! Get yours in by 11:59pm CT on November 4.
americansoccerinsights.com/c…
PySport retweeted
Asked @simonw what the biggest productivity boost has been, throughout his 20-year software engineer career. He answered, without hesitation:
Open source.
(Clip from @Pragmatic_Eng Podcast from the episode "AI tools for software engineers, but without the hype")
PySport retweeted
First use of mplbasketball.... incredible!
Had all the code to get the data already, just a matter of putting it onto the court. So simple to use
PySport retweeted
Excited to introduce 🏀𝐦𝐩𝐥𝐛𝐚𝐬𝐤𝐞𝐭𝐛𝐚𝐥𝐥🏀: SPL's Python plotting library for basketball data!
Bring your 2D and 3D visualizations to life with fully customizable courts across the NBA, WNBA, and NCAA, along with complete integration with matplotlib.
