Big Data Baseball with Python - Ichiro Suzuki hacks! #kwsk01
3 likes2,721 views
Ichiro Suzuki reached his 3,000th MLB hit on August 7, 2016. The author analyzes Ichiro's journey to 3,000 hits using Python and MLB play-by-play and pitch tracking data from 2015-2016. He extracts Ichiro's at-bat and pitch-level data to analyze trends in Ichiro's pitch selection and performance over the two seasons leading up to the milestone. The author plans to present further analysis and visualizations of Ichiro's path to 3,000 hits at an upcoming Python conference.
Big Data Baseball with Python - Ichiro Suzuki hacks! #kwsk01
- 1. Ichiro Suzuki - The journey to 3,000 hits Big Data Baseball With Python( Ruby 01 2016/8/20) Shinichi Nakagawa@shinyorke
- 2. Starting Member • Who am I ? • Hack ( ) • MLB Python Hack Ichiro Suzuki - The journey to 3,000 hits • &
- 3. Who am I? • Shinichi Nakagawa(@shinyorke) • Pythonista/Agile & Scrum & Kansan coach/Baseball Analyst • visasQ( ) Python Engineer • /Oakland Athletics/RADIOHEAD ※ (21) & Thom Yorke • Python . • Hack .
- 4. Hack? • PyConJP 2015 Python Hack • ( ) Python OSS • ” ” Web ( • • Python & ( Ruby )
- 5. Hack! Python PyCon JP 2015 http://www.slideshare.net/shinyorke/hackpython-pyconjp
- 6. Hack! Python PyCon JP 2015 http://www.slideshare.net/shinyorke/hackpython-pyconjp
- 7. Hack! Python PyCon JP 2015 http://www.slideshare.net/shinyorke/hackpython-pyconjp
- 8. Hack!→Big Data Baseball • etc… • • ( ) ( , etc…) • … •
- 9. MLB at BAT MLB • MLB • PC Apple TV • MLB.TV •
- 11. Ichiro Suzuki records 3000th MLB Hits!!! 3,000 (8/20) 3,007 ( )
- 12. Ichiro Suzuki - The journey to 3,000 hits ( Ichiro Suzuki hack) • 8/7 ( 3,000 ) Ichiro . • & . • 3,000 !?
- 13. • Python pitchpx 2015-2016 & Get • & Ichiro Suzuki • Jupyter + matplotlib
- 14. 2015-2016 ※ ( ) $ pip install pitchpx # MLB Dataset library(author: @shinyorke) $ pitchpx -s 20150401 -e 20151005 ./output/ # 2015 Season $ pitchpx -s 20150401 -e 20151005 ./output/ # 2016 Season $ ls -l -rw-r--r-- 1 shinyorke_mbp staff 363988 8 20 14:40 mlbam_atbat_20150401.csv -rw-r--r-- 1 shinyorke_mbp staff 357494 8 6 14:31 mlbam_atbat_20150402.csv ( ) -rw-r--r-- 1 shinyorke_mbp staff 3429 8 17 20:55 mlbam_umpire_20150912.csv -rw-r--r-- 1 shinyorke_mbp staff 3241 8 6 14:31 mlbam_umpire_20160723.csv
- 15. Ichiro & • pitchpx CSV At Bat Pitch f/x At Bat: Pitch f/x: • At Bat Pitch f/x • pandas . ( )
- 16. (At Bat) 30 50MB (Pitch f/x) 200 300MB
- 17. Ichiro Suzuki $ # $ ls -lah ichiro_suzuki*.csv -rw-r--r--@ 1 shinyorke staff 140K 8 18 01:42 ichiro_suzuki_2015_atbats.csv -rw-r--r--@ 1 shinyorke staff 687K 8 18 01:42 ichiro_suzuki_2015_pitchfx.csv -rw-r--r-- 1 shinyorke staff 74K 8 14 17:06 ichiro_suzuki_2016_atbats.csv -rw-r--r-- 1 shinyorke staff 402K 8 14 17:14 ichiro_suzuki_2016_pitchfx.csv $ wc ichiro_suzuki*.csv 442 7641 143176 ichiro_suzuki_2015_atbats.csv 1636 10368 703882 ichiro_suzuki_2015_pitchfx.csv 233 4026 75543 ichiro_suzuki_2016_atbats.csv 955 6002 411526 ichiro_suzuki_2016_pitchfx.csv 3266 28037 1334127 total
- 18. ?
- 19. ( ) ( )
- 20. ( ) Jupyter
- 21. 2015/2016 …
- 23. . (CU) (SL) ??? 2015 2016
- 24. : ( ) : : : 20: 21: 22: 23: 2015 2016
- 25. : ( ) : : : 20: 21: 22: 23: 2015 2016
- 26. • • • ( ) • • ( ) • ( ) • 3,000 • Pandas
- 27. PyCon JP 2016 9/22( )12:00 ? Python Hack