Starting Pitchers Get Pulled After 2 Walks.

Dr. Nel
Data Sciency Things
2 min readMar 17, 2016
Walks vs Times Substituted.

Baseball is a sport so rich in data and statistics, it should be called ‘The Disney World of Data Scientist’.

I’ve been wanting to do a fun project involving Baseball for a while and I was browsing through Kaggle.com the other day and found some interesting questions, along with a few data sets. My plan for this post is to walk you through my analysis.

How many walks does it take for a starting pitcher to get pulled?

The data set I have contains information about a pitchers’ performance for the entirety of the season. To find the starting pitchers out of the list of all pitchers, I looked at the amount of games started vs the amount of games played and only selected those who had pitch in at least 99% of the games. This set is small relative to the original one but as it turns out it still a good amount of data to do some analysis.

The data set didn’t contain information about how many times the pitcher was pulled in the season, so to get around that I looked at the total number of outs divided by 3 to obtain the innings played, then divided the innings by games played to obtain an average of innings. This allowed me to get rid of those few cases with phenomenal pitching.

I also divided the number of Walks (BB) by the number of games to find the average number of walks per game, and the proceeded to count to obtain the following:

We can see that starting pitchers tend to get pulled after the second BB.

You can review and vote for my Kaggle solution at https://www.kaggle.com/nelabdiel/d/kaggle/the-history-of-baseball/starting-pitchers-get-pulled-after-2-bb

Feel free to also check out the code on Github https://github.com/nelabdiel/walkitout

If you have any ideas for web apps, Data Science projects or some cool sports related analysis that you’d like to see, let me know and I will look into it!

Cheers,

El Nel

Like this post? Press the heart shape button just below to recommend it and/or share via social media. Feel free to highlight any part of it you like and/or leave me comments on it.

--

--

Dr. Nel
Data Sciency Things

AI Scientist / Quantum Topologist bouncing around DC, NYC and Miami