Okay, so today I wanted to dig into the Milwaukee Brewers vs. Mets game and really look at the player stats. I’m a huge baseball fan, and I love getting into the numbers.

Getting Started
First, I needed to find a good source for the data. I usually just do some looking around for sports data.
Gathering the Data
Once I find a good source, I started pulling the info I wanted. It’s kind of a manual process, to be honest. I looked for things like:
- Batting: At-bats, hits, runs, RBIs, strikeouts, etc.
- Pitching: Innings pitched, hits allowed, runs allowed, earned runs, strikeouts, walks.
- Maybe some fielding stats if anything interesting happened.
I basically just went through each player that played in the game and jotted down their key stats in a simple text, copying and pasting what statistics are shown.
Organizing My Notes
After getting all the raw stats, I started organizing them. I made simple tables, nothing fancy. One for the Brewers, one for the Mets. I listed the players and their stats in rows and columns.
for example:

player:xxx batting:xxx pitching:xxx
Looking for Standouts
With everything organized, I started looking for any players who had a particularly good or bad game. You know, like who had a ton of hits, who struck out a lot, which pitcher had a great outing, or who gave up a bunch of runs.
My Final Thoughts
It’s a pretty straightforward process, but it’s fun for me. By putting all these numbers into a simple table, I got a much better feel for how each player performed and how the game played out. I am happy to put the data together and make a comparison, next time, I will probably try other ways to make them more organized.