top of page
  • Writer's pictureEan Maloney

Home Runs in the 2020 MLB Season

Updated: Aug 4, 2021

Should baseball analysts really care about the 2020? Honestly it seems bound to be an outlier since the "COVID season" was shortened from 162 games to just 60 games for each team. (And there were actually two teams--Detroit and St. Louis--that only played 58 games.) Considering that baseball stats are generally derived from a sample that's relatively large for sports data, and considering that in this case we're cutting that sample by more than half, is it even fair to treat the data from the 2020 season as being on the same level as that from normal season? Perhaps not, but it's interesting data to play around with nevertheless, and since we can't redo the 2020 season, it's as good as we're gonna get.

All the data I used for this analysis was from Retrosheet's Game Log dataset. I loaded the data into a Pandas DataFrame in Jupyter Notebook, did some data transformations and manipulations with Pandas, and created charts using a couple different versions of Plotly. Anyway let's look at some charts...

My first idea was to group games based on the date they were played, so I totaled the number of home runs and the number of games played on each date. I'm ignoring the fact that these groups are of different sizes, and I simply divided the number of home runs by the number of games played for each data to get the following chart.

What this chart shows is, well...basically a mess. It's not easy to see any kind of trend in this data. However, we do see that on any given day, the average number of HR per game was between 1.5 and 5. (In case you're wondering what the aberration was on Aug. 13, there were only 6 games played that day and Mookie Betts hit 3 homers himself in one of them.)

My next idea was to look at the total number of HR vs. the total number of games over the course of the season. Note: the data points that this line chart is connecting are still grouped by date, so each point is the number of games played up to and including that date vs. the number of home runs hit up to and including that date.

As we can see, the result is basically a straight line. So it looks like there was no major variation in how many home runs were hit over the course of the season. It's hard to think after looking at this graph that HR hitting simply increases or decreases in one direction over the season, but I'll dig deeper into what this chart shows in my next post...


7 views0 comments

Recent Posts

See All


bottom of page