When I started out to write this post, I planned on addressing the pace of the game. But what actually evolved was an analysis of game length. The objective became to look at the length of games over time and present the analysis in a fashion that can spark some discussion.

MLB has been concerned with the pace issue and has implemented rules to it. There have been some positive impact, however, any effect on the length seems to have been minor and temporary.

For now, let’s look at the average game length (of a regulation 9-inning game) over time. (Note that there is no game length data currently available for 1949.)

I have always felt that the optimal length for a baseball game was 2 to 2 1/2 hours. After constructing this chart, I began to wonder if that was just because I was a product of the environment at the time that I became a fan. In the late 60’s, it turns out that the average game time was about 2 1/2 hours.

What is striking to me is that from the first recorded game length data in 1920, the average length of a 9 inning game has increased by over an hour from 1:48 to 2:52. That represents an approximately 60% increase. Of course, a major factor affecting both the pace and length of a game is television commercial time but that is not the only cause. I hope to be able to have the patience to sit through a game or two before the end of the season and record the actual game time vs. the down time (e.g. commercial, warm-up, etc.). If I do, I will make it the subject of a subsequent post.

MLB has made some temporary inroads in shortening the length of the game. In 2003, there was a decrease of 6 minutes per game, however, since then the game has given back 5 of those 6 minutes.

Another interesting way to look at game length is examining the extremes (i.e. games less than 2 hours and greater than 3 hours) rather than the averages. This results in a chart that illustrates change more dramatically.
Besides the obvious trends, there is an anomaly in the 1999 – 2003 time frame. There was a large spike in games exceeding 3 hours, followed by an even larger “correction”. At the moment, I am at a loss to explain this. Given some time, I may try to understand this interesting divergence.

The point of this post was to take a look at the length of games over time, and see what questions that analysis may raise. The charts clearly illustrate the lengthening of the game over time and achieve that objective. Now here’s your chance to provide your thoughts on what this analysis shows. Does it make sense? What are the likely causes? What about the anomalies? What do you think is the optimal game length? Does MLB need to place a focus on game length as well as pace? Please take the time to share your thoughts.

Pardon this short commercial message. The information I summarized in this post was based on data from Retrosheet. They have generated databases including Boxscores for every major league game from 1920 through 2009 as well as a wealth of information and may be used at no cost. Retrosheet, was founded in 1989 for the purpose of computerizing play by play accounts of as many pre-1984 major league games as possible (data for games since 1984 have been collected by Project Scoresheet/Baseball Workshop and other organizations).