FanPost

Have the new rules shortened games?

Brad Barr-USA TODAY Sports

Executive Summary

Well, yeah, a bit.

★★★

Full Report

Preamble and Rationale

During this offseason, much attention was paid to the length of games, and how to shorten them. As such, MLB (under new commissioner Robert D. Manfred, Jr.) instituted some new rules, both at the minor- and major-league levels, with the goal of cutting down on those long, drag-ass games that we've all come to loathe.

The most visible change at the major-league level has been the addition of clocks, one beyond centerfield and the other behind the plate, which count down as soon as the third out is made; for non-nationally-televised games, this is set to two mintues, 25 seconds (it's two minutes, 45 seconds for nationally televised games, I believe). In addition, batters must keep at least one foot in the batters' box between pitches, lest they wander away for hours, chasing a bright, shiny object floating in the breeze.

I was curious to find out if these changes had any effect on the length of games, or if MLB was just in the back pocket of Big Clock.

(That's clock, folks.)

Methodology

The always-handy Baseball-Reference.com was my source for all data. What would we ever do without it?!

I thought it would be a good idea if I looked at all games played in the first two weeks of the past few seasons, and compared them to the first two weeks of this season. I chose that number because, when I started this little project, it was about mid-April, and I wanted to compare apples-to-apples; do games later in the season take longer? Do weather conditions have an effect on game length in any way? I wanted to scrub-out as many extraneous effects as possible.

For each game, I collected the total number of runs, hits and plate appearances by both teams put together, and the length of the game. In the end, I didn't end up using the runs and hits data, but it was, um, fun typing them in, I guess? This ended up being, as they say in the biz, a shit-ton of data (and typing), which is why it's early May when I'm putting this little report together.

I lumped all of the games for each year together, then sorted them out by league and by team. I wanted to know if there was a difference in the length of American League versus National League games -- for interleague games, I considered a game in an NL park to be an NL game, and vice-versa. In addition, I thought it would be fun to pick on the Yankees and the Red Sox, as they have a bit of a reputation for being excessively slow.

My original data is here. If you'd like to slice-and-dice the data in a different way, be my guest. Google Drive was a little cantankerous in terms of previewing it; you can just download it if you like.

Results

I like graphs, so here's one.

As you can see, yes indeed, game times are shorter... than last year. In the AL, games are back to the length they were in 2012-13. In the NL, they're back to 2011-12 lengths.

How about our favourite culprits, the Red Yanks?

They've been above the league average for every year except 2014, when they (and the Tigers) were curiously clustered right around the league average. Of note is this year's Tiger games, sneaking in just below the AL average; I'd probably chalk that up to the plethora of low-scoring games in which they've been involved so far.

The above graphs deal with the game lengths, in total; one could view this as the effect of the between-innings clock. But, does the foot-in-the-box rule have any effect? To answer that, I graphed the average time per plate appearance.

It follows the same general shape as the total time graph, which isn't terribly surprising, I suppose. The data are clustered much more tightly around the overall average, in comparison to total time; regardless of league, a plate appearance is pretty much a plate appearance, so it would seem. A caveat here is that between-innings breaks factor into the total length of the game, which was divided by the number of plate appearances; if those breaks are to be scrubbed out, one would have to listen to all the games, note the exact times the breaks started and ended, add them up, and subtract from the game length. Ain't nobody got time for that.

However, this part of the analysis tries to wash-out the variance you'd get in game length by having extra-inning or rain-shortened games, and it appears to do so. There might be an extra layer you could put on top of this, e.g. time per pitch (divide game length by number of total pitches thrown by both teams), but I felt this was far enough to satiate my curiosity.

How about the New Bostons, you ask?

Yeah, pretty much as I suspected... although, the consistency of the Tigers' plate-appearance lengths, in comparison to the league average, is kinda fun.

Obviously, in these team-based graphs, the effect of that particular team can't quite be teased-out of this data; for example, in "NYY" games, the Yankees have only roughly half the plate appearances, on average, and thus the effect is somewhat blunted. But, it gives one some food for thought.

Now, as I was churning through this data (1082 games in total, holy crap), I thought to myself, Out of all these games, which one had the shortest time per plate appearance, and the longest? I had my hunches...

Shortest: something started by Mark Buehrle, this year
Longest: some Yankees/Red Sox game somewhere, probably on ESPN, a few years ago

Turns out I was wrong.

Shortest by time-per-plate appearance: April 16, 2012, Phillies at Giants. Philadelphia won, 5-2, in a game that featured a then-marquee matchup of Roy Halladay and Tim Lincecum. Halladay did have a reputation for working quickly (although I don't really know if Lincecum has/had one as well). There were seven runs, 18 hits, four walks and 77 total batters, and the game was played in a crisp two hours and 23 minutes. This resulted in a brisk pace of 1:51.43 per batter, which was a full two seconds faster than its nearest competitor. It wasn't even a getaway day, either.

Longest by time-per-plate appearance: April 11, 2015, Tigers at Racist Nicknames. Detroit beat Cleveland 9-6; the starters, David Price and Corey Kluber, were (and are) both their respective teams' aces, but only Kluber pitched into the seventh inning. The main culprit for the honkin' huge length of this game, I imagine, is the number of pitchers used for each team (five for Detroit, seven for Cleveland); the last four innings saw a combined 13 runs scored (and the game was close the whole time). There were 15 runs, 22 hits, 11 walks and 85 total batters, in 4:16. Total time per plate appearance, even with all the new fancy rules, was a whopping 3:00.71, a second slower than the next-slowest game.

Conclusions

The rules are having the effect of skooching back a couple of years, in terms of game lengths. But, will this change stick, or will times creep back upwards? Given the carpal tunnel syndrome this little project has produced in my wrists from all the clicking and typing, someone else can answer that a few years from now.

This is a FanPost and does not necessarily reflect the views of the <em>Bless You Boys</em> writing staff.