The Captain's Blog

“You can’t sit on a lead and run a few plays into the line and just kill the clock. You’ve got to throw the ball over the goddamn plate and give the other man his chance. That’s why baseball is the greatest game of them all.” – Earl Weaver

Young prospects aren’t the only ones being put to the test in the Arizona Fall League this year. Major League Baseball is also using the league to experiment with a series of measures designed to speed up the game. MLB has used Arizona as a laboratory for innovation in the past, but this time around, the proposed changes are intended to turn the clock back to a time when the “pace of the game” was deemed more fan friendly.

Average Length of Baseball Game, 1947-2014
Note: Includes shortened, nine inning and extra-inning games.
Source: Baseball-reference.com

The length of the average major league baseball game has increased over the last 70 years. In 1947 (the first year for which baseball-reference.com has nearly every game time), the typical game lasted two hours and 11 minutes, a far cry from the three hours and eight minutes needed to complete a contest this season. What’s the reason for this increase, and, is baseball’s desire to reverse this trend a case of trying to fix something that isn’t broken?

What probably concerns baseball suits the most are the last three years depicted in the chart above. Since 2011, the average length of a baseball game has increased by over 10 minutes, representing the first sustained breach of the three-hour mark in the period covered. What explains this sudden spike after years of gradual increases?

Replay Review Snapshot

Note: Official replay time is from retrosheet. Actual replay time is based on the time from when a manager enters the field until the next batter steps into the batter box.
Source: retrosheet.com and MLB.com video

Instant replay is an obvious culprit. However, based on retrosheet data, the typical review process added only 46 seconds to each game (1:46 per review multiplied by 1,275 reviews spread across 2,430 games). Of course, that only includes the amount of time used by the replay officials, not the events that occur before and after their deliberation. In a very small sample of three replay reviews (of various lengths occurring in Yankee games played in April, June and September), the real length of the delay was around two to three times longer than the reported total. Using the 2.4x multiplier average of those three events would yield a total delay of closer to one minute and 12 seconds extra per game. That’s not an extreme amount of time, but it does represent one-third of the year-over-year increase from 2013.

Relief Appearances per Game, 1947-2014
Note: Per game total includes both teams. Divergence in 1973 is likely attributable to the implementation of the DH. Data does not distinguish between pitching changes made at the beginning of an inning from those made mid-inning.
Source: fangraphs.com

Another recent trend impacting game length is the number of relief pitches used. Even though reliever innings have fluctuated over the last decade-plus, the number of pitchers used has maintained a steady accent. More pitchers means more pitching changes, and with them an increased number of in-game delays. Compared to 2011, this past season featured 563 additional relief appearances, which, when multiplied by the average of three minutes for a pitching change (again using a limited sample from the current season), equates to another per game increase of 41 seconds.

Replay and more relief pitchers have combined to add approximately two minutes per game, but that barely scratches the surface of the recent time increase. And, an increasing number of total batters faced, innings pitched and pitches thrown aren’t to blame either. In other words, most of the increase is probably related to non-systemic elements (i.e., delays, not breaks in action). That’s good news for MLB because it means it does not have to disturb the way the game is played in order to quicken the pace.

Batters Faced and Half Innings Pitched, Amount per Game and Average Time, 1947-2014
Source: fangraphs.com and baseball-reference.com

Pitches Thrown, Amount Per Game and Average Time, 2004-2014
Note: Data does not represent average time taken between pitches. Calculation is total pitches thrown divided total minutes played.
Source: fangraphs.com and baseball-reference.com

Simply enforcing the time limits between innings (2:05) and pitching changes (2:30) would significantly shorten the average game. Based on another anecdotal survey, enforcing these limits would shave one minute per half inning and a half minute per pitching change, or about 20 minutes per game (i.e., 17 half innings and average of six relief appearances per game in 2014). Minus this amount, the average game would shrink to approximately two hours and forty-eighty minutes, a length not seen since the mid-1980s.

In addition to limiting the breaks in action, MLB is also testing several other time saving initiatives. Some, like requiring batters to remain in the box and limiting the number of mound visits, are prudent, but others seem excessive. Eliminating the intentional walk is a gratuitous example. Not only wouldn’t the measure save much time, but it also unfairly penalizes the offense. As witnessed in the 2014 NLDS, successful intentional walks are not a foregone conclusion. What’s more, their deliberation can sometimes heighten the moment. Admittedly, the practice is mostly mundane, but considering the miniscule benefit, there’s little reason to eliminate it.

The proposed measure that could cause the most disruption is establishing a time clock between pitches (12 seconds when the bases are empty and 20 seconds when not). Unlike the other measures, which seek to eliminate dead time, this change would impact how the game is played on many levels. In particular, it would compromise the process of pitch selection, and distract players and fans from focusing on the pitcher/batter confrontation. This situation would be exacerbated when there are men on base. Upsetting the timing of a potential base stealer would be nearly impossible under the imposed restrictions, leading to either an excessive number of throws to first base (which would cause delays) or an increase in stolen base attempts. It’s debatable whether the latter would be desirable, but it is, nonetheless, a fundamental change.

Another problem with pitch time limits is ensuring consistent, accurate enforcement. Although every stadium would have a timer visible for all to see, protocols would be needed to ensure that the clock is reset properly. Other sports like basketball and football deal with this issue, but their clocks are longer and require fewer resets. In baseball, the pitch clock would need to be started about 300 times per game, often for a period of only 12 seconds. Also, the end of a baseball play is less obvious than triggering events in other sports. As a result, there’s likely to be more disputes about whether the clock was reset properly, which not only creates potential for unfair outcomes, but also introduces more delays.

Even if MLB could fairy enforce a pitch time limit, the penalty is another problem. Calling a ball when a pitcher takes too much time seems benign, but, what if the bases are loaded with a 3-2 count, a situation that requires the most concentration? Does MLB really want such an important moment decided by a clock? Whether it’s an automatic ball that plates a run, or a grand slam that results from a pitcher rushing to beat the timer, the feeling of dissatisfaction from such an outcome would easily offset the time saved.

Baseball doesn’t need to dramatically alter the game in order to shorten it. In fact, an argument could be made that there is no reason to shorten the game at all. After all, longer games mean more content for baseball to sell to its broadcast partners. Unless the length of games is forcing people to tune out, and local TV ratings suggest that’s not the case, there’s really no impetus to significantly reign them in. That’s all the more reason for MLB to take a more measured approach to shortening games. Although there’s nothing wrong with being proactive and eliminating wasted time, any attempt to undermine the natural flow of the game in deference to a stopwatch is misguided. The only people who would benefit from dramatically shorter games are those for whom watching a game is work. No wonder sportswriters and announcers are the most chronic complainers. For everyone else, there’s no need for baseball to hurry love.