FG on Fox: MLB’s Runs Per Hour Problem
New commissioner Rob Manfred is clearly not afraid of change. In his first 24 hours on the job, he postulated about a future of the sport that included pitch clocks but excluded defensive shifts, among other tweaks. Unlike the former commissioner — who famously hated computers — Manfred is a proponent of technology and wants to make sure baseball keeps itself relevant in a changing landscape of how fans consume sports and entertainment. In this day and age of screens everywhere, shorter is often better, and the commissioner seems serious about addressing the pace of play issue in Major League Baseball.
However, his comments about the shift came in the context of a stated desire to breathe some offense back into a sport that has shifted heavily towards the pitching side of the equation in recent years. With offense trending downwards, the league clearly feels there is a point at which rules may have to be adjusted to restore the balance between offense and defense, just as the league took action in 1969 (by lowering the mound) and again in 1973 (by introducing the Designated Hitter). While I’m among those who do not believe that restricting the shift would have much of an effect on increasing offense, the willingness to consider it as a remedy suggests that Manfred believes that current offensive levels are a potential problem for the sport.
So on the one hand, the league would like to speed up the games; on the other hand, the league would like the games to include more run scoring. This seems to be a bit of a paradox, given that the act of scoring runs inherently means that more time is spent doing things besides ticking off some of the 54 outs — or 51 outs, if the home team protects a ninth inning lead — allotted for each contest. More offense means more at-bats and often can mean more pitching changes, and those two things generally mean longer games.
But how closely does run scoring track with length of game? Is it such a clear relationship that any increase in offense would be immediately met with a corresponding uptick in the number of minutes in a contest? I wasn’t actually sure, so with the help of some data from our friends at Baseball Prospectus and Baseball-Reference, I lined up the average length of game with average team runs per game for each season since 1950. The results are displayed in the chart below.
Read the rest on Just a Bit Outside.
Dave is the Managing Editor of FanGraphs.
Robo umps and a smaller strike zone. Tweak strike zone (annually?) as required.
Why wouldn’t a smaller strike zone just lead to more walks? If there’s anything more boring than K after K, it’s BB after BB.
Because pitches aren’t trying to walk people, they’re trying to get guys out. They will be forced to throw the ball where it can be hit with authority.
Yeah, over the last five years the bottom has fallen out of the strike zone and the game has gotten longer and the scoring has gone down.
Strikeouts take lots of pitches. Groundball pitchers give faster games. (And I’m told are more democratic.)
How are extra-inning games impacting these results? In a higher run-scoring environment, you might expect to have fewer and shorter extra-inning games. Are we for some reason playing more extra inning games now than might be expected, and might these be inflating the total game time?
I wonder what the data would look like if we removed the extra-inning games and just looked at the expected time to watch 9 innings.
I really like how this is not the outcome Dave expected (at least as he said on the podcast). Commitment to data!
How does this factor when you remove the Padres from last years data? Watching the Padres was like watching paint dry.
http://blogs-images.forbes.com/maurybrown/files/2014/09/GraphAvgTimeMLB2014.jpg
The padres played the second fastest game son average. This is why runs per hour should be more important to look at than just game time. No one cares what the actual game length is if they “feel” like… as you put it… “watching paint dry”.
Let’s do something extreme with the fences…bring them all in 20 feet and simulate the Steroid Era OR push them all out 20 feet and we can bring back the go-go 80’s, Vince Coleman style
Easy solution to increase the offense- Aluminum bats.
That could greatly increase offense because no pitchers would be left alive to be any good.
I feel like a pitch clock would help out some, but keep in mind that the pitch clock is only on (or should only be on) when there is nobody on base. The pitcher has to have the ability to hold the runner and should have as much time as necessary to do so. When analyzing the possible effect of pitch clocks, are situations with base runners included? If so, then the effects of a pitch clock may be less drastic than originally thought.
Also, wouldn’t a decrease in game time naturally lead to an increase in runs per hour? Making games faster would reduce the denominator; a game where 5 runs are scored in 3 hours has a rph of 1.67, yet the same number of runs in 2.83 hours (a 10 minute reduction) has an rph of 1.76. While an increase in run scoring would be fine, I still feel that more focus should be placed on quickening the pace of play.
Baseball games are NOT too long, and plenty of runs are being scored. The recent downward trend in scoring already represents restored balance between offense and defense. Leave it alone.
And its exactly opinions like this that means we shouldn’t freakout and make some overly drastic changes like pitch clock.
I’m of the opinion that some games get just completely bogged down in the mud by late inning pitching changes and S-L-O-W relievers. So mild rule changes to speed that up would be nice.
I also think this new low strike is an overall bad thing for the game and it happened too fast for hitters to adequately adapt to it. Remember it takes an easy decade of training to determine a pitch’s location and the need to swing or not in just a small fraction of second, but this low strike call basically happened in 2-3 years. If it was a more gradual evolution over 10 years or more, it probably wouldn’t have impacted hitting as much, because the hitters would have had time to adapt to count it.
I feel like the best complimentary data to this analysis would be that of outs recorded by starting pitchers and relief pitchers over the same time frame. Also, worth researching would be the number of in-inning pitching changes year to year. The easiest way to add length to games is to add time where it’s not actually even being played. I would expect you’d find that starters go fewer innings than they did previously (since we already know this) and that reliever usage is also down, per appearance, particularly over the last five years. Should be an easily testable hypothesis.
Stop letting hitters out of the batters box after each pitch. and make the pitchers pitch. Its that simple. Big Papi adjusting his gloves after every at bat is ridiculous. In the old days batters would bat and pitchers would pitch without all the theatrics they use today.
Sorry. Every pitch not at bat for Big Papi
How about every pitcher must pitch two complete at bats unless injured or unless the first at bat ends the inning. That would easily save a few minutes in any close game and might stimulate offense. I know this would spell the end of the ROOGY/LOOGY, but I just don’t see that as changing the nature of the game.
I like just one pitching change an inning period.
I’d be interested to see numbers on how pitching changes are effecting length of game. In recent years, it seems the number of pitching changes has gone up with every team boasting several flame throwing specialists and (rightly) tending to pull starters earlier and earlier, decreasing offense when it’s most dramatic to fans, while increasing dramatically the time per inning.
If this is really what’s happening (and not just the way it seems), perhaps a simple fix would just be adjusting the rules for relief pitcher usage, like every reliever must face at least 3+ batters (barring injury) before being pulled. Or more extreme, the reliever who starts an inning must finish it barring an injury or some pitch number threshold.
The autocorrelation function of runs scored by year in MLB shows a stationary time series. The runs per game may seem a little down from the norm but there doesn’t appear to be any statistical difference in the “downward trend” that is the issue right now.