calling all sabrematicians… calling all sabrematicians…
So, there’s a lot of talk about how (Major League) baseball games are taking longer and longer to finish. I’ve even, to my surprise, seen a columnist try to back it up with data about how average length of games for one team (the Red Sox) has increased-- it wasn’t completely convincing, since the years were random enough they could have been cherry-picked, but at least it’s enough data to provisionally believe it.
So my question is, what does the data show about the relative importance of batter/pitcher delays versus batters taking more pitches?
I mean, if it’s batter delays, then Red Sox games should have shortened since trading Nomar Garciaparra, but they haven’t. And the Red Sox have been open about having a strategy of trying to increase the pitch count of the opposing team, while the entire league generally has increased its appreciation of walks and choosy batters.
It’s pretty easy to look at: just get historical totals for length of games and # of pitches/game, calculate time per pitch and do some comparisons). My google-fu only returned an short article (mostly about MLB’s efforts to speed up games) that acknowledged that both # of pitches and time per pitch contribute to slowing games.
Using the Red Sox is tough, because they (and the Yankees) are featured so much on TV, and have done so more and more often in the last decade. This is a situation where you definitely need more data.
I read that column (Bill Simmons was the author). I think there is some evidence that games are getting longer, if only because most of them these days are on TV. It used to be that most games weren’t televised; now they all are, and you could even say that most of them are “featured” at least on their home networks.
On the other hand, Baseball-Reference.com, the leader in these kinds of stats, argues that part of the reason games are longer recently is because the average number of pitches in a game has increased, from 136 per team per game in 1988 to 147 so far this year. The theory is that teams are trying to work counts longer (which is “sabermetrically correct”, so to speak). Adding 11 pitches per team per game would increase the length of a game by 8% alone, and then when you factor in the additional number of relievers, commercial breaks sparked by that, etc., that could mean a significant time increase.
You might get some more ideas from a column written before the season.
It brought up interesting things about the three slowest teams–NY, Boston, Dodgers. They also are 1-2-3 in number of pitches their players “take” while at bat, and they are 1-2-4 in number of runners on base.
So, a more successful team is gonna take longer/inning while at bat. It might be that simple.
Here’s a previous thread that’s related, though it mostly people floating ideas to shorten games.
obfusciatrist estimates game time by elements (pitches thrown, side changes, etc.) in post #37.
That’s only correct if you assume the game is just pitching-according to obfusciatrist’s post, only about 3/8’s of game time are wrapped up in between pitches, so maybe 3-4% longer for each extra 20-25 pitches.
That’s true, but as the Baseball Reference article argues, the increase in pitches has also meant that there have been more pitching changes. Starting pitchers threw 95 pitches per game in 1988, 96 in 2010; there have also been more relievers per game. That’s going to slow things down.
It’s also - probably largely - just that there’s more hitting. Players get on base more than they did in 1988.
In the NL in 2009 the average team had 12.55 baserunners per game, and lost about 1 baserunner per game to double plays and getting caught stealing. In 1988, the average NL team had 11.5 baserunners per game, against losing about one baserunner to double plays and failed steal attempts.
More baserunners not only equal more batters, but slow the game down; pitchers work slower with men on and make pickoff attempts.
I would also guess that much of the increase in game time (which isn’t really as dramatic as people suppose, but there is some) is due to there being more pitcher changes. Teams have NEVER changed pitchers so often. Pitching changes are time consuming.
Good point that pitching changes should also be a big impact.
So has anyone gotten the data on # of pitching changes/game, # of at-bats per game, and # of pitches per at-bat (including pick-off attempts), and seen how they track game length? Does anyone know where this data is?
And finally, does anyone have any idea how to find data on whether commercial breaks are having any effect at slowing things down?