The Slowest(-Working) Team in Baseball History
The Mariners and Rays played a Wednesday matinee that featured extraordinarily little in the way of offense. Following the conclusion, there was this simple throwaway tweet:
#Mariners fall to Rays 2-0. Tampa had 3 hits, Mariners 2. No idea why this one took 3 hours, 21 minutes.
— Greg Johns (@GregJohnsMLB) May 14, 2014
Seems long. Seems like too much. An isolated instance is an isolated instance, and you shouldn’t focus too much on anything unless it repeats, but it turns out, for the Rays, this is a reflection of the norm.
Granted, there was a variety of reasons for Wednesday’s duration. Mariners starter Brandon Maurer struggled, and Mariners manager Lloyd McClendon went and got himself into an on-field argument. The day before, the Rays won 2-1, and that game took less than three hours. But a few days ago, the Rays lost a nine-inning affair 6-5, and they played for 228 minutes. Not long before that, they lost a nine-inning affair 6-3, and they played for 250minutes. In terms of footspeed, the 2014 Rays presumably are not the slowest team in baseball history. In terms of game pace, they most probably are.
Now, this is going to require some assumptions on our part, since we have very incomplete data. The main assumption is that baseball now is slower-moving than ever. Over the course of the PITCHf/x era, we’ve seen Pace increase by more than a second, which is of interest. And, the team with the fastest average game this year is Kansas City, with an average duration of two hours and 58 minutes. In 1986, the team with the slowest average game was Cleveland, with an average duration of two hours and 56 minutes. Baseball’s taking a long time, and some of that has to do with television broadcasts, but some of it doesn’t. Let’s say that baseball has slowed down.
If we can accept that, then we can arrive at a sweeping conclusion. It’s time to play around with Pace, the FanGraphs measure of average time in between pitches, based on PITCHf/x. We have Pace data for batters, we have it for starters, and we have it for relievers. We can also put it all together, with a little simple math.
This year, the slowest batting pace belongs to the Dodgers, at 24.1 seconds. We find the Rays around the bottom of the upper third, at 23.2. The Rays’ starting rotation, however, has the slowest pace in baseball, by more than two seconds. And the Rays’ bullpen also has the slowest pace in baseball, by most of one second. The Rays’ pitchers have been working slowly, and the Rays’ hitters haven’t been working quickly, and if you put the paces together and weight them by number of pitches, you find that these Rays have reached a new level of, say, being deliberate.
Here are the top ten slowest overall team paces over the course of the 2008-2014 PITCHf/x era:
- 2014 Rays, 24.9 seconds
- 2014 Giants, 24.0
- 2013 Astros, 23.8
- 2014 Dodgers, 23.7
- 2014 Yankees, 23.7
- 2013 Rays, 23.6
- 2013 Red Sox, 23.4
- 2011 Red Sox, 23.4
- 2013 Angels, 23.4
- 2012 Red Sox, 23.3
Of course, we have the smallest samples in 2014, and over the smallest samples, you’ll expect to see the broadest spreads. Four of those ten teams are from this season, and one should expect them to regress, at least in part. But the Rays aren’t just in front — they’re in front by almost an entire second, and that’s almost an entire second per pitch. A game has a few hundred pitches.
According to conventional wisdom, it’s the Red Sox and Yankees whose games take forever. And recently, that’s been true, at least relatively speaking. But now they’ve been surpassed ever so quietly by a division rival, and in large part for this reason the average Rays game has taken three hours and 24 minutes. That’s the longest average game duration in baseball history, and though it’s also tied with the 2014 Dodgers, the average Dodgers game has packed in a little more action. A Rays game has averaged just under 78 plate appearances. A Dodgers game has averaged just under 79. It’s enough to make the Rays the leaders on an adjusted rate basis, so they’ve had presumably the slowest pace ever, and the longest duration ever. There’s a lot of baseball history in history.
And as it happens, this was just talked about the other day, on Rays.com. Joe Maddon’s explanation:
“I think part of that is the number of pitches we do see. You have to consider from the offensive perspective also. I don’t know if we lead the league in potential challenges, play on the field, I don’t know how that plays out.
“We have a couple of pitchers who probably can be a little more time efficient between pitches. But overall, I’m a little surprised that that’s true, quite frankly. The only thing I think that validates [the Rays leading the league in longest games], looking at the numbers is we do see a lot of pitches on a nightly basis as an offense. We’ve worked good at-bats the whole season.”
Another perspective:
Rays catcher Jose Molina, who has seen a lot having first played in the Major Leagues in 1999, said the only thing he’s noticed is “guys not throwing strikes.”
“Pitchers not throwing strikes or hitters having good at-bats and fouling off good pitches,” Molina said. “And that takes you to a long inning, and a lot of changes of pitchers.”
Maddon mostly focused on the hitters. It’s absolutely true that Rays hitters have worked some long plate appearances, and the deeper a count gets, the slower a pitcher tends to work. And the more guys that get on base, the longer a game goes in minutes. But counter to Maddon’s point, it isn’t the Rays’ hitters who’ve been unusually slow — it’s their own starters and relievers who stand out. As he noted in passing, perhaps those arms could be a little more time efficient between pitches.
As for Molina, it’s once again about long at-bats. Long at-bats and good at-bats increase both pace and duration. Pitchers slow down in deep counts, and pitchers slow down with runners on base. This year, Rays hitters are fourth in baseball in rate of pitches seen with men on, and Rays pitchers have baseball’s third-highest rate. Both rates are higher than they were in 2013, contributing to everything moving slower. The hitters have been more difficult to put away, and the pitchers have just been worse, in part due to worse health.
There are reasons why the Rays work so slowly. Given the exceptional nature of their numbers, by the end of the year they’ll probably look a little more normal. But they’re also on pace to be the slowest-working team in baseball history, most likely, and we’re a quarter of the way through the season. There’s a lot of time between pitches, and as a direct result, there’s a lot of time between starts and conclusions. It’s not a stat that’s going to determine their wins and losses. The Rays don’t seem to mind it too much. It’s just a stat that is. It’s a part of the Rays viewing experience.
Jeff made Lookout Landing a thing, but he does not still write there about the Mariners. He does write here, sometimes about the Mariners, but usually not.
For a second I thought this was gonna be about the Orioles front office and Dan Duquette.
If I was this guys boss I’d fire him. You get paid to cover baseball, not complain about how you don’t want to be there covering baseball.
If I were Tom B’s boss I’d fire him.
You get paid to stand in front of Little Caesars with a HOT PIZZA! sign, not make comments on articles that you didn’t understand.
But he came highly recommended from that “Cash 4 Gold” place down in the nearly abandoned strip center!
HA.
Except you make my point. I get paid to stand there with the sign, not COMPLAIN about how long I’ve been out there to potential customers.
Your unsatisfying personal relationships all share one common element.
Come on, Jeff. Just let the big red numbers speak for themselves – no need to give him a response.
Sometimes you don’t have much to do when you’re waiting at the dentist.
What?! You’re paid to cover baseball, not go to the dentist!
Is this real life? Is this really happening?!
Wow, the response here is pretty ridiculous. This man (the Mariner’s beat writer, I wasn’t talking about YOU Jeff….) speaks publicly as a voice of the newspaper that employs him, and complaining about the length of the games only strikes me as unprofessional?
So your response is a personal attack at me, Jeff?
Get a fucking grip.
(1) I don’t see a complaint in the tweet, just an observation.
(2) Do you get how your comment was unclear?
(3) Unnecessarily long games are a drag for writers who are working on an unforgiving deadline. Put yourself in their shoes. They don’t get to watch baseball like you get to watch baseball.
I don’t see how what I said was unclear, you don’t cover live games anymore? Right? Obviously I wasn’t referring to you then…
You only reference the tweet in the first place because of the complaint it raised. The length of the game. He didn’t reference it because it’s some how amazing that it was some arbitrary amount of time, but that he felt it was a loooong time.
It’s all good… your response just struck me as odd. I made a poorly constructed joke and lost 20 internet points. I’ll just have to try again tomorrow.
What you said was unclear because “this guy”, it seemed to me (and many others judging by the downvotes), was Jeff, and not Greg Johns. Both get paid to cover baseball, and neither seemed to be complaining, so your comment was a little confusing and unclear.
I don’t want to get in any arguments here; I’m sure it was just a misunderstanding. You just didn’t seem sure why your comment was unclear, so I’m trying to show you why it is.
It’s all good, you can purchase blocks of 100 internet points for $9.99.
Or by making a Ruben Amaro/old player quip.
Hey Murphy Brown, even if this guy tweeted “I don’t want to be here covering baseball” that would not be something he would be fired for. Also, shut up.
http://goo.gl/KlWft
This website needs a sign that says “Slow children crossing”.
go get a fro-gurt
http://www.youtube.com/watch?v=r_pqnsKWlpc
They are paid to cover baseball and they are propably not getting paid by the our. Wouldn’t you also complain if your job takes more time than usual?
I think maybe you did not read the thing.
Does the number of relief pitcher switches per game have a significant effect on the length?
Random Internet guy: No idea why [insert baseball minutia here].
Jeff Sullivan: Challenge accepted!
I really enjoy reading your articles, Jeff. You seem to have a very unique perspective of this great game, and I thank you for continuing to share it with the world.
I enjoy Sullivan’s articles too, but there is no f’ing thing as “very unique.” Unique does not take a modifier. Ever!
It just took one and liked it, though.
And did you see the way it was dressed? It was practically asking for it…
I don’t see why unique can’t have a modifier if it refers to things that are not common. There are different levels of uncommon-ness. It’s not like a light being on or off. You can’t say the light was very on or almost off, but you can say something is somewhat uncommon (Mike Trout’s defense) and something is extremely uncommon (Mike Trout’s WAR).
however, unique does not mean uncommon. it means one of a kind. and something cannot be more “one of a kind” than something else because there is only one.
A light could be almost off if it is controlled with a rheostat.
ryarriz, you must feel exceptionally stressed due to your pedantry. I recommend two vicoden followed by free-flowing beer. Temporarily relieves consciousness!
At least word police knows he has a problem.
@ryarriz/Word Police:
Let’s acknowledge the idea that words mean nothing objectively, only what concepts we’ve arbitrarily decided to link to them, so unique means whatever the hell we want it to. In common usage, the definition has expanded from “dissimilar in some way from everything else” to “particularly uncommon or unseen, or differentiable from otherwise similar things”. The former definition would not allow a qualifier as it is simply a binary, the second would as it is a degree.
More importantly, that definition of unique is near-useless. It’s impossible to argue that anything is unique from a trait perspective (i.e. I have this quality, other things have this quality, therefore this quality is not unique) or that anything is not unique from a composition perspective (i.e. nothing has the precise attributes that another thing has). It would be as impossible to prove something is or isn’t unique as it would be to prove that there are no red swans.
Therefore the “true” definition of unique is useless; either it applies to everything or it applies to nothing, so it cannot differentiate one group of things from another or meaningfully describe something, which are the purposes of an adjective. The more widely-used definition has some value, indicating something that is clearly distinguishable but not wholly different in a particular realm, and it supports the addition of the “very” qualifier.
Now, this length and degree of explication probably seems wholly unnecessary. But it appears as though you were mostly interested in providing a dictionary definition of a word in order to feel superior to someone else and their “incorrect” use of a word; everyone knows what he meant, and he probably knows the “real” definition of that word, but it’s a useful term that helped make his point. Since the point of language is to convey ideas, and “very unique” helped do that, what he did is totally fine. So it is important to me that any feelings of superiority you might still be deluding yourself with evaporate as quickly and completely as possible.
The reason to use precise meaning for words is to better convey what is intended. Many words are in the English language to offer subtle differences in meaning. If you say that words that have similar meaning can always be used interchangeably, that eventually dumbs down the language.
In Chinese Mandarin the word “ma” can have five different meanings depending on which of five tonal inflections are used. You could get by without the inflections and rely instead on context, and just assume the listener would always get the meaning. Much better to use the inflections and have precise meaning, and not call your friend’s mother a horse.
If unique has lost its specific meaning and can be used to generally mean uncommon or odd, then it is a loss to effective communication.
Unique never takes a modifier, unless of course, you are using the third definition of the word:
http://www.merriam-webster.com/dictionary/unique
Few things make an internet commenter look dumber than “correcting” someone that isn’t wrong.
@Quincy, your relativistic take on the meaning of words is reminiscent of Humpty Dumpty —
‘I don’t know what you mean by “glory”,’ Alice said.
Humpty Dumpty smiled contemptuously. ‘Of course you don’t — till I tell you. I meant “there’s a nice knock-down argument for you!”‘
‘But “glory” doesn’t mean “a nice knock-down argument”,’ Alice objected.
‘When I use a word,’ Humpty Dumpty said, in rather a scornful tone, ‘it means just what I choose it to mean — neither more nor less.’
‘The question is,’ said Alice, ‘whether you can make words mean so many different things.’
‘The question is,’ said Humpty Dumpty, ‘which is to be master — that’s all
———-
Pretty sure we don’t let writers choose the meaning of their words.
Wow, the miriam webster dictionary has a whole paragraph about how you’re wrong:
“Many commentators have objected to the comparison or modification (as by somewhat or very) of unique, often asserting that a thing is either unique or it is not. Objections are based chiefly on the assumption that unique has but a single absolute sense, an assumption contradicted by information readily available in a dictionary. Unique dates back to the 17th century but was little used until the end of the 18th when, according to the Oxford English Dictionary, it was reacquired from French. H. J. Todd entered it as a foreign word in his edition (1818) of Johnson’s Dictionary, characterizing it as “affected and useless.” Around the middle of the 19th century it ceased to be considered foreign and came into considerable popular use. With popular use came a broadening of application beyond the original two meanings (here numbered 1 and 2a). In modern use both comparison and modification are widespread and standard but are confined to the extended senses 2b and 3. When sense 1 or sense 2a is intended, unique is used without qualifying modifiers.”
Sorry, word police. “Very unique” is now a part of the English language as normally spoken and accepted by dictionaries.
You had better find a new occupation.
They do this to get Todd Kalas more air time
You looked at Pitchers and Batters, what about catchers? It could be that every Rays pitcher is taking a little longer, but it could also be that Molina is slowing the pace down.
The NFL has a play clock, and the NBA has a shot clock.
How long before we see MLB introduce some sort of pitch clock?
Naturally, as a non-timed game, this would be more difficult to do correctly, but there are a few things that could be done relatively painlessly:
30-second pitch clock to each new hitter (or during an AB with a SB attempt). 20 seconds after a foul ball. 15 seconds within an AB when it’s just the pitcher and catcher touching the ball.
Each team gets 5 time-outs during a game. These are used any time a coach/manager steps out of the dugout.
A pitch clock violation also uses a time out.
Once time-outs are used up, a clock reaching zero results in an immediate ball to the current (or next) batter. The clock is reset to 15 and starts counting again with the same consequences.
Potentially, if a team uses all its timeouts, and wants to make a pitching change that takes 1 minute to get pitchers swapped out, the new pitcher would start the AB 3-0 (30+15+15 = 3 balls).
Umps can stop the clock, at any time, at their discretion.
Clearly half-baked at this point, but with the talk of slow games getting much louder in the last few years, I’d be shocked if something drastic like this weren’t strongly considered in the next few CBA cycles.
A clock isn’t really in the nature of baseball, if you ask me. Strict timing rules would make the game almost unrecognizable, because baseball has never had the direct pressure of a clock.
I do think it’s in the game’s best interest to reverse some of the slowing down. A game takes almost twice as long to play as it did 100 years ago, which means a baseball fan, casual or otherwise, is being asked to invest almost twice as much time, and anyone with a short attention span will move to basketball, where points are scored every 10 seconds.
A speed-up plan could simply take the form of rules against dilly-dallying on the mound or in the batter’s box, repeated mound visits by other defenders (especially catchers), long times to the plate, etc. Baseball would be a lot faster if everyone were more like Mark Buehrle, and while some players will have to make significant adjustments, it will do very little to actually alter the spirit of the game.
“baseball has never had the direct pressure of a clock.”…..There used to be a clock, it was called the sun
The rule is already in place, actually.
[link]
Does anybody have a good reason as to why this rule is not enforced?
That we don’t yet have robot umps. Oh wait, you said a GOOD reason. Nope, sorry.
In some international softball turnaments they use a pitch clock. Not sure if it also happens at baseball turnaments.
Softball has the benefit of no lead for runners and therefor no holding of runners though. So it is much easier to enforce a pitch clock.
Plus everyone is ready to leave the game to have a beer or twelve.
Honestly shocked it isn’t the Twins. Pelfrey, Deduno take aaaaaaaages to throw the dang ball already. Painful to watch.
Of course, slow games are hardly the only reason the Twins are painful to watch these days….
🙁
Were Deduno a qualifying pitcher, he would actually rank 90th or so on the slowest thrower list.
Combine 3 injured starters, a manager who likes to mix and match bullpen matchups, the new instant replay rule and the manager’s slow walk out to argue while waiting to hear if a challenge should be made, and you get really slow games. Having watched most of them, I can tell you that instant replay should be scrapped as it slows down the game too much for the limited benefits, and the Rays are going to be in for a very long season if they keep running a long reliever, a guy who should be retired, and Odorizzi out there 3 out of every 5 games for much longer. I think the Rays average like 150 pitches a game or something crazy.
I read somewhere that replay has actually sped up the game by reducing the number of manager-ump arguments. Maybe on fangraphs somewhere.
Has anyone thought about the shifts that are employed now? The Rays, I thought, use the shift more than most teams. If that is the case then wouldn’t it stand to reason that moving fielders around to different positions on the field from one batter to the next would take time to do?
Think about it; a left handed pull hitter is up and the Rays shift guys to the right side of the infield for him and then a right handed batter is up next and they have to move the fielders to a more normal position for him. How long do those two moves take to do? Thirty seconds each? More? Less?
If 30 seconds is the average time it takes to move guys into position and there are 10 shifts a game then there is an extra five minutes. Add in pitchers not throwing strikes, hitters getting deep into counts, coaching visits to the mound, pitching changes and replay reviews and all that time adds up….
I haven’t seen many Rays games this year, but my recollection is that they were adjusting their shifts between pitches based on the count. Perhaps someone who sees them more regularly can confirm this. If this is the case the pitchers are probably giving the coaching staff an opportunity to signal the shifts in addition to the time it takes to actually move.
I was just looking at the 5/16/33 Senators-Indians box score, and that was a 12-inning, 21-run game with 11 pitchers. That game took 3:34. Something else has happened to baseball.
Maybe it’s TV breaks, (2 extra minutes for ads between each half inning is 35-40 minutes), or maybe it’s just the dudes playing it. Or both.
Is that before relief pitchers got warm-up pitches?
Maddon and Molina are both right too, the Rays are 4th in pitches thrown and 4th in pitches seen (total, I’m too lazy to calculate the rates, but eyeballing it looks like the rankings roughly hold up, +/- 1 spot)
I’ve noticed that foul balls result in a long delay compared to a swinging strike, even if no runners on base, the batter doesn’t leave the box, and no fielder moves away from his position. Why?? That’s one thing that bugs me constantly: to see the pitcher throwing every 10-15 seconds until we suddenly get a 45-second break in the rhythm after a ball gets tipped straight back into the seats. The pitcher needs rubbing time with a new ball and that seems to be an unsolvable problem, except there must be a better way. Can you use your advanced stats to see if game time correlates to the number of foul balls, or anything else for that matter?
And a batter should be called out automatically if he adjusts the same batting glove twice on the same pitch. Once per pitch is more than enough.
I feel like I read a few years ago that the actual number of pitches per game is increasing. I assume that would have to correlate to more “balls” being thrown, as you would expect more “strikes” to speed things up.
Is there anywhere to look that data up? I feel like I should be finding it on BBRef but I can’t put it together.
Pitchers nibbling for the ever looming strikeout stat, combined with the increased time between pitches, answers can be had!
League, team and individual pitch data for 2014:
http://www.baseball-reference.com/leagues/MLB/2014-pitches-pitching.shtml
Click through for other years.
I find it hard to believe that the Rays’ overall pitcher pace will “look more normal” as the year goes along. At some point Hellickson will be back, and isn’t he one of the slowest-paced pitchers in baseball?
Slowest working and no mention of Josh Beckett?
Who the hell is slowing down the Giants? I’m going to have words with him. Give me a name.
Vogelsong, I bet. Thank god there are no more Moscoso games to put us to sleep this season, though.
No wonder no one goes to their games.
Erik Bedard is a common element between the 2013 Astros and the 2014 Rays. Like the guy, but he sure does take his time.
He contributed to the 2011 Red Sox as well.
I watch many Rays games, and the pitchers pace is sllllllooooowwww. It’s unbearable, and probably if they had better pacing they would throw more strikes?
The Rays and also the Yankees practice defensive shifting to a high degree. The Rays sometimes change their shifts depending on the count. I am not surprised that this shows up as the Rays’ pitchers being slow. They have to allow for the changing shifts called from the dugout. I have not noticed the Yankees shifting much within a single at-bat, but they do rank high in the % of the time they use shifts.