Baseball for the Thinking Fan

You are here > Home > Baseball Newsstand > Baseball Primer Newsblog > Discussion
Baseball Primer Newsblog
— The Best News Links from the Baseball Newsstand

## Friday, August 15, 2014

#### How Bullpens Took Over Modern Baseball

The broader trend that goes back half a century is clear. In 1964 (four years after the save rule first came to baseball), teams used an average of 2.58 pitchers per game, including the starter; today, they’re using 3.92 pitchers per game. In ’64, relievers tossed an average of 2.64 innings per game; today, they’re throwing an eyelash more than three innings per game.
...
As a result, managers choose a few relievers from a phalanx of fireballers, then go get a few more if some of them break down.

In other words, the pitchers might be on the mound for fewer and fewer pitches, but the trend of harder throwers looks like it’s here to stay.

Doing the math, the average reliever in 1964 threw 1.67 IP/G as opposed to 1.03 IP/G today.

Joyful Calculus Instructor Posted: August 15, 2014 at 01:03 PM | 64 comment(s) Login to Bookmark
Tags: jonah keri, relief pitching

Go to end of page

Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.

1. shoewizard Posted: August 15, 2014 at 11:45 PM (#4772257)
Simple solution to a lot of todays woes in the game. No more than two relievers may appear in one inning, and no more than 3 relievers in back to back innings. ( so if you use two relievers in the the 7th you can only use 1 in the 8th)

Wont be a dramatic shift, but just enough to move the needle and also add some interesting strategy.

That and of course keep hitters in the box
2. madvillain Posted: August 16, 2014 at 12:11 AM (#4772262)
Nah, those rules will never fly imo. Less invasive (and thus simpler in a more important way) solutions to length of play are available. What if a reliever gets injured? What if he's wildly ineffective? Too many variables in play imo. Just put hitters and pitchers on the clock and go to booth replay, boom you just cut down the average game 20 minutes.
3.  Posted: August 16, 2014 at 12:14 AM (#4772264)
Trying to read the article...and the first four paragraphs put me to sleep... what does it ultimately say?
4.  Posted: August 16, 2014 at 12:27 AM (#4772265)
1. First chart talks about increase of top speed of pitchers...basically pitchers know they are going to go no more than an inning and are putting more effort at 100%. It's nice seeing confirmation of what most people suspect that there are more pitchers throwing high heat.

2. There is a huge problem using ops+ to analyze changes from year to year, but still it's a good chart to show the difference between performance of relievers and starters. I would argue that the best conclusion to draw from that chart is that there is a good possibility that pitchers who would have been starters are now becoming relievers. Where in the past you could rank teams best pitchers as 5 of a teams 6 best pitchers were starters, it's now probable that on average 3 out of 6 are starters.

I think discussing relief usage is a good topic.
5. Pasta-diving Jeter (jmac66) Posted: August 16, 2014 at 12:28 AM (#4772267)
Trying to read the article...and the first four paragraphs put me to sleep... what does it ultimately say?

if you look at Figure 3 (they're not numbered, but it's the 3rd fig). the OPS+ allowed by relievers has plummeted downwards in the last 50 years--in the 60's and early 70's the OPS+ allowed by relievers was worse than that of starters, but now, it's just the opposite. OPS+ by relievers is around 94, while that for starters is around 103. Which suggests (I hate to say it) the modern use of relievers is effective (yech)
6.  Posted: August 16, 2014 at 09:59 AM (#4772330)
Which suggests (I hate to say it) the modern use of relievers is effective (yech)

which anyone who has looked at the issue for more than five minutes has known for a long time. I've been arguing it at least since 2007 when I showed similar charts to the ones in this article at a conference in San Francisco.

Part of the reason relievers have become so effective is that the expansion of their use without a change in roster size means that managers have fewer bench options from which to choose. Because you need more relievers in today's usage model (and the effectiveness of the model means that there is no internal incentive to move away from them), you wind up carrying seven relievers - and either four or five bench players, one of whom (the backup catcher) you probably aren't going to use unless you must.

Here's what I would do:

1. Increase roster size to 30, of which only 25 can be designated as active for a day's game;
2. The five-man taxi squad can have no more than two pitchers;
3. No more than 10 pitchers can be active for any one game.

Would a couple more hitters on the bench make a difference? I don't know, but I'd like to see it tried. I don't particularly like rules that say you can't use more than "x" pitchers in an inning because that puts a manager in a position where he "can't" act. I'd rather see if there's a way to give a manager more options to counter a reliever.

-- MWE
7. McCoy Posted: August 16, 2014 at 10:07 AM (#4772334)
My solution has always been no substitutions during the inning unless it is for an injury. If you want to bring in a reliever, hitter, or a defensive sub you have to do so after 3 outs and before the start of the next half inning. If a player gets pulled because of an injury he must sit out at least 3 games if he is a positional player, 6 games if he is a reliever, or go on the DL. Problem solved.

The playoffs would require different rules. If you pull a player because of injury he cannot play the rest of the series. If you pull a player during the series because of injury and MLB thinks you pulled them even though they weren't injured you forfeit your first round draft pick, player possibly gets suspended, manager/coaches get suspended, GM gets suspended, owner gets suspended, and the team gets hit with a huge fine.
8.  Posted: August 16, 2014 at 10:20 AM (#4772335)
Would a couple more hitters on the bench make a difference?

I doubt it. For every Manny Mota there were a dozen Ed Armbristers, guys we only remember for reasons that had nothing to do with their hitting prowess.
9.  Posted: August 16, 2014 at 11:29 AM (#4772358)
In ’64, relievers tossed an average of 2.64 innings per game; today, they’re throwing an eyelash more than three innings per game

Doing the math, the average reliever in 1964 threw 1.67 IP/G as opposed to 1.03 IP/G today

So bullpens face just one more batter per game now than 50 years ago; but individual relievers face two batters fewer.

One thing that has clearly disappeared is the quick hook, which accounts for some of this dynamic. IOW, despite a huge drop in CGs, bullpens aren't on balance pitching a lot more of games, because starters so rarely leave early. This doctrine has become pretty inflexible (and perhaps for excellent reasons of staff management overall).

Take two seasons almost at random, by Cub pitchers having undistinguished years: Dick Ellsworth 1964 vs. Jeff Samardzija 2013. Both were regular rotation starters all year with below-average ERAs and W-L records. Ellsworth made 36 starts, Samardzija 33.

Ellsworth's starts ranged between 11.1 and 1.2 innings, 50 and 12 batters faced. He had four starts with more than 40 BF, and five with fewer than 20 BF (parts of the range Samardzija didn't even approach).

Samardzija, by contrast, threw in a range between 9 and 3.1, 33 and 22 BF. He threw 24 starts with BF between 25 and 30 inclusive. By contrast, Ellsworth in 1964 had only eight such 25-30 starts.

The standardization in starter usage nowadays is to me even more striking than the bullpen usage. A starter goes out nowadays, and unless something extremely freaky happens, he will get to 100 pitches and be gone soon thereafter, almost without regard to the score of the game (which includes pinch-hitting strategy, of course; in 1964 I'd say it was much more common to hit for a starter if you were behind at all in the middle innings, no matter what his pitch count at that point).
10. valuearbitrageur Posted: August 16, 2014 at 11:58 AM (#4772383)
What if a reliever gets injured? What if he's wildly ineffective? Too many variables in play imo

The manager can still yank the reliever even if team is over his limit, but it's a balk.

Just forcing relievers to face two batters would be a significant change, it would eliminate running relievers out back to back to maximize matchups, and speed slowest part of game.

Your suggestions are good, they speed game but don't address offensive imbalance or high K rates.

11. PASTE, Now with Extra Pitch and Extra Stamina Posted: August 16, 2014 at 12:50 PM (#4772410)
3. No more than 10 pitchers can be active for any one game.

Only one starter will be active, so you're still allowing for a nine man bullpen here. Unless your unspoken suggestion is that the team can only change its designated pitchers once every week or two.
12. Pat Rapper's Delight (as quoted on MLB Network) Posted: August 16, 2014 at 01:39 PM (#4772427)
One thing that has clearly disappeared is the quick hook, which accounts for some of this dynamic.

I was talking about that recently on another forum. Drives me nuts that managers will leave an ineffective starter (perhaps at the beginning stages of nursing an injury...) to absorb a line like 4-2/3 12 8 8 4 2 just so he can "get his innings in" and "save the 7-man bullpen" because bullpen roles are now so regimented that they don't know what to do if the game dictates they should bring in a reliever before the 5th inning. So they do nothing.
13. PASTE, Now with Extra Pitch and Extra Stamina Posted: August 16, 2014 at 02:00 PM (#4772433)
As a believer in keeping workloads as stable and consistent as possible, I would be inclined to keep a pitcher who has nothing in the game until he throws 90 pitches, in the interests of acclimating his mind and body to going 90-110 pitches every fifth day like clockwork (so long as he isn't hurt, of course). You could pull him from the game and have him throw pitches in the bullpen a while before he takes a shower, but usually in the case where he's getting lit up the game is effectively over by the third inning anyway.

I don't know whether any of the managers who prefer to let a struggling starter struggle think along these lines, or if they all just do it to "save the bullpen".
14. valuearbitrageur Posted: August 16, 2014 at 02:23 PM (#4772437)
I don't know whether any of the managers who prefer to let a struggling starter struggle think along these lines, or if they all just do it to "save the bullpen".

Kirk Gibson does it because his starters "owe him 100 pitches". Despite throwing 44 starts and 300 innings at a 3.00 ERA (130 ERA+ ish) over his first 2 seasons, the next year Daniel Hudson got in Gibby's dog house early because in his first 9 starts only went 6 innings or more 4 times. The other starts he went 5 IP (83 pitches), 4 IP (90 pitches), 3.2 IP (100 pitches), and 1.2 IP twice (40 & 50 pitches). He blew his arm out on the 50 pitch outing.

By my count Hudson still is in deficit 90 pitches to Gibby, which I'm sure he'll demand if and when Daniel returns from back to back Tommy John surgeries.
15.  Posted: August 16, 2014 at 02:24 PM (#4772438)
1. Increase roster size to 30, of which only 25 can be designated as active for a day's game;
2. The five-man taxi squad can have no more than two pitchers;
3. No more than 10 pitchers can be active for any one game.

I've been pushing something similar 28 instead of 30, with 25 being designated as active for a series. It's not as drastic of a suggestion that you are putting out there, and I also didn't have a roster requirement. My thought process would be that each team would put their two starting pitchers that wouldn't pitch in the series on the inactive roster, along with any position player that is having a day to day injury or just not good enough, teams would probably then replace one of the starting pitchers with a reliever giving most teams 3 starting pitchers and 8 relievers for a series, but at least one of the spots from the dropped starting pitcher will hopefully be filled with a bat.

I actively oppose any rule that would argue for forcing a pitcher to pitch to a number of batters, that is too much loss of managerial control in my opinion.

16. Lance Reddick! Lance him! Posted: August 16, 2014 at 02:25 PM (#4772439)
Part of the reason relievers have become so effective is that the expansion of their use without a change in roster size means that managers have fewer bench options from which to choose. Because you need more relievers in today's usage model (and the effectiveness of the model means that there is no internal incentive to move away from them), you wind up carrying seven relievers - and either four or five bench players, one of whom (the backup catcher) you probably aren't going to use unless you must.

A return to the 11-man staff while maintaining a 7-man bullpen can be achieved organically by the first team to think of the obvious:

If starters are only being asked to throw 100 pitches, why do they still need the fourth day of rest we decided to give them when they were throwing 130? Shorten the rotation, with the fifth starter becoming a long reliever (or acting in a platoon with the fourth starter), and keep six dedicated short relievers. Voila, bench bat.

3. No more than 10 pitchers can be active for any one game.

Only one starter will be active, so you're still allowing for a nine man bullpen here. Unless your unspoken suggestion is that the team can only change its designated pitchers once every week or two.

His spoken suggestion, directly above that one: 2. The five-man taxi squad can have no more than two pitchers.
17.  Posted: August 16, 2014 at 02:32 PM (#4772441)
If starters are only being asked to throw 100 pitches, why do they still need the fourth day of rest we decided to give them when they were throwing 130?

Because your body still needs the time to heal.
18. Lance Reddick! Lance him! Posted: August 16, 2014 at 02:40 PM (#4772443)
The body needs time to recover, but why does it need *as much* when it's been taxed less in the first place? The answer is that it doesn't, or else one-inning relievers would be getting shelved for days at a time.
19. McCoy Posted: August 16, 2014 at 02:51 PM (#4772444)
Thank you, doctor.
20.  Posted: August 16, 2014 at 02:52 PM (#4772445)
The body needs time to recover, but why does it need *as much* when it's been taxed less in the first place? The answer is that it doesn't, or else one-inning relievers would be getting shelved for days at a time

The difference between 130 and 110 isn't that big of a difference, basically it means that a modern pitcher probably threw 180 pitches in the game while a pitcher from the 70's threw 210 pitches. It's still a workout. And of course pitchers in the day were getting career ending injuries at a higher rate than modern pitchers.
21. Ziggy: The Platonic Form of Russell Branyan Posted: August 16, 2014 at 05:39 PM (#4772498)
Since the back end of the bullpen is basically AAA guys anyway, why don't teams (more often) artificially increase their roster size by shuffling tired relievers back and forth through AAA? Just tell the minor league manager "don't use these guys, we're recalling them in three days, they need their rest". Hell, it doesn't have to be AAA, since they're never going to pitch there, let it be rookie ball or something. In fact, you don't even need to have them physically go to the minor league team, officially assign them there but keep them physically with the big league club so they can be "called up" whenever necessary.

This has the added advantage that your relievers don't accrue as much service time. Maybe not a big deal for lots of them (bad relievers don't usually make much anyway), but it'll make a difference in some cases.
22. tfbg9 Posted: August 16, 2014 at 05:40 PM (#4772500)
This thread represents to me why this site is still worth coming to. Good thread. Just baseball.
23.  Posted: August 16, 2014 at 05:44 PM (#4772504)
ince the back end of the bullpen is basically AAA guys anyway, why don't teams (more often) artificially increase their roster size by shuffling tired relievers back and forth through AAA? Just tell the minor league manager "don't use these guys, we're recalling them in three days, they need their rest".

When a player gets sent down, barring sending someone to the disabled list, they have to stay down for 10 days.
24. Pasta-diving Jeter (jmac66) Posted: August 16, 2014 at 05:47 PM (#4772507)
When a player gets sent down, barring sending someone to the disabled list, they have to stay down for 10 days.

rather specifically to prevent teams from doing what was suggested by Ziggy
25. Ziggy: The Platonic Form of Russell Branyan Posted: August 16, 2014 at 05:49 PM (#4772508)
Well rats. It seemed like such a good idea.
26. McCoy Posted: August 16, 2014 at 06:37 PM (#4772520)
You could still do it but even if the rule didn't exist it would have limited value. You don't really want AAA relievers throwing innings at the major league level. If you have 7 or 8 middle relievers with options of comparable ability and skill you could shuttle them all between MLB and the minors but the options would be the key. How many teams are going to have a bunch of relievers with options on their 25 man roster or on their 40 man roster.
27.  Posted: August 16, 2014 at 07:20 PM (#4772535)
You don't really want AAA relievers throwing innings at the major league level

NOW you tell me. Hang on while I copy that sentence and send it to Ron Washington.
28. puck Posted: August 16, 2014 at 07:26 PM (#4772538)
And of course pitchers in the day were getting career ending injuries at a higher rate than modern pitchers.

Wait, is this known? It seems far from an "of course" piece of knowledge.
29.  Posted: August 16, 2014 at 07:27 PM (#4772540)

Wait, is this known? It seems far from an "of course" piece of knowledge.

Considering there was no Tommy John surgery, or other ways of making it back to the league, it seems pretty obvious.
30. snapper (history's 42nd greatest monster) Posted: August 16, 2014 at 07:36 PM (#4772542)
I've been pushing something similar 28 instead of 30, with 25 being designated as active for a series. It's not as drastic of a suggestion that you are putting out there, and I also didn't have a roster requirement. My thought process would be that each team would put their two starting pitchers that wouldn't pitch in the series on the inactive roster, along with any position player that is having a day to day injury or just not good enough, teams would probably then replace one of the starting pitchers with a reliever giving most teams 3 starting pitchers and 8 relievers for a series, but at least one of the spots from the dropped starting pitcher will hopefully be filled with a bat.

I actively oppose any rule that would argue for forcing a pitcher to pitch to a number of batters, that is too much loss of managerial control in my opinion.

Hate this idea. You're just giving them MORE RPs to use. Every team will choose to replace the two SPs with 2 more RPs.

1) No more than 11 pitchers on the 25-man roster. Ever.
2) Each pitcher used must face 3 batters. If a pitcher is injured and must leave before the 3 batters, he may not be used for 7 days.

31.  Posted: August 16, 2014 at 07:43 PM (#4772547)

1) No more than 11 pitchers on the 25-man roster. Ever.
2) Each pitcher used must face 3 batters. If a pitcher is injured and must leave before the 3 batters, he may not be used for 7 days.

I can support the first thing, but can't support the second. Roster construction doesn't really affect the game in front of the fans for the most part, but the second thing absolutely does, and I hate it. One batter, one pitcher isn't really that big of a deal. Only reason to propose it is for time of game, and realistically speaking, we are talking less than two minutes, per game on average...you want to save time, you concentrate on the batting box antics.
32. JJ1986 Posted: August 16, 2014 at 07:53 PM (#4772552)
The easiest way to adjust pitching changes to speed up the game is to remove the warmup pitches. The pitcher gets one to test the mound and then is ready to go.
33. valuearbitrageur Posted: August 16, 2014 at 08:03 PM (#4772557)
Each pitcher used must face 3 batters. If a pitcher is injured and must leave before the 3 batters, he may not be used for 7 days.

Why not just make it similar to a balk, batter takes first, all baserunners advance? A manager can't angle shoot that very easily, if ever, even if he planned to intentionally walk the batter, he's still advancing other baserunners.
34. valuearbitrageur Posted: August 16, 2014 at 08:22 PM (#4772566)
The easiest way to adjust pitching changes to speed up the game is to remove the warmup pitches. The pitcher gets one to test the mound and then is ready to go.

Another good idea. Obviously they would fully warm up in the bullpen first so should not a safety issue and doesn't just speed play. Since manager has to warm up relievers more often to prepare for use and their first pitches are more tentative, also reduces reliever effectiveness.
35.  Posted: August 16, 2014 at 08:28 PM (#4772571)
The easiest way to adjust pitching changes to speed up the game is to remove the warmup pitches. The pitcher gets one to test the mound and then is ready to go.

You can tell when someone posts something that has probably never played the game. They might not need eight pitches, but just one? That is insane.

Let's be perfectly honest. No rule that is going to reduce commercial time is ever going to happen. Sometimes when I see these suggestions, I wonder if anyone making them are actually baseball fans.

36. JJ1986 Posted: August 16, 2014 at 09:33 PM (#4772600)
You can tell when someone posts something that has probably never played the game. They might not need eight pitches, but just one? That is insane.

Pitchers get mostly ready in the bullpen. If he needs 7 more pitches out there, then he's ready to come in a minute or two later.
37. Howie Menckel Posted: August 16, 2014 at 09:39 PM (#4772604)
I have mentioned before my 1970 Orioles example, but it's worth it again I think (the shoulder seasons are pretty similar) because to anyone under 40 this might seem unfathomable - or at least unfathomable that us current posters actually remember such a time. team went 108-54 and won World Series.

Your SPs are 20-game winners Cuellar, McNally and HOF Palmer starting 40-40-39 games for 119.
Then Phoebus starts 21 of 27 G to get to 140, and Hardin starts 19 of 36 to get to 159 starts (Phoebus had the 4th and final slot til late June, had a bad outing, and Hardin pretty much took over the 4th SP role til Sept when Phoebus - who had only tossed 18 total IP in July/Aug (might also be injury) - got more chances).

The other 3 starts were by Marcelino Lopez - 1 in Aug and 2 in Sept for a team that won the pennant by 15 G.

In the bullpen, no one threw as many as 62 IP and the entire pen only needed to throw fewer than 400 IP thanks to 60 CG - and more than that were nearly completed.

Richert, DHall (4 IP in August), Watt, Leonhard (who rarely pitched in the 2nd half), Lopez, Drabowsky (traded to KC in June) and Hardin threw almost all the R IP. Only 12 pitchers hurled for these Orioles all season - the 12th was Fred Beene, who relieved 4 times (2 in July, 2 in Sept).

The Orioles alternated between a 9- and 10-man staff that season.

Then they dumped Phoebus, prospect Enzo Hernandez, Beene, and Severinsen on Padres in the offseason for Pat Dobson - who joined the 3 above as Orioles 20-game winners in 1971.
38. Walt Davis Posted: August 16, 2014 at 09:44 PM (#4772607)
Since the back end of the bullpen is basically AAA guys anyway, why don't teams (more often) artificially increase their roster size by shuffling tired relievers back and forth through AAA?

This is exactly what teams do.

Relievers throw about 490-500 innings a year and there are 7 bullpen slots. So that's 70 innings per slot.

In 2013, only 35 individual pitchers even hit the 70 IP mark. Another 57 made it to 60. We're at an average of 3 relievers per team at this point and we're already a bit short of the 210 innings we need out of those 3 slots.

Needless to say, everything after that is a mess as there are another 36 guys who hit the 50 mark. We're already 4 slots in and we're sitting at probably around 250 of the 280 innings we need out of them.

So we've got to fill 250 relief innings out of 3 bullpen slots and no single pitcher is even going to hit 50 innings. All told last year, there were 327 pitchers, at least 95% of their appearances in relief, who pitched at least 5 innings. You can add a few guys who get bounced out of the rotation that are missed by the 95% thing.

There's infinite churn out there for replacing tired and injured guys. It seems rather likely that a large percentage of reliever DL trips are "tired veteran that I want to keep but don't have an option on".

Here's Robert Coello who threw 17 innings for the Angels in 2013 (and some in earlier years too). His first appearance for them was on May 12. Within that first WEEK, he made 5 appearances and threw 7 innings. He then had 4 full days off then made 3 appearances, 3.1 IP in a week; then 4 appearances 3.2 IP in a week, then one more before going down (or DL or somewhere). The guy was around for roughly 27 games, made 13 appearances and 15 IP -- conveniently exactly 1/6 of a season. So if you extrapolated that out, that's 78 games and 90 innings, a massive load in today's game.

Then they brought him back in Sept.

It's hard to find a good reliever who can give you 90 innings a year. It is not so hard to find 6 Robert Coellos. For example, Micheal Roth on the 2013 Angels. From game 11-25, he made 6 relief appearances and one (looks emergency) start -- 9.1 innings total, 6 in relief including 2 2-IP stints. He popped back on 13 May, threw 2.1 innings of relief then 2 innings on 15 May then back down. Pops back up in July, makes 4 sppearances in 8 games, then a few days off, then 2 appearances in 2 days. I can't say for sure but he was on the roster for possibly as few as 32 games and made 14 relief appearances covering 17 innings -- or a 70 game, 85 inning pace.

JC Gutierrez, 30, seems to have started the season as the Royals' mop-up reliever and had just 25 appearances and 29 IP through 90+ games. He was traded to the Angels at the deadline. Over the next 29 games for the Angels, he made 16 appearances and 15 IP -- again an 80-90 IP pace. He also made 11 appearances over their last 22 games. Cory Rasmus was picked up in late Aug and made 6 appearances in 8 games, got a few days off, then 10 appearances in 20 - total 16 appearances, 15 IP in 32 games.

Baseball hasn't completely gotten rid of the relievers who throw 80+ innings a year exactly, they just cobble them together from the detritus of the waiver wire and AAA.
39.  Posted: August 16, 2014 at 09:45 PM (#4772608)
Pitchers get mostly ready in the bullpen. If he needs 7 more pitches out there, then he's ready to come in a minute or two later.

They get their body ready, every mound, every night is different. Depending on humidity, the specific groundscrew, and in some cases, the mounds are tailored based upon the hometeam starting pitcher...so yes, it takes more than one throw to get a proper feel for the mound. This isn't pickup softball, these are elite professionals who can tell the difference between 1/4 ounce weight of the bat etc.

40.  Posted: August 16, 2014 at 09:49 PM (#4772609)
I have mentioned before my 1970 Orioles example, but it's worth it again I think (the shoulder seasons are pretty similar) because to anyone under 40 this might seem unfathomable - or at least unfathomable that us current posters actually remember such a time. team went 108-54 and won World Series.

The thing about mentioning the Orioles, is that even in their day and age, it was an outlier that EVERYONE at the time talked about. Grab the 1968 Tigers, and look at how many of their pitchers are pitching effectively in 1972. In fact grab any team from the late 60's and you will see the same thing. Pitchers just didn't have careers in the 70's. For every Nolan Ryan you can find, you could find a half dozen John Fulghams.
41. bobm Posted: August 16, 2014 at 10:00 PM (#4772618)
The standardization in starter usage nowadays is to me even more striking than the bullpen usage. A starter goes out nowadays, and unless something extremely freaky happens, he will get to 100 pitches and be gone soon thereafter, almost without regard to the score of the game (which includes pinch-hitting strategy, of course

From B-R:

Starts by Pitch Count
```Year P<=89 90<=P<=99 100<=P<=109   P>=110      UNK  Total
1991  1537       765         820     1085        1   4208
2013  1248      1494        1482      637        1   4862
```

% of Starts (rounded) by Pitch Count
```Year P<=89 90<=P<=99 100<=P<=109   P>=110      UNK  Total      90<=P<=109
1991   37%       18%         19%      26%       0%   100%             38%
2013   26%       31%         30%      13%       0%   100%             61%
```

42. Walt Davis Posted: August 16, 2014 at 10:03 PM (#4772622)
The OPS+ chart is reminiscent of one I did a long, long time ago for an article here on reliever usage -- for which you can find some of the unformatted HTML code here (http://www.baseballthinkfactory.org/primate_studies/discussion/wdavis_2003-01-15_0) :-) ... don't ask me.

It's no longer relevant I suppose but in the early days of this shift it seemed important to look at 1-inning relievers vs. multi-inning. Some of that drop he shows from mid-90s to early 00s is, if I recall, much more dramatic if you try to remove the mop-up guys. That is, the guys you'd actually use in a close game were substantially more dominant relative to starters than that chart makes it appear.

It's somewhat moot now that pretty much everybody is a 1-inning guy but in the transition from fireman to closer there were still a fair number of set-up guys, etc. pitching multiple innings (I think I used 1.1 IP/appearance as my cutoff). I used ERA+ at the time (don't think we had OPS+ by pitcher in those days) but my memory is that for short relievers, it was on the order of a 115-120 ERA+ average (vs. 100ish for starters) and the overall reliever vs. starter comp was pulled back to about 108-110 by horrible mop-up guys.

teams used an average of 2.58 pitchers per game, including the starter; today, they’re using 3.92 pitchers per game. In ’64, relievers tossed an average of 2.64 innings per game; today, they’re throwing an eyelash more than three innings per game.

I hope this sinks in because this is the major change we've seen in bullpen usage. Starters aren't, on average, leaving games that much earlier -- about 1/3 of an inning although I suspect that's dropping a bit again. It's the number of relievers used to cover those last 3 innings that has skyrocketed but, as MWE, this guy and even me back in the day pointed out, that seems to be because it's really effective.

The other change is the one discussed above that starters are essentially required to go at least 5 innings in every start. So while the mean IP/start hasn't changed a lot, the variation around that mean has been substantially reduced. And of course the shift from 4-day rotations to 5-man rotations reducing the number of starts per season.

But do keep that in mind next time somebody points out that all this "babying" of pitchers doesn't seemed to have reduced injuries. On a per-start basis, they aren't being babied -- they're pitching nearly the same IP/start. Why the drop in the number of starts hasn't had more obvious benefit is a good question but in the old days, unless it was Fergie or similar, those CG they threw were balanced by getting yanked in the 3rd when it wasn't their day.
43. The District Attorney Posted: August 16, 2014 at 10:15 PM (#4772632)
Good Lord, I've only ever heard of one of the four guys mentioned in #38. (Which itself is not ideal from the fan point of view, IMHO. I don't want it to be like football where an individual player might have such a small and specialized role that I don't even know his name.)

I think that:

• In life, you're less likely to get hurt when you work out frequently and less strenuously, as opposed to infrequently and more strenuously. If you combined a four-man rotation with a slightly quicker hook than currently (i.e., the "danger zone" is more like 80 pitches, rather than 100), you'd almost surely win more games, and I think you'd see injuries decline if anything -- certainly not increase. Of course, I can't prove this. But it makes sense to me, and I don't think anyone can disprove it either. Baseball is unfortunately not a science where you get to run controlled experiments. I do at least firmly believe science supports my first statement.

• As far as the bullpen goes, there's no doubt that when a guy knows he's only going to pitch one inning, he can go all-out, and will pitch better in that one inning. However, there is a tradeoff involved. In order to make it possible to do that with every reliever, you have to carry fewer hitters. Hitters also benefit from rest (albeit not as much as pitchers), and there is more to be gained from platooning hitters than from platooning relievers. I can justify carrying "extra" relievers if the purpose is to spread around the relief workload and make each guy better. I don't think that the LOOGY who's going to pitch 30 innings is worth the tradeoff.

• I don't care about the rules nearly as much, but I guess I'll offer an opinion. Limiting relief use via the rules does feel artificial to me. I'd like to see both a smaller roster and no warmup throws from the mound, but it is probably pointless to even bring them up, because one costs the union jobs, and the other costs everyone money. So I end up thinking that practically speaking, the way to speed up the game is to eliminate in-game dawdling, which has no beneficiaries.
44. 6 - 4 - 3 Posted: August 16, 2014 at 10:44 PM (#4772640)
Call the strikezone as in the rulebook
=> increase the likelihood of a taken pitch being called strike
=> decrease the likelihood of a player reaching via walk
=> batters more aggressive early in the count
=> starters throw fewer pitches, last longer
=> fewer walks and strikeouts
=> lions and zebras live together in perfect harmony
45.  Posted: August 16, 2014 at 10:55 PM (#4772641)
Call the strikezone as in the rulebook
=> increase the likelihood of a taken pitch being called strike
=> decrease the likelihood of a player reaching via walk
=> batters more aggressive early in the count
=> starters throw fewer pitches, last longer
=> fewer walks and strikeouts
=> lions and zebras live together in perfect harmony

Not sure what this helps. As it stands strike outs are increasing, walks are decreasing already. Not sure we want anything that would accelerate those actions.

46. Hysterical & Useless Posted: August 16, 2014 at 10:58 PM (#4772643)
Pitching changes provide commercial revenue, as well as a chance to get a snack/take a leak/update your scorecard. Rule changes to limit pitching changes are not needed and would be detrimental to the game. Virtually the entire problem with game time/pace is that every hitter nowadays thinks he's Mike ####### Hargrove. Why don't we petition MLB to have the umpires enforce the rules? In baseball, players can only ASK for time to be called; the umpire is not obligated to grant the request, and they should be instructed NOT to grant it after every ####### pitch. If the hitter steps out and time isn't called, pitcher can lob one to the catcher and it should be called a strike. Simple. No more 15 pitch innings lasting 10 minutes.
47. GGC for Sale Posted: August 17, 2014 at 10:23 AM (#4772744)
Good Lord, I've only ever heard of one of the four guys mentioned in #38. (Which itself is not ideal from the fan point of view, IMHO. I don't want it to be like football where an individual player might have such a small and specialized role that I don't even know his name.)

This. The Red Sox used 48 players last year and 26 of them were pitchers. Contrast this with 1963, were they used 31 players, 16 of them pitchers. They were a bad team in 1963 and world champs last year. I think part of this was it was harder back then to call up players from the minors, but that's still an increase in players used of roughly 50%
48. McCoy Posted: August 17, 2014 at 10:56 AM (#4772755)
Last year the Cubs used 56 players with 31 of them being pitchers. In 1879 they used two pitchers for the entire season and in only one game did the starter not complete the game!
49.  Posted: August 17, 2014 at 11:17 AM (#4772766)
1)No more than 11 pitchers on the 25-man roster. Ever.

Wait, wouldn't such a rule dictate that mean Adam Dunn would never again pitch to a batter? Why do you hate America?
50.  Posted: August 17, 2014 at 11:22 AM (#4772770)
Last year the Cubs used 56 players with 31 of them being pitchers. In 1879 they used two pitchers for the entire season and in only one game did the starter not complete the game!

In that same year, the pennant winning Providence Grays also used only two pitchers for the entire season. A 19 year old pitcher, Monte Ward, won more games than the Cubs all by himself, going 47-19 in 587 innings.
51. snapper (history's 42nd greatest monster) Posted: August 17, 2014 at 12:30 PM (#4772812)
I can support the first thing, but can't support the second. Roster construction doesn't really affect the game in front of the fans for the most part, but the second thing absolutely does, and I hate it.

Why do you hate it? I can't stand seeing a RP used for one batter.

If you can't get both righties and lefties out, you don't belong in MLB. To me, anything that reduces specialization is good for the game.

And, yes, I want the DH abolished. If you're too old, slow, fat or crippled to play the field, you don't belong in MLB either.
52.  Posted: August 17, 2014 at 12:42 PM (#4772819)
Why do you hate it? I can't stand seeing a RP used for one batter.

Because you are taking away managerial options. I don't like limiting things just because something is un-pleasing. It's fundamentally changing the concept of the game. Baseball already has a roster limit that managers have to abide by, no return to the game once they left, this provides enough strategy to make teams be wary of too much changing out of players. If you limit the number of 'pitchers' on the roster, you have already done enough to eliminate some one batter appearances, but you haven't completely eliminated the option, I prefer the game to have options.

If you can't get both righties and lefties out, you don't belong in MLB. To me, anything that reduces specialization is good for the game.

I'm the opposite, I can't stand that every team has basically 3 utility players and no speed burners or power bench bats from the right side etc. The lack of specialization leads to lower quality play.
53. snapper (history's 42nd greatest monster) Posted: August 17, 2014 at 01:15 PM (#4772841)
Because you are taking away managerial options. I don't like limiting things just because something is un-pleasing. It's fundamentally changing the concept of the game. Baseball already has a roster limit that managers have to abide by, no return to the game once they left, this provides enough strategy to make teams be wary of too much changing out of players. If you limit the number of 'pitchers' on the roster, you have already done enough to eliminate some one batter appearances, but you haven't completely eliminated the option, I prefer the game to have options.

Yeah, I don't watch baseball for the managing. The less I know about what the managers are thinking, the happier I am.

I'm the opposite, I can't stand that every team has basically 3 utility players and no speed burners or power bench bats from the right side etc. The lack of specialization leads to lower quality play.

But, that's true because of the specialization of relief pitching. I don't care for pinch-running specialists, but the lack of bench bats, and platoon bats is directly the result of 7-8 man bullpens.

Force GMs/managers to have a 6 man bench, and you'll get a lot more variety on that bench. And, none of those players, even a platoon 1B or LF, will be as specialized as a LOOGY or ROOGY.
54.  Posted: August 17, 2014 at 01:32 PM (#4772852)
But, that's true because of the specialization of relief pitching. I don't care for pinch-running specialists, but the lack of bench bats, and platoon bats is directly the result of 7-8 man bullpens.

I know that, but it's still a loss of specialization. The first part that I agreed with from the post you quoted was a suggestion of limiting the roster number of pitchers. This expands the benches a little bit, and also brings down the number of available relievers which limits the likelihood of the one batter appearances, but doesn't eliminate it as a possibility.
55. snapper (history's 42nd greatest monster) Posted: August 17, 2014 at 01:38 PM (#4772856)
I know that, but it's still a loss of specialization. The first part that I agreed with from the post you quoted was a suggestion of limiting the roster number of pitchers. This expands the benches a little bit, and also brings down the number of available relievers which limits the likelihood of the one batter appearances, but doesn't eliminate it as a possibility.

Right, but I think it's a good loss of specialization. Just like eliminating the DH is a good loss of specialization.

To me, the goal should be that any attempt at specialization should come with you having to pay the natural price for the advantage you gain.

If you want to use a LOOGY to get out a very dangerous lefty, fine, but he's going to be exposed to the RH batting next. If you want to get a big bat/no glove guy in the lineup, fine, but you're going to have to live with his glove for at least 6 or 7 innings. If you want to put an all glove/no bat guy at CF or SS, you have to let him bat 2 or 3 times. If you want to play a platoon at a position, that guy has to face RPs of the opposite hand, at least some of the time.

It's all nicely symmetric, except the LOOGY/ROOGY and DH situations, where you get specialization w/o cost.
56. nick swisher hygiene Posted: August 17, 2014 at 01:52 PM (#4772864)
I'm with snapper on this one.......basically, "in-game strategy" is pretty dull. it consists of a few simple sets of choices which baseball fans obsess over.

but that said, I'd rather see the timing rules that are already part of the game enforced before making rule changes.

to step up the pace of the game, stop the stepping out and stepping off; if that doesn't work, THEN take bigger steps.
57.  Posted: August 17, 2014 at 02:25 PM (#4772882)
Just for the record, in 2013 there were 1163 relief appearances that was one batter only. This averages out to just under 39 per team(with the most being 73 for the Giants and fewest being 16 for the Nationals.

Obviously this is happening more than it was in the past, but even as recently as 1980, teams averaged over 17 one batter faced performances per year. I support trying to reduce that number of times one batter relievers are used, but I just can't support eliminating that option.
58. snapper (history's 42nd greatest monster) Posted: August 17, 2014 at 02:28 PM (#4772884)
This averages out to just under 39 per team

So, every other game we're wasting 4-6 minutes on an OOGY. Seems like a good place to lop 3 minutes off the average game.
59.  Posted: August 17, 2014 at 03:01 PM (#4772908)
I just think that there are other ways to fix that problem than eliminating it as an option.
60. snapper (history's 42nd greatest monster) Posted: August 17, 2014 at 03:02 PM (#4772910)
I just think that there are other ways to fix that problem than eliminating it as an option.

OK, but why is that option important to baseball? Managers, for generations, basically operated as if hey didn't have the option, and baseball didn't seem to suffer.
61.  Posted: August 17, 2014 at 03:14 PM (#4772917)
I'd be fine with a rule against mid-inning pitching changes. There are already constraints: you can't pull back a pitcher once introduced. Simply extend that stipulation. It's not an unfair tactical constraint, it's just a different one.
62. snapper (history's 42nd greatest monster) Posted: August 17, 2014 at 03:15 PM (#4772920)
I'd be fine with a rule against mid-inning pitching changes. There are already constraints: you can't pull back a pitcher once introduced. Simply extend that stipulation. It's not an unfair tactical constraint, it's just a different one.

Agree about the fairness, but I think an outright ban goes too far. If a guy starts and inning and has nothing, he shouldn't be left out there to get pummeled. That's why I like the 3 batter rule. I could be sold on 4 batters.
63.  Posted: August 17, 2014 at 03:21 PM (#4772925)
OK, but why is that option important to baseball? Managers, for generations, basically operated as if hey didn't have the option, and baseball didn't seem to suffer.

Because I just don't see what the purpose is of removing options that exist within the fundamental rules. There is a cost to using a one batter faced option, and that is the loss of availability of a relief pitcher later in the game. That is sufficient penalty. If you combine it with a rule limiting teams roster size based upon makeup(no more than 11 pitchers) and it's a real detriment to using a specialist reliever. It's basically one bullet that you have to choose the best time to use.

Just like if you have a great pinch hitter, you are limited on when the best time to utilize him. I just cannot support rules that limit the between the lines management. Before the game limitations sure, equipment limitations most definitely, but not saying 'we have a special rule, specifically for this situation, simply because we don't like seeing this situation' it seems like a petty rule in my opinion.
64. PASTE, Now with Extra Pitch and Extra Stamina Posted: August 18, 2014 at 07:44 AM (#4773292)
Agree about the fairness, but I think an outright ban goes too far. If a guy starts and inning and has nothing, he shouldn't be left out there to get pummeled. That's why I like the 3 batter rule. I could be sold on 4 batters.

Ah, that's my cue to point out yet again how well a nice, simple rule that the pitcher can be removed midinning only once he's allowed a run of his own would be.

There's never any problem, in my mind, with changing pitchers between innings--so if you want to put a LOOGY in there with two outs in the inning to strike Pedro Alvarez out and then change to a real pitcher for the next inning, that's no problem. You're not interrupting the action.

You must be Registered and Logged In to post comments.

<< Back to main

### News

Old-School Newsstand

### Support BBTF

Thanks to
James Kannengieser
for his generous support.

### Bookmarks

You must be logged in to view your Bookmarks.

### Hot Topics

NewsblogMLBTR: Dodgers Acquire Curtis Granderson
(28 - 1:54am, Aug 21)
Last: akrasian

NewsblogOTP 14 August 2014: The American Pastimes of Rock ’n’ Roll, Baseball and Poetry
(2843 - 1:53am, Aug 21)
Last: Sebastian

NewsblogCC Sabathia went from lowest low to highest high in two weeks | New York Post
(15 - 12:35am, Aug 21)
Last: Lars6788

NewsblogJeffrey Loria has ‘no sad feelings whatsoever’ about selling the Marlins - Sun Sentinel
(9 - 12:07am, Aug 21)
Last: LA Podcasting Hombre of Anaheim

NewsblogMLB umpire protest: Umps unlikely to get sympathy from fans | SI.com
(18 - 11:37pm, Aug 20)
Last: Robert S.

NewsblogAin't no WEEKEND OMNICHATTER until someone gets punched in the face, for August 19-20, 2017
(257 - 10:06pm, Aug 20)
Last: What did Billy Ripken have against ElRoy Face?

NewsblogDaren Willman:"Breaking down the quality of contact chart (exit velocity / launch angle)
(3 - 9:53pm, Aug 20)
Last: Pat Rapper's Delight (as quoted on MLB Network)

NewsblogReusse: Twins' altercation in 1967? Part about a gun on team bus was avoided
(1 - 9:34pm, Aug 20)
Last: winnipegwhip

(1809 - 9:29pm, Aug 20)
Last: 2 dudes are better than STIGGLES; i'm both of em

Sox TherapyThe Gauntlet
(77 - 8:46pm, Aug 20)
Last: the Hugh Jorgan returns

NewsblogJake Arrieta is pitching like an ace again — just in time for free agency - Chicago Tribune
(9 - 7:57pm, Aug 20)
Last: Dag Nabbit at ExactlyAsOld.com

Gonfalon CubsNow or Never
(106 - 7:06pm, Aug 20)
Last: Andere Richtingen

NewsblogOT: New Season August 2017 Soccer Thread
(159 - 7:00pm, Aug 20)
Last: SPICEY WITH A SIDE OF BEER ON A BABYYYYYYY

NewsblogArthur: Buster Posey Has Quietly Become A Lock For Cooperstown
(114 - 6:18pm, Aug 20)
Last: cardsfanboy

NewsblogGregg Olson: Orioles' emphasis on talented, structured bullpens is nothing new - Baltimore Sun
(4 - 5:23pm, Aug 20)
Last: Sunday silence

Page rendered in 0.4759 seconds
47 querie(s) executed