Baseball for the Thinking Fan

Login | Register | Feedback

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

Monday, May 21, 2018

Posnanski: Modern Romonce

Is this the future of baseball?

Now, let’s say right off that I’m quite sure this Romo strategy is a winning one. Many of us have been talking about it for years; Joe Sheehan most prominently. Teams rarely use fewer than three pitchers in a game; they average about four pitchers per game. There have been 17 complete games all season. Seventeen. The concept of a starting pitcher as we knew it in the 1970s and 1980s and even 1990s has been blown to smithereens, so the question has been obvious for a while now: When will a team just finish the job and reinvent the whole pitching order of a game?

Baldrick Posted: May 21, 2018 at 09:39 AM | 117 comment(s) Login to Bookmark
  Tags: game theory, pitching, posnanski, traditions

Reader Comments and Retorts

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.

Page 1 of 2 pages  1 2 > 
   1. eric Posted: May 21, 2018 at 01:10 PM (#5676505)
I was excited by the way Romo was being used. It reflects a potential complete overhaul of the traditional baseball deployment of resources. If such a usage pattern sticks, it will be exciting to see where and how the game develops.
   2. snapper (history's 42nd greatest monster) Posted: May 21, 2018 at 01:13 PM (#5676508)
I don't like it. Much like I don't really like the shifts.

I prefer teams to play it straight and let the talent determine the outcome. Not excited by managerial chicanery having more impact on the game.

I like analytics to help understand the game. In most cases, when they influence how the game is played, they make it worse.

The last thing we need is more anonymous short relievers racking up Ks.
   3. Panic Posted: May 21, 2018 at 01:29 PM (#5676526)
I don't like it. Much like I don't really like the shifts.

I prefer teams to play it straight and let the talent determine the outcome. Not excited by managerial chicanery having more impact on the game.

I like analytics to help understand the game. In most cases, when they influence how the game is played, they make it worse.

The last thing we need is more anonymous short relievers racking up Ks.


Despite your preferences, talent still determines the outcome as much as it ever has. This does not change that. It just provides a new way to deploy the talent.

And some of these so-called "anonymous relievers" are still really good pitchers. I didn't know who Josh Hader was before this year, but I will stop and watch him pitch any day of the week, and I'll get just as much enjoyment out of it as I do any other "known" pitcher.
   4. snapper (history's 42nd greatest monster) Posted: May 21, 2018 at 01:34 PM (#5676529)
And some of these so-called "anonymous relievers" are still really good pitchers. I didn't know who Josh Hader was before this year, but I will stop and watch him pitch any day of the week, and I'll get just as much enjoyment out of it as I do any other "known" pitcher.

They're not "really good pitchers". If they were expected to throw 7 innings, most of them would be terrible.

To your example, Josh Hader up up a 5.37 ERA/7.05 FIP is 12 AAA starts last year. That's not a very good pitcher.

They are poor to mediocre pitchers who are deployed in a way (short, maximum effort stints, where they never face the same batter twice) that allows them to be very effective.

If baseball ever moves to a series of 1-3 IP pitchers with no starters, I'm done with the game.
   5. SoSH U at work Posted: May 21, 2018 at 01:46 PM (#5676549)
Now, let’s say right off that I’m quite sure this Romo strategy is a winning one.


I'll need a little more evidence than a split with the Angels to be convinced of this. Sergio Romo is a limited pitcher. Sometimes he looks good (like his 30 IP with Tampa last year). Sometimes he doesn't (like his 25 with LAD).

Moreover, just redistributing which inning you give to your short man and which one you give to your "starter" doesn't really strike me as something that should result in massive changes to win expectancy. I don't know where the value is coming from.

Now, a move toward all short stints could change things, but I'm almost positive there would be some long-term fallout from such a move.

   6. Panic Posted: May 21, 2018 at 01:50 PM (#5676553)
And if baseball starts using AAA stats and ignores accomplishments against the best players in baseball, I'm done with the game.
   7. snapper (history's 42nd greatest monster) Posted: May 21, 2018 at 01:54 PM (#5676555)
And if baseball starts using AAA stats and ignores accomplishments against the best players in baseball, I'm done with the game.

What does this even mean? Baseball has been using minor league stats to evaluate players for 100 years. Hader has an ERA over 5 in two years at AAA. That means he's not a very good pitcher.

He can succeed in the short reliever role, but that does not make him a very good pitcher. There basically has never been agood SP who didn't excel in short relief. The list of the 100 best pitchers in baseball contains 100 starters.
   8. Tin Angel Posted: May 21, 2018 at 01:56 PM (#5676557)
I don't know where the value is coming from.


That's what I don't get either. How is this any different than saving your "best" guy for the 9th since it's (incorrectly) seen as more high leverage than the 7th or whatever?
   9. eric Posted: May 21, 2018 at 02:04 PM (#5676567)
The value is coming from getting to pick your matchups at the beginning of the game. Romo mauled the top of the Angels order, putting the Rays in that much better position from the start.

The value also comes from your "starter" having to face the tough hitters fewer times in the game. If the first guy he faces is really batter #6, then he can be pushed to a third time through the order to face the 6-9 hitters, rather than having to face 1-4. Then the next, rested, short reliever can plow through #1-4 in the seventh or eighth.

   10. Ithaca2323 Posted: May 21, 2018 at 02:04 PM (#5676569)
Despite your preferences, talent still determines the outcome as much as it ever has.


How effective you are in a specific role is an entirely different question as to your talent level. This is especially true for pitchers

Trevor Hoffman was effective in his role, but Mike Mussina was an inarguably more talented pitcher. That's why they got put in those roles to begin with.
   11. jacksone (AKA It's OK...) Posted: May 21, 2018 at 02:09 PM (#5676575)
That's what I don't get either. How is this any different than saving your "best" guy for the 9th since it's (incorrectly) seen as more high leverage than the 7th or whatever?


In the first game I believe the value gained was in the right handed Romo pitching to the right handed top of the Angels order. That allowed the lefty Yarbrough to have an easier start to his day. Not so sure what the benefit was in the 2nd game, other than "hey it worked last night, might as well..."
   12. Moses Taylor, aka Hambone Fakenameington Posted: May 21, 2018 at 02:11 PM (#5676578)
Hader has an ERA over 5 in two years at AAA. That means he's not a very good pitcher.

He can succeed in the short reliever role, but that does not make him a very good pitcher. There basically has never been agood SP who didn't excel in short relief. The list of the 100 best pitchers in baseball contains 100 starters.


This is dumb, I mean, I get what your overall point is trying to be, but this is a bad example.

Hader pitched 121 total innings in AAA, as a 22/23 year old. He played in the PCL, in Colorado Springs, which IIRC, is going to inflate his numbers a little. He's now pitched 75 innings in the bigs as a 23/24 year old. Hader is also not your typical short reliever, in that he's pitching well over an inning each appearance.

The point about facing guys multiple times applies, but also, he's still pitching at a level significantly above any of your standard middle relief mirage types. So, still, he's probably a bad example. He's also a more interesting case study than Romo randomly starting games.
   13. Slivers of Maranville descends into chaos (SdeB) Posted: May 21, 2018 at 02:11 PM (#5676579)

They're not "really good pitchers". If they were expected to throw 7 innings, most of them would be terrible.


And if Kershaw were expected to throw 30 innings per game, he'd be terrible too. What does it matter? Is a marathon runner automatically better than a sprinter?
   14. snapper (history's 42nd greatest monster) Posted: May 21, 2018 at 02:18 PM (#5676587)
And if Kershaw were expected to throw 30 innings per game, he'd be terrible too. What does it matter? Is a marathon runner automatically better than a sprinter?

If you have to cover 26 miles he is. Teams need to pitch 8+ innings every day. A guy who can pitch effectively 5-7 of them is automatically more valuable than one who can only pitch 1-2.

   15. jacksone (AKA It's OK...) Posted: May 21, 2018 at 02:31 PM (#5676600)
A guy who can pitch effectively 5-7 of them is automatically more valuable than one who can only pitch 1-2.


Who would you rather have on your team, 2017 Rick Porcello with 203.1 innings of 99 ERA+ or 2017 Craig Kimbrel with 69 IP at 322 ERA+?
   16. Jose is an Absurd Doubles Machine Posted: May 21, 2018 at 02:33 PM (#5676601)
If you have to cover 26 miles he is. Teams need to pitch 8+ innings every day. A guy who can pitch effectively 5-7 of them is automatically more valuable than one who can only pitch 1-2.


Wakefield or Rivera? One of them is on the shortlist of my all time favorite players but I'd still rather have the one who pitched fewer innings.

That said I agree with you more than I don't. I'm not convinced this is a winning strategy. As you note you have to get through 9 innings to win a game. I think trying to do this over a full season is folly. Pitchers will break down and just on a game to game basis you are going to be finding the one guy that doesn't have it all too often.
   17. Ithaca2323 Posted: May 21, 2018 at 02:33 PM (#5676606)
And if Kershaw were expected to throw 30 innings per game, he'd be terrible too. What does it matter?


Starting pitchers can do what relievers do. In fact, they do do what relievers do, often in their team's most important games of the season. Kershaw, Verlander, and Scherzer all came out of the bullpen in the playoffs last year. Bumgarner in 2014. Mussina in 2003. Johnson in 2001. Pedro in 1999, etc....

Relief pitchers cannot do what starters do. That's why they're made into relievers.
   18. snapper (history's 42nd greatest monster) Posted: May 21, 2018 at 02:45 PM (#5676613)
Who would you rather have on your team, 2017 Rick Porcello with 203.1 innings of 99 ERA+ or 2017 Craig Kimbrel with 69 IP at 322 ERA+?

Porcello. He also makes quite a bit more than Kimbrel.

It's hard to find starting pitchers who can throw 200 IP. It's easy to fine relievers who can throw 60 great innings. Most teams find 2-4 of them every year.
   19. Jess Franco Posted: May 21, 2018 at 02:48 PM (#5676621)
Pitchers will break down and just on a game to game basis you are going to be finding the one guy that doesn't have it all too often.

Look at the Dodgers. It's not that hard to find a bunch of pretty good interchangeable relievers. Other teams have already followed suit. You can find these guys much easier, and much less expensively, than a Kershaw or a Scherzer.

The novelty is interesting, but the further economic rationalization of the game won't be so much. A team like Tampa will pinch pennies and still field a .500 team full of mediocrities who can neutralize the game's best hitters.

There's no stopping progress, Snapper.
   20. SoSH U at work Posted: May 21, 2018 at 02:55 PM (#5676631)
The value is coming from getting to pick your matchups at the beginning of the game. Romo mauled the top of the Angels order, putting the Rays in that much better position from the start.


Saturday he did. At the end of the day, he's just as likely to repeat his 2017 first half as his 2017 second half, and put you in a hole against that tough top of the lineup. Because he's simply not a great pitcher, the occasional small-sample bursts and reliever-advantage stretches notwithstanding.
   21. snapper (history's 42nd greatest monster) Posted: May 21, 2018 at 02:56 PM (#5676632)

There's no stopping progress, Snapper.


Well, it's not progress if it makes the game aesthetically worse. And, MLB can stop it if they want to.

Limit teams to 10 pitchers on the roster. Boom, you now needs SPs who can give you 6-7 IP. There's no possibility of cobbling together enough innings from relievers.
   22. Ithaca2323 Posted: May 21, 2018 at 02:57 PM (#5676634)
Whenever I think of the starter/reliever debate, I remember that the signature performance that defines Mariano Rivera as the Greatest Closer Ever (three shutout innings in Game 7 of the 2003 ALCS) was a lesser version of the same trick that Mike Mussina—regarded by many as perhaps the 7th or 8th best starter of his own generation*—pulled off like, two hours earlier in his first relief appearance ever.

*As evidenced by the fact that five of them were elected to the Hall before him, and two more would have been had it not been for off-the-field issues
   23. Misirlou doesn't live in the restaurant Posted: May 21, 2018 at 03:11 PM (#5676644)
Who would you rather have on your team, 2017 Rick Porcello with 203.1 innings of 99 ERA+ or 2017 Craig Kimbrel with 69 IP at 322 ERA+?


It depends? Who is Porcello replacing? Who is Kimbrel replacing?

It's easy to fine relievers who can throw 60 great innings. Most teams find 2-4 of them every year.


Not consistently. Kimbrel's value is being consistently effective, and frequently outstanding. Hector Rondon was great, until he wasn't. Justin Wilson's ERA+ are 172, 86, 132, 103, 133, 114. Carlos Marmol had the same pattern. Justin Grimm was lights out in 2015. the rest of his career he was pretty worthless. Brian Dunsing has been great the last 2 years. He was useless to pedestrian before that, and I expect him to go back before his career is over. Neil Ramirez, like Justin Grimm, was great for exactly 1 year.

And that's just Cub examples from the last 5 years.
   24. Nasty Nate Posted: May 21, 2018 at 03:17 PM (#5676653)
Very few teams last year got even just 2 relievers who threw 60 great innings, never mind 2-4 for most teams, unless "great" means 130 ERA+.
   25. SoSH U at work Posted: May 21, 2018 at 03:20 PM (#5676654)
Whenever I think of the starter/reliever debate, I remember that the signature performance that defines Mariano Rivera as the Greatest Closer Ever (three shutout innings in Game 7 of the 2003 ALCS) was a lesser version of the same trick that Mike Mussina—regarded by many as perhaps, the 7th or 8th best starter of his own generation*—pulled off like, two hours earlier.


On the other hand, it was much better than the trick pulled off like three hours earlier by one of the top four starters of his own generation.

And I'm a little confused why Mo's 3 IP, 0 Run, 2 Hit, 3K, 0 Walk outing was a lesser version of Mussina's 3 IP, 0 RUn 2 Hit, 3 K, 0 Walk outing? Looks like an eerily identical version to me.

   26. snapper (history's 42nd greatest monster) Posted: May 21, 2018 at 03:20 PM (#5676655)
Not consistently. Kimbrel's value is being consistently effective, and frequently outstanding.

I'm not sure that consistency is all that valuable when every team is employing 8 RPs. You just need 2-3 of them to be really good every year. It doesn't have to be the same 2-3. As long as you don't pigeon hole yourself into sticking with your "closer".
   27. snapper (history's 42nd greatest monster) Posted: May 21, 2018 at 03:24 PM (#5676658)
Very few teams last year got even just 2 relievers who threw 60 great innings, never mind 2-4 for most teams, unless "great" means 130 ERA+.

A 130 ERA+ is pretty great.

Last year there were 82 RPs with >= 50 IP, and a ERA- of 80 or less. 56 with an ERA- of 70 or less.
   28. SoSH U at work Posted: May 21, 2018 at 03:24 PM (#5676659)

I'm not sure that consistency is all that valuable when every team is employing 8 RPs. You just need 2-3 of them to be really good every year.


It wouldn't be valuable if you knew which relievers were going to suck before they sucked, and thus kept the ball out of their hands. But that's not the way it works.

   29. Misirlou doesn't live in the restaurant Posted: May 21, 2018 at 03:25 PM (#5676661)
I'm not sure that consistency is all that valuable when every team is employing 8 RPs. You just need 2-3 of them to be really good every year. It doesn't have to be the same 2-3. As long as you don't pigeon hole yourself into sticking with your "closer".


Yeah, it doesn't work that way. If a guy is inconsistent year to year, it follows that he is likely inconsistent day to day. Look at Justin Wilson. He's been outstanding in his career, and he's been lousy. He has outstanding outings, and then he walks 3 in the bottom of the 11th. He walked Billy "Freakin" Hamilton with the bases loaded. Your Kimbrel's rarely do that.
   30. SandyRiver Posted: May 21, 2018 at 03:26 PM (#5676662)
Who would you rather have on your team, 2017 Rick Porcello with 203.1 innings of 99 ERA+ or 2017 Craig Kimbrel with 69 IP at 322 ERA+?

Of course, this would not be an even exchange, as one would need 4 "Kimbrels", one that you have already plus 3 more to eat Porcello's average 6-inning start, and that's allowing numbers 2-4 having to pitch 2 innings each. I could see the philosophy resulting in teams having perhaps 15 pitchers, meaning (assuming the 25-man limit stays) that the rest of the roster would be the 8 position players plus a spare C and Brock Holt. I'd rather not. (And there's a bit of Russian roulette in trotting out a whole sequence of RPs; one Barnes/Hembree on a bad day and you're down a quick 3-4 runs.)
   31. Nasty Nate Posted: May 21, 2018 at 03:28 PM (#5676663)
Last year there were 82 RPs with >= 50 IP, and a ERA- of 80 or less. 56 with an ERA- of 70 or less.
Out of curiosity, how many at 60 innings like you first said?
   32. Nasty Nate Posted: May 21, 2018 at 03:29 PM (#5676665)
And I'm a little confused why Mo's 3 IP, 0 Run, 2 Hit, 3K, 0 Walk outing was a lesser version of Mussina's 3 IP, 0 RUn 2 Hit, 3 K, 0 Walk outing? Looks like an eerily identical version to me.
You weren't asking me, but Mussina came in with first and third and no outs and stranded the runners, while Mo came in to start an inning.
   33. Misirlou doesn't live in the restaurant Posted: May 21, 2018 at 03:31 PM (#5676668)
51.
   34. Ziggy's screen name Posted: May 21, 2018 at 03:32 PM (#5676669)
Stick in the mud is stick in the mud. Teams trying new stuff is fun. Even better if it turns out to be clever and so gives them an advantage (we'll see about that; I'm not yet sold on this being a good idea). Plus, Sergio Romo can't vulture any wins anymore, since he started the game and didn't go five innings.

And regarding shifts, they're nothing new. From the 1860 Beadle's Dime Base Ball Player: "[The shortstop's] position is generally in the center of the triangle formed by the second and third bases and the pitcher's position, but he should change it according to his knowledge of the striker's style of batting." Here.
   35. SoSH U at work Posted: May 21, 2018 at 03:33 PM (#5676670)
You weren't asking me, but Mussina came in with first and third and no outs and stranded the runners, while Mo came in to start an inning.


I guess that's fair. I've tried to block out the details from that nightmare.
   36. PreservedFish Posted: May 21, 2018 at 03:42 PM (#5676677)
I love it. While I agree with most of the community about the shame that is the long the decline in SP innings and concomitant increase in RP importance, I also love the creativity on display in this move.
   37. snapper (history's 42nd greatest monster) Posted: May 21, 2018 at 03:43 PM (#5676678)
Out of curiosity, how many at 60 innings like you first said?

58 at 80 or better. 42 at 70 or better.

   38. SoSH U at work Posted: May 21, 2018 at 03:48 PM (#5676683)
Stick in the mud is stick in the mud. Teams trying new stuff is fun. Even better if it turns out to be clever and so gives them an advantage.


For the record, I have no problem with them doing it. I just question Joe's conclusion that it is a winning tactic. It seems more of a deck chairs reshuffle than a groundbreaking strategy.
   39. Misirlou doesn't live in the restaurant Posted: May 21, 2018 at 03:49 PM (#5676684)
Out of curiosity, how many at 60 innings like you first said?

58 at 80 or better. 42 at 70 or better.


What does this mean?

Last year there were 74 relievers with a 130 ERA+ of better in 50+ IP, 51 in 60+ IP, 19 in 70+, and 5 in 80+
   40. Don Malcolm Posted: May 21, 2018 at 03:50 PM (#5676686)
It is useful for teams to experiment with these things...first and foremost, because if a team actually stayed with a counterintuitive strategy for a sufficient length of time, we'd get some real world data with which to work. Will the results ultimately reveal an "edge" for some unorthodox practice? The odds are probably against it, but why shouldn't teams try it?

Joel Sherman gives Brian Kenny credit for what he calls the "opener"--but people who haven't let the "metrics revolution" glaze over their sense of history will recall that Earnshaw Cook dreamed up this idea in the early 60s as a way to get an extra at-bat for a hitter in the #9 slot early in the game. Naturally he had no inkling that at least one league would do away with that particular strategic opportunity. Today, such an approach can be for a variety of reasons--economic (what's the cheapest agglomeration of pitchers used in the most unorthodox way that creates acceptable results), strategic (Sherman cites the idea of countering the top of the order with a pitcher with better potential performance for that individual inning), or tactical (if one wants to stretch fewer starters through a particular patch of the season, for whatever conceivable reason).

It would be interesting if a team like the Marlins or the Reds tried to something like this, only they made an effort to do so via the Earnshaw Cook variation, with regular starters for 1-3-5, relievers capable of pitching a two-IP "opening" in 2-4, and a fourth starter who gets a start in lieu of the reliever every second time through. So basically a three-and-a-half man rotation with a twist. It has some of the flavor of pre-60s baseball but with wildly different usage increments within an individual game than the old days.

Note that the Rays seem also to have made Matt Andriese into some kind of "spot starter" after having been used in a more traditional starting role previously. (OTOH, look at his 2016 games logs to see an odd usage pattern that seemed to throw him off. He's made two starts and been pulled in the fourth inning--yet another variation that might cluster around the "don't let lesser pitchers go more than 1 (2) times through the rotation" theory.

BTW, Andriese followed Romo to the mound yesterday--and took the loss despite giving up no hits in his two innings of work.
   41. Nasty Nate Posted: May 21, 2018 at 03:50 PM (#5676688)
What does this mean?

Last year there were 51 relievers with a 130 ERA+ of better in 60+ IP, 74 in 50+ IP
I think he was doing some ERA- threshold, not ERA+.
   42. snapper (history's 42nd greatest monster) Posted: May 21, 2018 at 03:52 PM (#5676690)
Last year there were 51 relievers with a 130 ERA+ of better in 60+ IP, 74 in 50+ IP

ERA-, as Nate says. The Fangraphs stats filter is easier to use. And ERA+ is a crappy metric anyway. You want to ask, "How much better than the league was pitcher X?", not, "How much worse than pitcher X was the league?".
   43. Misirlou doesn't live in the restaurant Posted: May 21, 2018 at 03:54 PM (#5676694)
The 5 who did it in 80 IP bolsters my point. Petit led with 91 IP. Last year was his only good year. Chris Rusin was outstanding with a 190 in 80 IP. This year he's at 69. Alex Claudio was at 188 least year. This year so far 96.
   44. PreservedFish Posted: May 21, 2018 at 03:59 PM (#5676697)
Surprised to see snapper engage in this fallacy. A lot of these great relief performances are just sample size flukes - you cannot predict or plan around them.
   45. snapper (history's 42nd greatest monster) Posted: May 21, 2018 at 04:12 PM (#5676707)
Surprised to see snapper engage in this fallacy. A lot of these great relief performances are just sample size flukes - you cannot predict or plan around them.

You don't have to if you're flexible in your bullpen management. Throw enough live arms at the pen, and you'll end up with enough high level performances to get you through the season.

And it's not like those "consistent" relievers are very consistent. Betances was other wordly for several years, but by last October he was unusable when he was needed.

   46. Slivers of Maranville descends into chaos (SdeB) Posted: May 21, 2018 at 04:16 PM (#5676712)

If you have to cover 26 miles he is. Teams need to pitch 8+ innings every day. A guy who can pitch effectively 5-7 of them is automatically more valuable than one who can only pitch 1-2.


More valuable, yes, but not necessarily better.


Starting pitchers can do what relievers do. In fact, they do do what relievers do, often in their team's most important games of the season. Kershaw, Verlander, and Scherzer all came out of the bullpen in the playoffs last year. Bumgarner in 2014. Mussina in 2003. Johnson in 2001. Pedro in 1999, etc....


And Tampa doesn't have a team full of Bumgarners and Pedros, so they're trying to use the pitchers they have more effectively. I still don't know why this is such a travesty.
   47. Howie Menckel Posted: May 21, 2018 at 04:16 PM (#5676713)
the problem with Kimbrel and the 322 ERA+ is that there isn't much, if any, additional value compared to a 150 ERA+.

they have one job - to protect a lead in the 9th inning of 1, 2, or 3 runs. both closers above might well wind up with about the same success rate - 38 out of 41 or 42, let's say. so how did Kimbrel striking out so many more guys help the Red Sox win more games? hurting the self-esteem of the opposing batters?

sure, for predictive value, the 322 guy is more likely to excel the next season - but that's a different question.

meanwhile, it's amazing that we're trying the RP in the 1st inning - arguably a minor advantage, as explained above - but managers still prefer to have their best RP throw in the 9th in a 4-1 game rather than a 2-2 game. I mean, it doesn't take an Ivy League mathematics professor to figure out that inanity.
   48. Nasty Nate Posted: May 21, 2018 at 04:25 PM (#5676719)
the problem with Kimbrel and the 322 ERA+ is that there isn't much, if any, additional value compared to a 150 ERA+.

they have one job - to protect a lead in the 9th inning of 1, 2, or 3 runs. both closers above might well wind up with about the same success rate - 38 out of 41 or 42, let's say. so how did Kimbrel striking out so many more guys help the Red Sox win more games? hurting the self-esteem of the opposing batters?
This is a good point, but you are overstating it. The lower ERA pitcher is likely to have a better success rate. And even for closers with the same success rate, all blown saves don't hurt the team's chances equally. Furthermore the performance in the other 20 (non-save) innings matters, especially the tie games.
   49. Don Malcolm Posted: May 21, 2018 at 04:26 PM (#5676721)
Digging further into the Rays stats, they've also used reliever Andrew Kittredge as a starter three times, beginning with the third game of the year (3/31) when his usage (3.1 IP) would not have raised any eyebrows.

They started him again on 4/8 against the Red Sox...he went 2 IP, relieved by Ryan Yarbrough.

And one more time on 5/4 against the Blue Jays...again, 2 IP, relieved by Yarbrough.

Kittredge has since been sent to the minors: while he's been a reasonably effective "opener" (3.68 ERA), he's been a woefully ineffective reliever (14.46 ERA). Though, yes, some of this is sample size--he was shipped out after being left in to give up 6 runs against the Orioles when he entered the game in the fourth inning. He was brought in the middle of the inning, so probably didn't have as much time to warm up as would presumably be the case when he "opened."
   50. snapper (history's 42nd greatest monster) Posted: May 21, 2018 at 04:31 PM (#5676724)
And Tampa doesn't have a team full of Bumgarners and Pedros, so they're trying to use the pitchers they have more effectively. I still don't know why this is such a travesty.

I don't think it's a travesty. I just think that, outside of the rare occurrence where you have a RP who's great against RH, but sucks against LH, and the other team has three RH to start the game, I don't really see the advantage.
   51. My name is Votto, and I love to get blotto Posted: May 21, 2018 at 04:33 PM (#5676726)
And ERA+ is a crappy metric anyway.


honest question, but why?

   52. Barry`s_Lazy_Boy Posted: May 21, 2018 at 04:39 PM (#5676732)
I love seeing the game evolve.

   53. snapper (history's 42nd greatest monster) Posted: May 21, 2018 at 04:40 PM (#5676735)
honest question, but why?

Because it measure the league ERA as a % of the pitcher's ERA. So a 3.00 ERA in a 4.00 ERA league is expressed as a 133 ERA+. But, it's not 33% better than league average, it's 25% better than league average, which is what ERA- shows at 75.

The problem become worse at the extremes. A 1.50 ERA pitcher in the same context will show a 266 ERA+. But he's not 133% better than the league, he's 62.5% better than the league.
   54. Baldrick Posted: May 21, 2018 at 05:03 PM (#5676751)
The problem become worse at the extremes. A 1.50 ERA pitcher in the same context will show a 266 ERA+. But he's not 133% better than the league, he's 62.5% better than the league.

This is, of course, a critical distinction, as represented in the following anecdote:

"Boy," I said to my friend, "this pitcher sure is great. "Can you believe he's 133% better than the league?"
"Gosh," my friend said. "I remember when I was a kid watching Koufax pitch and thinking that there would never be another guy who was 133% better than the league."
"You idiots!" a man yelled from behind us. "He's only 62.5% better than the league!"
   55. cardsfanboy Posted: May 21, 2018 at 05:08 PM (#5676753)
Moreover, just redistributing which inning you give to your short man and which one you give to your "starter" doesn't really strike me as something that should result in massive changes to win expectancy. I don't know where the value is coming from.


I think the value is that you can use a short reliever in the inning that has the most scoring, allowing him to help reduce some of that scoring. Another advantage is that it takes the best hitters on the other team and allows the lineup rotation another third of the way through before your starter has to face them a third and/or a fourth time. I can see some value in both of these.
   56. Baldrick Posted: May 21, 2018 at 05:09 PM (#5676754)
With less snark: ERA+ and ERA- are literally the same statistics. You can derive one from the other, and vice versa. Neither is 'better' or 'worse' than the other. It's pure aesthetics. 1 is 50% of 2. But 2 is 200% of 1. Which is more 'accurate'? Answer: they're both accurate because that's how numbers work.
   57. cardsfanboy Posted: May 21, 2018 at 05:22 PM (#5676758)
With less snark: ERA+ and ERA- are literally the same statistics. You can derive one from the other, and vice versa. Neither is 'better' or 'worse' than the other. It's pure aesthetics. 1 is 50% of 2. But 2 is 200% of 1. Which is more 'accurate'? Answer: they're both accurate because that's how numbers work.


They are not the same statistic, but ERA+ makes the extremes appear to be more extreme than they really are. ERA- measures in the same direction that era does, so it's more intuitive and matches...and of course you cannot ever be below zero, just like real era. Mariano Rivera in 2005 had an era of 1.38, had an era+ of 308 and an era - of 32... I guess it's a matter of aesthetics, but being 68% better than the league sounds better than the league was 104% worse than him.
   58. Tin Angel Posted: May 21, 2018 at 05:26 PM (#5676762)
I think the value is that you can use a short reliever in the inning that has the most scoring, allowing him to help reduce some of that scoring.


So, if that's the case, it only makes sense to do this if you think the reliever you use is better than your scheduled starting pitcher for that day.
   59. Barry`s_Lazy_Boy Posted: May 21, 2018 at 05:31 PM (#5676768)
So, if that's the case, it only makes sense to do this if you think the reliever you use is better than your scheduled starting pitcher for that day.


The first Romo start was a perfect use of the strategy. They had a lefty "starter" slated, but the Angels have a righty-heavy top of the order. Using Romo to start, and allowing the lefty to appear in the second improved the matchups for the Rays.

I agree with others that the second start seemed to be a case of "it worked yesterday".
   60. PreservedFish Posted: May 21, 2018 at 05:37 PM (#5676773)
I think there is an excellent argument for using a reliever to open the game even without special matchup considerations. In fact, I've argued as much right here:

11. PreservedFish Posted: November 24, 2015 at 01:25 PM (#5101447)
I think you should begin most games with a reliever facing the top of the order. One of your guys that throws the ball real hard. Maybe your third best reliever. In reality, he's probably a 3.25 ERA guy and thus better than most of your starters over a single inning...

You would then generally switch to the starter in the 2nd inning. When you get to the dreaded "third time through the lineup," it's the opponent's lesser hitters that are getting that important good third look first. The better hitters would rarely get a third crack at any pitcher.

This way you don't reduce the number of innings that your starter throws, but you do hopefully make his job easier.

This would be especially useful for those typical 3rd starter types, the guys that are durable with mediocre stuff, the Jeremy Guthries and Jason Vargases.
   61. An Athletic in Powderhorn™ Posted: May 21, 2018 at 05:38 PM (#5676774)
honest question, but why?
This article explains it. The short version:

"The denominator of ERA is innings pitched; the denominator of ERA+ is earned runs. This is a point that many people have overlooked (embarrassingly, I made this mistake myself in 2007 , although I had come to my senses by early 2008 ). The practical consequence is that if you are averaging ERA+ across samples, you cannot weight it by innings pitched--you must weight it by earned runs."

There are other reasons, but I think of that as the most important one.
   62. Howie Menckel Posted: May 21, 2018 at 05:40 PM (#5676776)
The lower ERA pitcher is likely to have a better success rate. And even for closers with the same success rate, all blown saves don't hurt the team's chances equally. Furthermore the performance in the other 20 (non-save) innings matters, especially the tie games.

well, not only would I claim that the SV percentage for both historically has been very close, the details of the games would matter, too. how many 1-run games? how many against a stacked lineup? how many in bad-weather conditions?

I think we have found that effective relievers close out games pretty equally regardless of much further dominant they become. but I could be wrong.
   63. SoSH U at work Posted: May 21, 2018 at 05:50 PM (#5676783)
You would then generally switch to the starter in the 2nd inning. When you get to the dreaded "third time through the lineup," it's the opponent's lesser hitters that are getting that important good third look first. The better hitters would rarely get a third crack at any pitcher.


Or, routinely facing a team's top three hitters exposes that third-best bullpen arm for the lesser pitcher we already know him to be.

   64. villageidiom Posted: May 21, 2018 at 05:54 PM (#5676785)
I'm adding nothing to the current discussion; rather, I'm just jumping in to say a Don Malcolm sighting is always a good thing 'round here.
   65. PreservedFish Posted: May 21, 2018 at 05:54 PM (#5676786)
Or, routinely facing a team's top three hitters exposes that third-best bullpen arm for the lesser pitcher we already know him to be.

This doesn't seem responsive to me. If we expect the reliever to be better than the starter over the course of a single inning, then he's better.
   66. Randomly Fluctuating Defensive Metric Posted: May 21, 2018 at 05:55 PM (#5676787)
It makes no sense to try getting five good innings every five days from mediocre fourth and fifth starters when pitchers are generally more effective the first time through the order. Teams deep in quality starting pitching can afford to take the risk. But I see the #4 and #5 starters being phased out of rotations in the next ten years. They will be replaced by tandem swingmen and one batter specialists. During the most optimistic moments of spring training, it'll be a common observation that, "these guys are good enough to have five starters if they wanted, like the old days."
   67. SoSH U at work Posted: May 21, 2018 at 06:06 PM (#5676793)
This doesn't seem responsive to me. If we expect the reliever to be better than the starter over the course of a single inning, then he's better.


We expect a reliever to be better than a starter over the course of a single inning because relieving is easier. If you make relieving harder, like forcing your No. 3 reliever to routinely face the other team's best hitters, then we might not expect him to be better than the starter anymore.

These guys are inferior pitchers to the starters, who have better stats because they have easier jobs (and small sample sizes).


It makes no sense to try getting five good innings every five days from mediocre fourth and fifth starters when pitchers are generally more effective the first time through the order. Teams deep in quality starting pitching can afford to take the risk. But I see the #4 and #5 starters being phased out of rotations in the next ten years. They will be replaced by tandem swingmen and one batter specialists. During the most optimistic moments of spring training, it'll be a common observation that, "these guys are good enough to have five starters if they wanted, like the old days."



Possibly. We don't know what effect such deployment will have on long-term effectiveness of these relievers.
   68. Nasty Nate Posted: May 21, 2018 at 06:14 PM (#5676798)
I think we have found that effective relievers close out games pretty equally regardless of much further dominant they become. but I could be wrong.
I think you're right, but I just wanted to get in my nitpick that 2 closers with identical save percentages can have different values even assuming identical orthodox usages.
   69. PreservedFish Posted: May 21, 2018 at 06:15 PM (#5676799)
If you make relieving harder, like forcing your No. 3 reliever to routinely face the other team's best hitters, then we might not expect him to be better than the starter anymore.


Yes, it would hurt the reliever's numbers. And boost the starter's. But that's superficial. The reliever would still be better. Unless we are wildly incorrect about why relief pitchers have better numbers than starters, and it's actually some sort of unaccountable mirage.
   70. SoSH U at work Posted: May 21, 2018 at 06:29 PM (#5676803)
Yes, it would hurt the reliever's numbers. And boost the starter's. But that's superficial. The reliever would still be better. Unless we are wildly incorrect about why relief pitchers have better numbers than starters, and it's actually some sort of unaccountable mirage.


We're already seeing a superficial difference today. If your plan is to a) not change the number of innings a starter is pitching, buy b) give the hardest innings to the relievers, then right now we must be seeing the starter getting more of the best hitters and the reliever getting more of the worst. That's going to play a part in why the relievers are better (as well as sample size stuff and ease of task, offsetting the fact that relievers are actually shittier pitchers).

Moreover, how many relievers are going to be tasked with this? If you simply split the job between two relievers (each starting the first every other day), then you're looking at bumping the sverage reliever innings by about 20 innings for two guys. And, as Misirlou's 43 shows, those guys who throw 80 innings of lights-out ball one year may not be the best bets to do it again next year.

Pitcher usage has both short- and long-term ramifications. Even if your solution, or others proposed, increase the WE today, they may be damaging over the long haul.


   71. PreservedFish Posted: May 21, 2018 at 06:50 PM (#5676814)
We're already seeing a superficial difference today. If your plan is to a) not change the number of innings a starter is pitching, buy b) give the hardest innings to the relievers, then right now we must be seeing the starter getting more of the best hitters and the reliever getting more of the worst. That's going to play a part in why the relievers are better

Yes, it must play a part. But I think it's a small effect compared to the "ease of task" factor.

Moreover, how many relievers are going to be tasked with this? If you simply split the job between two relievers (each starting the first every other day), then you're looking at bumping the sverage reliever innings by about 20 innings for two guys.


Dunno. I was mostly thinking about the impact on "WE today." But in reality I would probably allow my good starters to open up the game.

And, as Misirlou's 43 shows, those guys who throw 80 innings of lights-out ball one year may not be the best bets to do it again next year.


If we are right that relieving is easier than starting, we still assume that the reliever is going to be better in the first inning at least when compared to non-exceptional starters. You keep referring to small sample size, but my argument has nothing to do with small sample size - it is based on the fact that relievers are (as a whole) significantly more effective than starters. That's based on a massive sample, and it's even more striking when we acknowledge that they are lesser pitchers.

Even if your solution, or others proposed, increase the WE today, they may be damaging over the long haul.

Quite possible, yes.
   72. snapper (history's 42nd greatest monster) Posted: May 21, 2018 at 06:53 PM (#5676817)
I think you should begin most games with a reliever facing the top of the order. One of your guys that throws the ball real hard. Maybe your third best reliever. In reality, he's probably a 3.25 ERA guy and thus better than most of your starters over a single inning...

You would then generally switch to the starter in the 2nd inning. When you get to the dreaded "third time through the lineup," it's the opponent's lesser hitters that are getting that important good third look first. The better hitters would rarely get a third crack at any pitcher.


Yes, that makes sense. But, now you've used your 3rd best reliever in a game that may end up not being close.

You've only got 70 IP out of that guy, and now you've guaranteed his LI is 1.0.
   73. PreservedFish Posted: May 21, 2018 at 07:05 PM (#5676822)
I didn't think about the specifics too much. But if you're talking about replacing your 4th or 5th starter, your 7th best reliever might be good enough to give you the bonus.
   74. SoSH U at work Posted: May 21, 2018 at 07:07 PM (#5676825)
Yes, that makes sense. But, now you've used your 3rd best reliever in a game that may end up not being close.


Yes, that's an important factor. The current 60 innings is maximized in a way this method wouldn't allow for.

If we are right that relieving is easier than starting, we still assume that the reliever is going to be better in the first inning at least when compared to non-exceptional starters. You keep referring to small sample size, but my argument has nothing to do with small sample size - it is based on the fact that relievers are (as a whole) significantly more effective than starters. That's based on a massive sample, and it's even more striking when we acknowledge that they are lesser pitchers.


But look at the guy who they used in this way. He's not a better pitcher than an average starter, even with that advantage. Relieving is easier, which allows us to see at the end of the season that Team X had three guys who had great ERA+ numbers (or ERA-, for those into that sort of thing). But we don't necessarily know in April who those dominant guys are going to be as many of them don't duplicate that from one year to the next. Because they're not really great pitchers.

So I mention small sample sizes because it's a small sample of work that makes a lot of these individual pitchers look better than they are (and worse, for that matter), which we see from one year to the next. Romo was just the kind of dominant relief pitcher who you would want to use this way during the second half of the year in Tampa. During the first half, he was crap in Los Angeles. Use him (or any number of guys like him that you think are your best bullpen arms) exclusively against top of the order hitters, and I think a lot of that expected advantage is gone.
   75. PreservedFish Posted: May 21, 2018 at 07:11 PM (#5676827)
In the AL last year, starters had a 4.54 ERA and relievers - the demonstrably inferior group! - had a 4.11 ERA. The reliever bonus is real and significant.
   76. Baldrick Posted: May 21, 2018 at 07:16 PM (#5676828)
Yes, that's an important factor. The current 60 innings is maximized in a way this method wouldn't allow for.

So I mention small sample sizes because it's a small sample of work that makes a lot of these individual pitchers look better than they are (and worse, for that matter), which we see from one year to the next. Romo was just the kind of dominant relief pitcher who you would want to use this way during the second half of the year in Tampa. During the first half, he was crap in Los Angeles. Use him (or any number of guys like him that you think are your best bullpen arms) exclusively against top of the order hitters, and I think a lot of that expected advantage is gone.

Either these guys are good enough that it's important to maximize their innings in important situations or they're so unpredictable that they can't be trusted in important situations. It can't be both.

Well, that's not actually right. There is, of course, a third possibility: this would mostly come out even, meaning this usage wouldn't really make much of a difference. That's probably where I fall.
   77. PreservedFish Posted: May 21, 2018 at 07:23 PM (#5676833)
this would mostly come out even

That would be my own quibble with my strategy here. Sure, you skip the third time through the order with the 1-4 hitters. But if SP innings don't change, you're giving it to the other guys. Giving a 900 OPS hitter a 50 point bonus is no better or worse than giving a 700 OPS hitter a 50 point bonus. I don't know if there's research that gives us more detail to say otherwise.
   78. SoSH U at work Posted: May 21, 2018 at 07:27 PM (#5676837)
Well, that's not actually right. There is, of course, a third possibility: this would mostly come out even, meaning this usage wouldn't really make much of a difference. That's probably where I fall.


That's what I said from the start.

I don't think this kind of deployment would end horrifically. Just that it won't matter much, and that every small edge you think you'll get is probably offset somewhere else.

   79. cardsfanboy Posted: May 21, 2018 at 07:48 PM (#5676852)
I didn't think about the specifics too much. But if you're talking about replacing your 4th or 5th starter, your 7th best reliever might be good enough to give you the bonus.


I think that is the plan. In the book they talk about not letting your below average starters go three times through the lineup, and this seems to be a step in that direction. You aren't going to be doing it with the Kershaws or Verlanders, but you might do it with the guy who is your 4th or 5th best starter.


Moreover, just redistributing which inning you give to your short man and which one you give to your "starter" doesn't really strike me as something that should result in massive changes to win expectancy. I don't know where the value is coming from.



This is one thing I like about this, it gives you certainty of deployment for a solid leverage situation for your third best reliever. Of course at the end of the game it might not matter, but at the beginning it has some value. I'm not in the camp that thinks this is going to be the new system, but at the same time, I can see it being deployed more frequently(especially after the success that it just had) I would seriously consider it in the right circumstances, you have your fourth or fifth starter going, you have the next day off, and you have a reliever you haven't used in a while and you want to get them some work. Heck right now I don't think I would mind seeing the Cardinals do this with Holland on days that Gant is scheduled to start.
   80. Zach Posted: May 21, 2018 at 07:54 PM (#5676855)
I don't think the strategy works in the long run. You end up squandering too many of your premium relief innings on games where the "starter" (the guy who pitches the most innings) gets rocked. Or where the other team's starter gets rocked.

You'll end up with a relief ace who faces very high quality opponents (good) with terrible leverage (very bad). So your leverage multiplier goes from ~2x to ~.7x (just spitballing here), and the reliever's value drops by 2/3.
   81. snapper (history's 42nd greatest monster) Posted: May 21, 2018 at 07:57 PM (#5676857)
I don't think the strategy works in the long run. You end up squandering too many of your premium relief innings on games where the "starter" (the guy who pitches the most innings) gets rocked. Or where the other team's starter gets rocked.

You'll end up with a relief ace who faces very high quality opponents (good) with terrible leverage (very bad). So your leverage multiplier goes from ~2x to ~.7x (just spitballing here), and the reliever's value drops by 2/3.


That's what I think happens.

This is one thing I like about this, it gives you certainty of deployment for a solid leverage situation for your third best reliever.

The first inning is not a solid leverage situation. 1.0 is well below what your 3rd best RP will typically see.
   82. Howie Menckel Posted: May 21, 2018 at 07:58 PM (#5676859)
I think you're right, but I just wanted to get in my nitpick that 2 closers with identical save percentages can have different values even assuming identical orthodox usages.

not a nitpick - you're just keeping me honest!
meanwhile, are you "Nasty Nate" like the big guy is nicknamed Tiny?
   83. cardsfanboy Posted: May 21, 2018 at 08:01 PM (#5676861)
I don't see how a 0-0 game is terrible leverage.
   84. Nasty Nate Posted: May 21, 2018 at 08:10 PM (#5676865)
meanwhile, are you "Nasty Nate" like the big guy is nicknamed Tiny?
I hope so!
   85. PreservedFish Posted: May 21, 2018 at 09:13 PM (#5676925)
I think the leverage argument works if the team is using a trusted top reliever in the first. But if it's random hardthrowing dude #6, well then, who cares? A leverage index of 1.00 is better than Max Scherzer or Clayton Kershaw will ever do.
   86. The Honorable Ardo Posted: May 21, 2018 at 09:15 PM (#5676930)
I loved the move with Sergio. Let your top three starters do their thing, but use the "first-inning gambit" with spots 4 and 5. This gives the RP about 65-70 games and an equal number of IP, which is a normal seasonal workload anyways.
I don't see how a 0-0 game is terrible leverage.
Agree. It earns you the generic reliever bonus and the advantage of having a max-effort pitcher work the inning with the highest run expectancy. Especially in road games, that's a huge advantage - securing, in effect, two chances to score first.
   87. bookbook Posted: May 21, 2018 at 09:28 PM (#5676949)
In the AL last year, starters had a 4.54 ERA and relievers - the demonstrably inferior group! - had a 4.11 ERA. The reliever bonus is real and significant.


The bonus is real, but some of that difference is due to inherited runners (runs charged to the starter).
   88. Walt Davis Posted: May 22, 2018 at 03:43 AM (#5677083)
In the AL last year, starters had a 4.54 ERA and relievers - the demonstrably inferior group! - had a 4.11 ERA. The reliever bonus is real and significant.

Yes and maybe. Relievers facing batters for the first time (and generally facing no more than 5 batters) have a 4.11 ERA. Who knows how well starters facing batters the first time fare ...

AL 2017 starters, first time through, gave up a line of 252/316/423; relievers facing a guy for the first time gave up a line of 243/316/398. The difference in the walk rates is mainly IBB; the main reason for the SLG diff is a lower HR rate. Those two stats of course are correlated as IBBs tend to go to tougher batters. The BABIP difference is a pretty trivial 297 vs 294. The K-rate difference is not very big, 22.2 vs 23.5. Given we are generally comparing a guy going 9 batters the first time (and having to pace himself for another 13-15) vs. a guy usually facing 1 to 5 batters in an outing, those differences are trivial. (It's the AL, almost no pitchers batting in the first 9.)

What's missing here is that there are two classes of reliever. There are leveraged relievers, used from the 7th on with a lead or otherwise close game ... then there are the back-end of the pen and other low-leverage innings pitched by guys you'd rather leverage cuz somebody has to. In a peek a few months ago, it seemed that a typical team will have 5 relief slots (about 350 innings) produce at above-replacement and the last 3 slots (about 210 innings) are truly atrocious. Those top 5 slots produce at about a 120 ERA+ (or 83 ERA- for the double nerds).

Now let's suppose that we deploy this strategy for starters #4 and #5 regularly. Presumably we use relievers #4 and #5 -- but those guys aren't putting up 120 ERA+, they are probably around 100-105 ... still better than the #4 or #5 starter, even first time through probably. Of course they won't be 100 ERA+ facing the top 3-5 batters all the time but neither are starters #4/5.

But the problem does go deeper. Those guys will now each spend 30 of their 70 (if fully healthy) innings in low-leverage 1st inning situations. Just as importantly, 4/5 starters only go about 5 innings now anyway. Depending on recent usage of relievers 1-3, you're likely going to be putting reliever #6/7 into more high-leverage situations because everybody else is tired.

This suggests something like: employ this strategy if (a) it's your 4/5 starter's turn; (b) relievers 1-3 are rested; (c) relievers 4 and 5 are sufficiently rested. Then there seems little harm in this strategy. If (b) doesn't hold then #4 has become #3 and you're likely starting #6 in one of those two games and there's likely no advantage in that. The right conditions probably come up about 20-30 times a year.

Basically, your #4/5 relievers are the guys whose primary job in today's usage is to pitch 6th innings in close leads, 7th and 8th innings in close deficits, reasonably often the 7th in a close lead because one of 1-3 is tired, and soak up some garbage innings. Instead you're trading some of each of those for 30 low-leverage first innings. If leverage matters, this loss of leverage -- replacing some 6th/7th innings in close games with either the 4/5 starter or the #6 reliever -- surely counter-acts (on average) the fairly small advantage they have over the first few batters of the game.

Note b-r doesn't have starter/reliever by batting order splits by times faced -- shouldn't be too hard for a retrosheet genius. A fairer comp than the one I did would be pitches 1-25 by S/R since relievers rarely throw more than 25 pitches in an outing (average about 18 and that includes garbage time) but you'd still want to adjust for the fact that starters are always facing batters 1-4 (and usually 5 and 6) in those 25 pitches.

the inning with the highest run expectancy

I suspect this has the highest run expectancy because sometimes your starter doesn't have it and it's obvious pretty quickly and he gets pulled early. Rarely does a massively ineffective pitcher pitch the 3rd and 4th innings. If you're going to start relievers, often times those relievers are going to stink out of the box too. At best, all you expect is the "reliever bonus" ... which needs to be adjusted for pitches thrown and batting order spot faced and the fact that you would be using relievers #4/5 at which point I doubt there's any generic reliever bonus vs the average starter but maybe still a small one vs #4/5.

Context matters and a sizable chunk of the "reliever bonus" is the context they'd be losing by starting the 1st inning.
   89. Blastin Posted: May 22, 2018 at 05:33 AM (#5677091)
Just popping in to say this:

If you have to cover 26 miles he is.


I am hereby better than Usain Bolt.
   90. cardsfanboy Posted: May 22, 2018 at 06:02 AM (#5677093)
I suspect this has the highest run expectancy because sometimes your starter doesn't have it and it's obvious pretty quickly and he gets pulled early. Rarely does a massively ineffective pitcher pitch the 3rd and 4th innings. If you're going to start relievers, often times those relievers are going to stink out of the box too. At best, all you expect is the "reliever bonus" ... which needs to be adjusted for pitches thrown and batting order spot faced and the fact that you would be using relievers #4/5 at which point I doubt there's any generic reliever bonus vs the average starter but maybe still a small one vs #4/5.


I don't think that is the reason, the reason is because it's the only time in the order that the team is guaranteed to have the most number of "good" hitters batting. The second inning has the lowest run expectancy for the reason that it has the highest chance of the bottom of the order or lesser hitters will bat.
   91. Russlan thinks deGrom is da bomb Posted: May 22, 2018 at 09:20 AM (#5677124)
Relievers facing batters for the first time (and generally facing no more than 5 batters) have a 4.11 ERA.

How much of the difference in ERA is related to the fact that relievers often don't start the inning.

For example, a relievers comes into the 7th inning with two out and nobody on. He allows a triple to the first batter he faces but retires the next hitter. His ERA is 0.00 for that outing but he was greatly helped by the fact that he only had to avoid giving up a run before getting one out.
   92. SoSH U at work Posted: May 22, 2018 at 10:25 AM (#5677165)
AL 2017 starters, first time through, gave up a line of 252/316/423; relievers facing a guy for the first time gave up a line of 243/316/398.


To be fair to the PF, cfb side of the argument, the first-time through vs. first-time through is not the apt comparison, but first-time through against third-time through, since that's what their plan is designed to avoid.

   93. Tom Nawrocki Posted: May 22, 2018 at 10:43 AM (#5677183)
It's worth pointing out that Sergio Romo is not actually a good pitcher anymore. He's just a run-of-the-mill reliever who happened to match up well against the top of the Angels' batting order. So it's not like Cash was burning his closer to try to ensure to a clean first inning.
   94. McCoy Posted: May 22, 2018 at 10:55 AM (#5677192)
I couldn't believe that Romo was still in the league but then I realized I was thinking of Sergio Mitre who upon looking up on BRef I discovered has probably the coolest photo of in all of BRef.
   95. SoSH U at work Posted: May 22, 2018 at 10:59 AM (#5677193)
I was thinking of Sergio Mitre who upon looking up on BRef I discovered has probably the coolest photo of in all of BRef.


What kind of cap is he wearing?

   96. What did Billy Ripken have against ElRoy Face? Posted: May 22, 2018 at 11:01 AM (#5677195)
I couldn't believe that Romo was still in the league but then I realized I was thinking of Sergio Mitre
Jane says she's done with both of them.
   97. McCoy Posted: May 22, 2018 at 11:03 AM (#5677197)
Bravos de Leon hat from his time with them last year.
   98. Nasty Nate Posted: May 22, 2018 at 11:19 AM (#5677204)
Jane says she's done with both of them.
It's too bad for her baseball career that she takes a swing, but she can't hit!
   99. ERROR---Jolly Old St. Nick Posted: May 22, 2018 at 11:19 AM (#5677207)
Here's my only problem with ERA-

BB-Reference doesn't use it.

I wish they would, but it's a pain in the ass to have to go somewhere else for a single stat when 99% of what you want is on BB-Reference.
   100. Misirlou doesn't live in the restaurant Posted: May 22, 2018 at 11:21 AM (#5677211)
How much of the difference in ERA is related to the fact that relievers often don't start the inning.

For example, a relievers comes into the 7th inning with two out and nobody on. He allows a triple to the first batter he faces but retires the next hitter. His ERA is 0.00 for that outing but he was greatly helped by the fact that he only had to avoid giving up a run before getting one out.


Eh, how often is there a pitching Chang with 2 out no one on? I’m not saying it never happens, but it’s rare
Page 1 of 2 pages  1 2 > 

You must be Registered and Logged In to post comments.

 

 

<< Back to main

News

All News | Prime News

Old-School Newsstand


BBTF Partner

Dynasty League Baseball

Support BBTF

donate

Thanks to
Mike Emeigh
for his generous support.

Bookmarks

You must be logged in to view your Bookmarks.

Hot Topics

NewsblogOT - 2018 NBA Summer Potpourri (finals, draft, free agency, Colangelo dragging)
(1240 - 4:16pm, Jun 19)
Last: aberg

NewsblogOT - Catch-All Pop Culture Extravaganza (April - June 2018)
(3759 - 4:15pm, Jun 19)
Last: Don August(us) Cesar Geronimo Berroa

NewsblogGrumpy Gossage Napalms His Bridges:BWitz:NYT (reg req.)
(79 - 4:15pm, Jun 19)
Last: snapper (history's 42nd greatest monster)

NewsblogFormer MLB pitcher Kevin Brown reportedly held two mail thieves at gunpoint until police arrived
(184 - 4:13pm, Jun 19)
Last: The Yankee Clapper

NewsblogSoccer? OMNICHATTER hardly knows'er! for June 19, 2018.
(34 - 4:12pm, Jun 19)
Last: Jess Franco

NewsblogOTP 2018 June 18: How Life Imitates the Congressional Baseball Game
(490 - 4:11pm, Jun 19)
Last: DavidFoss

NewsblogOT: Soccer Thread (World Cup)
(577 - 3:18pm, Jun 19)
Last: McCoy

NewsblogKelvin Herrera Trade Start of Something Big for Nationals
(7 - 3:13pm, Jun 19)
Last: McCoy

NewsblogPrimer Dugout (and link of the day) 6-19-2018
(45 - 2:54pm, Jun 19)
Last: vortex of dissipation

NewsblogDodgers' surprise ace makes All-Star case in win over Giants
(97 - 1:49pm, Jun 19)
Last: McCoy

NewsblogIt's not a crime when OMNICHATTER does it! for June 18, 2018
(80 - 1:14pm, Jun 19)
Last: Moses Taylor, aka Hambone Fakenameington

Sox TherapyA Pleasant Trip So Far
(9 - 1:07pm, Jun 19)
Last: Jose is an Absurd Doubles Machine

NewsblogJuan Soto Makes His Second MLB Debut Tonight
(18 - 12:20pm, Jun 19)
Last: Mike Emeigh

Hall of MeritMost Meritorious Player: 1944 Discussion
(10 - 12:16pm, Jun 19)
Last: MrC.

Gonfalon CubsClicking
(52 - 12:16pm, Jun 19)
Last: Moses Taylor, aka Hambone Fakenameington

Page rendered in 0.7020 seconds
46 querie(s) executed