The unpredictability of relief pitching

[Trov]

Coming into the season we thought we had a closer, or some did, in Taylor Rodgers.  The Twins pulled of the day 1 trade flipping him for Pagan and Paddock basically.  Others were involved but they were all current MLB players changing right before season started.  At first, both Rodgers and Pagan had success, but Pagan was getting by with giving up a ton of walks and some hard contact.  There was many fans upset because Rodgers through 20 gams was nothing short of amazing.  Giving up a run, striking out about a runner per inning, and walking only 4, but hitting 3 batters as well.  He was still 17 of 17 in saves.

Meanwhile, Pagan started to struggle, Paddock got put on IL with Tommy John.  The trade was beginning to look very bad for the Twins, despite still years of control.  Also, other pen pitchers for Twins were doing poorly like Tyler Duffey.  It was clear we needed a trade and looked like we should be regretting our early trade.  However, since that first 20 games for Rodgers he has been pretty bad, and by WAR worse than Duffey, but most likely in part because he is in higher leverage situations too.

Since the first 20 games Rodgers has jumped from a 0.44 ERA to a 5.00.  He has converted 11 more saves, but blew 7 saves, not a very good conversion rate.  His FIP has not not jumped in line with his ERA, so some would say he is getting unlucky, but he has been similar in past seasons having ERA much higher than FIP since 2019 season.  That is a long streak of bad luck, and fact is he is giving up runs as a late inning guy.  He got demoted from closer, and traded for a expected upgraded, in Josh Hader. 

On the Twins front we cut Duffey and traded for Michael Fulmer and Jorge Lopez.  Both were upgrades to what we had, but Lopez has blown 2 saves so far, and Fulmer took loss against Dodgers.  Pagan got lucky the other day against Angels for 1 inning, but then quickly gave up the walk of HR.  

The point I am making is we cannot ever predict how a pen will do.  Maybe we win a game or two more if we had not traded Rodgers, but assuming he progressed same way with Twins we would be calling for him to be cut as well.  Maybe we do not make the trade for Lopez, maybe that would have been a good thing.  I have long said you can never expect a pen guy to be dominate all the time.  They will all have bad stretches, you just hope it does not cost you games.  I am not addressing the move of Rodgers at all, so please lets not rehash those comments.  I am just saying you never know what the pen guy will do until they get out there and to expect they will bounce back or regress is not something you should expect. 

[notoriousgod71]

So if no one can trust anybody in the bullpen we should try to get as many innings as possible out of the top half of the rotation, yeah?

 

If you can't trust anyone you have failed as a GM .

[bean5302]

It's Taylor ROGERS. Like Mr. Rogers.

Not RODGERS like Aaron Rodgers.

Anyway, I love for people to wait until a relief pitcher has a really rough game to jump all over them (this seems to happen frequently for Rogers around here). The guy owns a 2.24 FIP and 3.32 xFIP this year. Peripherals look outstanding as well. He's been a reliably good closer for years now, but he's just seemingly been unlucky, and with the small number of innings a reliever has, luck can be everything.

2.81 FIP, 3.32 xFIP, 11.00 K/9 (28.5%) vs. 2.00 BB/9 (5.2%) BABIP .349
Rogers' stuff is working, too. He's essentially tied for the lead in called strike percentage. Hitters are being frozen watching strike 3 over and over again and he's just above average in swinging strike rate. Contact rate on Roger's stuff out of the zone is excellent. The BABIP is killing him.

Let's look at our closer acquisition Jorge Lopez.
2.97 FIP, 3.26 xFIP, 9.79 K/9 (26.6%) vs. 3.21 BB/9 (8.7%) BABIP .246
K rate is worse. BB is much worse. Chase rate is worse. Contact rates are worse. Swinging strike rate is worse, called strike rate is worse. First pitch strike is lower. Basically, across the board Lopez is inferior to Rogers.

How about our super stud Jhoan Duran.
3.04 FIP, 2.32 xFIP, 11.55 K/9 (32.3%), 2.13 BB/9 (6.0%). BABIP .289
Duran's given up an extra home run or two more than expected. His overall line is better than Rogers this year. Duran has been pretty elite, but you can see the K rate, BB rate, and FIPs are similar to Rogers.

When you're talking about 20 or 30 or 40 innings, you're talking about 5-7 starts for a starter (162 starts for Baldelli managed teams), all it takes is a couple unlucky base runners and a home run to jack ERA's up. Rogers has allowed 0 earned runs in 32 and 1 earned run in 8 of his 46 appearances. That's 87% of the time Rogers allowed 0-1 runs (earned or unearned) total.

 

[Mike Sixel]

Pagan has been not good for years. His HR rate is insane. He should not be on this roster.

[Vanimal46]

In theory yes, there are not many relievers in baseball anymore who are consistently good month after month, year after year. Even guys like Kenley Jansen who are multi time all stars have bad years then find it again. They are the kings of SSS and I am in agreement with Falvine’s philosophy of not spending big money on the position. 

The problem in recent years is not moving on quickly enough from veteran relievers. Colomé stuck around all of 2021 after his horrific start to the season. Pagan is still here when he’s been the root cause of 6+ losses. If you’re going to rely on your pen completing 4 innings every game, you need a large stable of arms to rotate through, and have a quick hook when it doesn’t work out. 

[jkcarew]

The only exception to the rule…

If the Twins have recently traded for it…you know exactly how it will perform. ?

[Game7-91]

The way pitching staffs are used, they are only as good as the weakest link in the chain. On any given night, 3 or 4 relievers are given a chance to turn victory into defeat. Law of averages says at least one of those links will oblige on any given night. As notoriousgod71 says above, a big part of bettering the pen is increasing performance from the SP's. Is it really going to be acceptable year after year that your highly compensated SP is expected to get 15 outs, giving up 3 runs or less, a "quality outing", while the BP is then expected to get 12 outs, including the outs 21-27, when so many close games are decided. Dont pay the pen? They are tasked with getting nearly the same amount of outs as the SP's, and those outs are higher leverage, less margin for error than are outs 1-15.  And these are usually the guys who are inferior pitchers...the reason they are RP's is usually because they dont have the secondary pitches to begin with. And we expect a sequence of 3-4 guys like this to protect 1-2 run leads? Talk about being set up to fail.

Rotation usage is bass ackwards imho.

The SP's need to step up. 110 pitches is not going to break their arm. Most of them break down averaging 90. Maybe teams need to look at 6 man starting rotations, with higher pitch count expected, and 18 outs expected. Reduce BP usage to 2 guys per game. At least gives the BP less chance to blow it, and shortens the number of outs needed to get to the closer.

[Doctor Gast]

SD do not know how to manage their pitchers. Rogers decline is due IMO to SD's mismanagement. Twins have discovered that Rogers does not pitch well on consecutive days. I imagine that SD pitched him as much as possible ,.

[Number3]

The way starting pitchers are used in the game today, I wonder if the days of the huge contracts for starters is over. There is the documentary about the history of baseball that has a segment about the old Negro Leagues and a comment by one of the pitchers always stands out. He said (paraphrase), "We knew when we went to the mound that we were paid to pitch 9 innings (or maybe more) period.". One thing the old timers would do was pace themselves until they got into trouble and then bear down. Now its all out all the time. There is a drill in the military (or used to be) where you simply ran with full gear as fast as you could as far as you could until you basically dropped. Now that is the way starters pitch and the managers almost never give them a chance to pitch out of a jam much after the 3rd inning. No wonder bull pens are totally unpredictable day to day. If it weren't for existing contracts and the archaic designation of "starting pitcher" why not pitch every game by committee? Might as well. There are only 1 or 2 Max Scherzers (sp) left anyway.

[Rosterman]

16 hours ago, Game7-91 said:

The SP's need to step up. 110 pitches is not going to break their arm. Most of them break down averaging 90. Maybe teams need to look at 6 man starting rotations, with higher pitch count expected, and 18 outs expected. Reduce BP usage to 2 guys per game. At least gives the BP less chance to blow it, and shortens the number of outs needed to get to the closer.

Six man rotation? Then you need a bunch of relief piotchers who caa do multiple innings and some day-after-day. 

Maybe we need to go with a 4-man rotation to have that extra bullpen arm since starters are only pitching twice thru an order. Would that change things too much?

The life of a pitcher is dictated by bad games. You have a bad (5-inning) start and it can destroy your ERA et al. Same with a bullpen arm. You put three guys on basem get pulled, the next guy comes in and finishes the inning giving up a couple of hits, but ALL your runners score.

A relief pitcher gives up two runs in an inning. He has to pitch three mroe innings of perfect ball to get his ERA to 4.50. Nine inning total of relief to look brilliant again, which may take up to nine appearances.

 

 

[TwinsDr2021]

2 hours ago, Number3 said:

 Might as well. There are only 1 or 2 Max Scherzers (sp) left anyway.

There is no way to know how many Scherzer types are left because teams aren't training them to do that. I mean Sandy Alcantara can't be the only 26 year old pitcher in the world that can average over 7 innings, can he? The Twins will never know how much stamina their pitchers have or how good they truly can be, because they are only asking 5 innings from their starters in the minors, may goodness here on TD they talk about how great a game a minor league pitcher had even if he pitched 5 innings.

I will say this though, back when I was in college (I pitched a little bit, real little bit) my coach said to us pitchers if you have to go to your third or fourth pitch to get people out in the first couple of innings, know you are going to be a relief pitcher. The best pitchers where able to get guys out early in the game with their best couple of pitches, so when they faced them later in the game they could throw all their pitches and keep the batters off balance. Maybe with video, only a few of the best pitchers can still do that anymore.

I think it was Ryan's start against Colorado where Gladden said, this could be a long/short day for Ryan because he is already going to his secondary stuff in the first inning. (he went on to pitch 102 pitches in 5 innings that day)

[Trov]

20 hours ago, bean5302 said:

It's Taylor ROGERS. Like Mr. Rogers.

Not RODGERS like Aaron Rodgers.

Anyway, I love for people to wait until a relief pitcher has a really rough game to jump all over them (this seems to happen frequently for Rogers around here). The guy owns a 2.24 FIP and 3.32 xFIP this year. Peripherals look outstanding as well. He's been a reliably good closer for years now, but he's just seemingly been unlucky, and with the small number of innings a reliever has, luck can be everything.

2.81 FIP, 3.32 xFIP, 11.00 K/9 (28.5%) vs. 2.00 BB/9 (5.2%) BABIP .349
Rogers' stuff is working, too. He's essentially tied for the lead in called strike percentage. Hitters are being frozen watching strike 3 over and over again and he's just above average in swinging strike rate. Contact rate on Roger's stuff out of the zone is excellent. The BABIP is killing him.

Let's look at our closer acquisition Jorge Lopez.
2.97 FIP, 3.26 xFIP, 9.79 K/9 (26.6%) vs. 3.21 BB/9 (8.7%) BABIP .246
K rate is worse. BB is much worse. Chase rate is worse. Contact rates are worse. Swinging strike rate is worse, called strike rate is worse. First pitch strike is lower. Basically, across the board Lopez is inferior to Rogers.

How about our super stud Jhoan Duran.
3.04 FIP, 2.32 xFIP, 11.55 K/9 (32.3%), 2.13 BB/9 (6.0%). BABIP .289
Duran's given up an extra home run or two more than expected. His overall line is better than Rogers this year. Duran has been pretty elite, but you can see the K rate, BB rate, and FIPs are similar to Rogers.

When you're talking about 20 or 30 or 40 innings, you're talking about 5-7 starts for a starter (162 starts for Baldelli managed teams), all it takes is a couple unlucky base runners and a home run to jack ERA's up. Rogers has allowed 0 earned runs in 32 and 1 earned run in 8 of his 46 appearances. That's 87% of the time Rogers allowed 0-1 runs (earned or unearned) total.

 

First, sorry for the misspelling.  Second, my post was not an attack on Rodgers at all, but pointing out even those that are considered good, have stints of poor play.  I have for years talk about in relief pitchers you can try to argue luck all you want, but at some point you need to look at results.  I mean Duran gave up a HR in LA that according to broadcast would have been a HR only in that park, that in any other park would have been an out on the track or in front of track.  At Target Field most likely not even on the track.  Was that bad luck because he was in a stadium that was smaller in that spot than every other ballpark?  FIP would say no, he gave up a HR and luck had nothing to do with it, even if every other ball park it is an out.  Also, for Rogers, to my knowledge FIP does not account for HBP, at least the formula I have seen it does not, he has hit 8 guys this year. 

Rogers may have elite stuff per the numbers you site, but he still has blown 7 saves in 35 chances, and it has been 7 blown out of his last 18 chances.  I am not discounting he had a dominate first 2 months, but since then he lost closer job and has been good for a total -.9 wins probability added on the season, which reflects his high leverage outings.  If you look at his career, where you would argue he is getting unlucky, he has totaled -2.3 WPA, since end of 2019.  He did have positive 0.7 2021.  His FIP and ERA got out of whack in 2020.  

Again, the purpose of my post was not to attack Rogers, I had issues with him blowing saves, holds, and losing games when tied in 2020, and people said he was getting bad luck, pitching back to back days and many other excuses.  Point was, he has bad stretches, and he has had a bad stretch since June really.  He is not the worst in the league and would be more than happy to pitch with Twins over Pagan, but for whatever reason he is not doing well as of late.  Maybe luck, maybe not. He is giving up a hard hit rate of 40.3% this year, which is only a little above his career average of 38.7% and line drives are actually below his career average.  Maybe he is just having a long run of bad luck, but he is not getting outs with enough regularity.  Maybe he bounces back, I am not saying he is cooked.  I am pointing out that if his season had flipped his start to now, our view of him would be very different. 

All pen guys are subject to SSS and what have you done lately. Maybe they get longer leash if it appears to be more luck than anything else, but results are all that really matters. 

[bean5302]

I'm saying the unpredictability is inherent to the game when it comes to small sample sizes. Relievers are always small sample sizes.

[heresthething]

19 hours ago, Vanimal46 said:

In theory yes, there are not many relievers in baseball anymore who are consistently good month after month, year after year. Even guys like Kenley Jansen who are multi time all stars have bad years then find it again. They are the kings of SSS and I am in agreement with Falvine’s philosophy of not spending big money on the position. 

The problem in recent years is not moving on quickly enough from veteran relievers. Colomé stuck around all of 2021 after his horrific start to the season. Pagan is still here when he’s been the root cause of 6+ losses. If you’re going to rely on your pen completing 4 innings every game, you need a large stable of arms to rotate through, and have a quick hook when it doesn’t work out. 

Given how the FO uses the bullpen how can they not spend big money or prospect capital on it?  We've been through enough seasons where they didn't spend enough money on the BP and it cost us.  

[Vanimal46]

2 hours ago, heresthething said:

Given how the FO uses the bullpen how can they not spend big money or prospect capital on it?  We've been through enough seasons where they didn't spend enough money on the BP and it cost us.  

I would argue they haven’t invested in enough people, not specifically money, in recent seasons. Even the highest paid relievers go through bad seasons, and bad investments like that can anchor a team like the Twins. Edwin Diaz, for example has ERAs of 5.59, 1.75, 3.45, and 1.33 during his Mets tenure. We simply can’t hide a $10-15 million reliever if he ends up with a 5+ ERA. 

Id rather go with the quantity over perceived quality approach in the bullpen. We should be more aggressive calling up fringe SP prospects and transition them to the pen. We should sign 4 or 5 Joe Smith types instead of 1. And shuffle through them until a couple over perform. 

[h2oface]

3 hours ago, Trov said:

First, sorry for the misspelling.  Second, my post was not an attack on Rodgers at all, but pointing out even those that are considered good, have stints of poor play.  I have for years talk about in relief pitchers you can try to argue luck all you want, but at some point you need to look at results.  I mean Duran gave up a HR in LA that according to broadcast would have been a HR only in that park, that in any other park would have been an out on the track or in front of track.  At Target Field most likely not even on the track.  Was that bad luck because he was in a stadium that was smaller in that spot than every other ballpark?  FIP would say no, he gave up a HR and luck had nothing to do with it, even if every other ball park it is an out.  Also, for Rogers, to my knowledge FIP does not account for HBP, at least the formula I have seen it does not, he has hit 8 guys this year. 

Rogers may have elite stuff per the numbers you site, but he still has blown 7 saves in 35 chances, and it has been 7 blown out of his last 18 chances.  I am not discounting he had a dominate first 2 months, but since then he lost closer job and has been good for a total -.9 wins probability added on the season, which reflects his high leverage outings.  If you look at his career, where you would argue he is getting unlucky, he has totaled -2.3 WPA, since end of 2019.  He did have positive 0.7 2021.  His FIP and ERA got out of whack in 2020.  

Again, the purpose of my post was not to attack Rogers, I had issues with him blowing saves, holds, and losing games when tied in 2020, and people said he was getting bad luck, pitching back to back days and many other excuses.  Point was, he has bad stretches, and he has had a bad stretch since June really.  He is not the worst in the league and would be more than happy to pitch with Twins over Pagan, but for whatever reason he is not doing well as of late.  Maybe luck, maybe not. He is giving up a hard hit rate of 40.3% this year, which is only a little above his career average of 38.7% and line drives are actually below his career average.  Maybe he is just having a long run of bad luck, but he is not getting outs with enough regularity.  Maybe he bounces back, I am not saying he is cooked.  I am pointing out that if his season had flipped his start to now, our view of him would be very different. 

All pen guys are subject to SSS and what have you done lately. Maybe they get longer leash if it appears to be more luck than anything else, but results are all that really matters. 

I had to laugh. Brains can be stubborn. Immediate resistance, but the learning curve is happening. I see a guy with one of the highest posting totals on the site spell names wrong all the time. Still can't figure out the double l in Kirilloff, and others, like the e in Duffey, and that really doesn't matter anymore, and we all know who they are talking about. At least for the challenged, Presley/Pressly duo is long gone and that is over. 

[h2oface]

On 8/15/2022 at 3:18 PM, Game7-91 said:

......The SP's need to step up. 110 pitches is not going to break their arm. Most of them break down averaging 90. Maybe teams need to look at 6 man starting rotations, with higher pitch count expected, and 18 outs expected. Reduce BP usage to 2 guys per game. At least gives the BP less chance to blow it, and shortens the number of outs needed to get to the closer.

I agree, except it is not their choice when they are pulled at 80 pitches in the 5th, and they have the 2 run lead with 2 outs and a runner on first and third. I am sure many of them many times would have rather continued, worked out of the inning themselves, pitched the 6th, too, instead of watching the blowpen come in and give up a hit or a homer to the first batter or even on the first pitch and blow their game and the teams lead up in one careless pitch. 

They want to step up. They don't get to make that decisison.

[h2oface]

I thought KC showed the baseball world in 2014 and 2015 how important the bullpen was with Herrera/Davis/Holland. Others started going for the lockdown last 3 innings and paid for it. And were successful. The starter was still usually going the first 6 instead of maybe the first 5. To ignore the last 3 innings and just have a hope and a prayer, and you will likely hover around .500 at best.

[jorgenswest]

Reliability in relievers is difficult to measure at the run level. ERA needs a sample size of several reliever years to stabilize so any unreliability is likely the measure and not the reliever.

To look for a decline in performance or unreliability I think you would need to consider performance at the pitch and plate appearance level.

[heresthething]

3 hours ago, Vanimal46 said:

I would argue they haven’t invested in enough people, not specifically money, in recent seasons. Even the highest paid relievers go through bad seasons, and bad investments like that can anchor a team like the Twins. Edwin Diaz, for example has ERAs of 5.59, 1.75, 3.45, and 1.33 during his Mets tenure. We simply can’t hide a $10-15 million reliever if he ends up with a 5+ ERA. 

Id rather go with the quantity over perceived quality approach in the bullpen. We should be more aggressive calling up fringe SP prospects and transition them to the pen. We should sign 4 or 5 Joe Smith types instead of 1. And shuffle through them until a couple over perform. 

Quantity over quality could work but how do you manage the 40 man roster?  I don't have enough background on the 40 man roster to know but from what I've learned on this site it could be problematic.  

[USAFChief]

On 8/16/2022 at 3:04 PM, Vanimal46 said:

I would argue they haven’t invested in enough people, not specifically money, in recent seasons. Even the highest paid relievers go through bad seasons, and bad investments like that can anchor a team like the Twins. Edwin Diaz, for example has ERAs of 5.59, 1.75, 3.45, and 1.33 during his Mets tenure. We simply can’t hide a $10-15 million reliever if he ends up with a 5+ ERA. 

Id rather go with the quantity over perceived quality approach in the bullpen. We should be more aggressive calling up fringe SP prospects and transition them to the pen. We should sign 4 or 5 Joe Smith types instead of 1. And shuffle through them until a couple over perform. 

Quantity over quality is what this FO does. Sign a bunch of minor league FAs and waiver claims, stash them in AAA, and shuffle them through the pen. 

That's why this team is always short in the pen. 

"Shuffle through them until a couple overperform" is a terrible idea. First, it's unlikely to work, and second, even if you end up with a couple usable guys, you've spent a good portion of the season blowing games while you shuffled. Meanwhile a couple other flash-in-the-pans have lost the magic pixie dust that led to a couple successful innings and you're looking for their replacements.

This FO needs to rethink its pen strategy, particularly given their clear decision to limit starters. "It'll work itself out" isn't a viable strategy. 

 

[Number3]

Yesterday it did actually end up as pitching by committee as a result of injury to Mahle in the 3rd inning so the extreme of pulling starters early did happen. Maybe Baldelli should have 2 lineup cards. 1 for position players and 1 for pitchers. Have 3 pitchers scheduled for 3 innings each and a pitcher in reserve for each 3 inning segment.  Divide the game into 3 separate games and try to win 2 of the 3. Forget starting pitchers all together.