Tim Lincecum: Now a Reliever, Maybe Needs to Close

Tim Lincecum is headed to the bullpen. After a miserable start to the second half — opposing batters are hitting .331/.422/.622 against him since the All-Star break — the Giants have finally removed him from the rotation and will experiment with Lincecum as a relief pitcher. Of course, Lincecum famously dominated out of the pen in the 2012 playoffs, and ever since, speculation has mounted that this was going to be the path to Linecum’s career revival. Bullpens are full of failed starters, some of whom have gained significant velocity while pitching in shorter stints and have turned into lights-out bullpen arms.

The Giants would be thrilled if Lincecum turned into their version of Wade Davis, for instance; as a starter, he allowed a .341 wOBA over his career, but hitters have posted just a .222 wOBA off him in his relief work. Some guys just need the boost that comes from throwing 20 pitches instead of 100, and it’s not hard to draw a correlation between Lincecum’s decline in velocity and performance. If moving Lincecum to the bullpen gets his fastball back to the mid-90s, he might be able to reinvent himself in the new role.

However, Lincecum’s struggles present a perhaps unique challenge in turning him into an ace reliever. As I wrote for Fox a few months ago, almost the entire portion of Lincecum’s struggles can be chalked up to struggles with men on base. I think the tables that were shown in that article are worth showing again, though I’ve updated the 2014 and total lines to take into account the more recent data.

Lincecum, career, bases empty.

Season BB% K% FIP xFIP
2007 9% 25% 3.04 3.44
2008 11% 27% 3.05 3.50
2009 7% 29% 2.20 2.77
2010 9% 26% 2.93 3.01
2011 10% 25% 3.22 3.31
2012 9% 26% 3.80 3.31
2013 9% 25% 3.93 3.34
2014 9% 23% 3.30 3.43
Total 9% 26% 3.15 3.24

Tim Lincecum, career, men on base.

Season BB% K% FIP xFIP
2007 12% 23% 4.43 4.32
2008 7% 31% 2.17 2.73
2009 8% 29% 2.55 2.90
2010 8% 26% 3.41 3.18
2011 9% 24% 3.11 3.41
2012 14% 20% 4.67 4.47
2013 10% 20% 3.49 3.84
2014 11% 17% 5.47 4.23
Total 10% 24% 3.48 3.53

Lincecum’s career wOBA allowed with the bases empty is .299; this year, it’s .313. That’s worse, but the magnitude of the difference is hardly anything to care about. However, his career wOBA allowed with men on base is .296, and this year, it’s .381. Out of the wind-up, Linecum has mostly been the Lincecum he’s always been, but out of the stretch, he’s been throwing batting practice.

The easy answer seems to be that pitching out of the stretch is causing his stuff to flatten out, but as I noted in the Fox piece, the data doesn’t really support that conclusion; his velocity is steady regardless of the number of baserunners. The primary culprit seems to be worse location with men on base, which could be a mechanical thing or a mental thing, or even just randomness. Diagnosing a cause is difficult.

However, we can note that Lincecum’s struggles have only really manifest themselves in this specific scenario, and generally, the goal of moving a guy to the bullpen is to maximize his usage in situations that he’s set up to succeed in. You take a pitcher with big platoon splits out of the rotation and use him as a match-up specialist to take advantage of his skills against same-handed hitters, for instance. But if Lincecum’s kryptonite is pitching from the stretch, then a middle relief role might not serve to limit those opportunities, since the guys pitching in front of the closer often need to come in with men on base and try and kill a rally.

Lincecum’s splits suggest that perhaps the best way to “fix” him is to let him pitch with the bases empty as often as possible, which means starting the inning and not cleaning up after others. And there’s only one role in the bullpen that is generally afforded that luxury; the closer. Sure, Bruce Bochy could just use Lincecum as a middle reliever who wasn’t used as a fire extinguisher, and could slot him into a rigid start-the-7th-inning-only role or something, but over the long term, no manager wants to be handcuffed to having a setup men unavailable in the highest leverage situation. Limiting Lincecum to middle innings work with no one on base necessarily means that he’s going to be pitching in lower leverage outings, and thus, won’t be of that much value even if he does pitch well.

But while save situations are often overrated, they are normally high leverage outings. Closers, as a group, post higher leverage numbers than setup guys, even though they are occasionally called on to protect three run leads. There are more one or two run leads to protect, however, and overall, closers do pitch in more meaningful innings than setup guys, even though they don’t come into situations with men on base and try to kill rallies. If you want Lincecum to have significant value in a bullpen role, it seems that giving him the ninth inning might be the best way to accomplish that goal.

But I’m guessing it’s probably not that easy to take a guy who has turned every hitter into Miguel Cabrera for the last month and hand him the ball in a critical situation. Usually, when you kick a starter to the bullpen, the goal is to get him some low leverage stints to rebuild some confidence, and I’m guessing that’s how the Giants will use Lincecum initially.

But if it works, they’ll be faced with a bit of a dilemma; put him back in the rotation where he’ll have to try and pitch out of jams again, or consider whether a quick ascension to the closer role is a bridge they want to cross. And part of that decision will have to come down to their interest in seeing Lincecum as a starter again. If they do give Lincecum some high leverage relief innings, and he pitches well in those outings, it’s going to be difficult to argue that he should ever start again, much in the same way that Wade Davis’ performance out of the bullpen has made it unlikely that he’ll ever be asked to do anything else. Once they move him into the closer’s role, he’s probably there for good, or at least until he’s no longer good enough to close either.

Given the history of the starter-reliever conversion, this seems to have a pretty good chance of working out pretty well for Lincecum. Especially if he’s allowed to primarily pitch with the bases empty, and perhaps in shorter stints, he’d even figure out how to work his way out of his own jams. But Lincecum’s struggles to strand runners make a middle relief role a bit awkward, and that might force the Giants hand a bit. If they think he’s a reliever for good, make him the closer and don’t look back.




Print This Post



Dave is a co-founder of USSMariner.com and contributes to the Wall Street Journal.


46 Responses to “Tim Lincecum: Now a Reliever, Maybe Needs to Close”

You can follow any responses to this entry through the RSS 2.0 feed.
  1. Chris Perez says:

    High time they did that.

    +32 Vote -1 Vote +1

  2. DeQuarium Lumpkin says:

    It’s odd to talk about Lincecum’s bases empty/men on base splits as if it were two different pitchers. You write:

    “Lincecum’s splits suggest that perhaps the best way to “fix” him is to let him pitch with the bases empty as often as possible, which means starting the inning and not cleaning up after others.”

    For a starter, every inning starts with the bases empty. Having him pitch higher-leverage innings with his problems with men on base does not follow from your argument.

    Now, Lincecum’s velocity may improve in relief, leading to performance improvement, but you cannot make an argument based on empty/on base splits.

    Vote -1 Vote +1

    • CJ says:

      I thought the argument was relatively clear.

      1) TL pitches better with no men on base.
      2) Starting pitchers who are converted to relievers tend to perform better in all contexts.
      3) Pitching badly tends to put men on base (implicit).

      So if you convert TL to relief: he pitches better due to the starter/reliever performance adjustment, leading to less men on base than he would generate as a starter, which leads to better results.

      That is, theoretically, pitchers who perform worse with runners on will demonstrate a larger increase in performance in moving to relief than the average starter/reliever convert.

      Vote -1 Vote +1

      • DeQuarium Lumpkin says:

        I follow all of that, but it is an argument against bringing in Lincecum with runners on base, not an argument for him to pitch high-leverage innings.

        Vote -1 Vote +1

        • Yirmiyahu says:

          Isn’t that the argument Dave is making? They should try to minimize bringing him in with runners on base, and as a practical matter that means closer.

          +7 Vote -1 Vote +1

        • Jason B says:

          …but as he is saying, Lincecum as a starter got to start EVERY inning with bases empty, and that hasn’t exactly translated to success this year.

          Vote -1 Vote +1

        • Catoblepas says:

          Right, but it’s two separate thoughts. Lincecum to relief –> better Lincecum. Reliever Lincecum to closer –> better reliever Lincecum.

          Vote -1 Vote +1

        • Jason B says:

          I’m not sure you just go toss him in the closer role and hope for the best though, right? Start him in lower leverage situations to see how he’s adjusting to the pen while Casilla closes, and if reliever Lincecum does indeed look better than starter Lincecum, THEN consider him to close.

          Vote -1 Vote +1

    • agam22 says:

      If you pitch seven innings, you will inevitably have many situations with men on base. Closing, especially with an ability to throw harder knowing you have to only pitch one inning, would give you a better chance of getting him in and out of the game without having men on base. It would also give him more situations where he could ignore a runner. As a starter, you are trying to limit as many runs as possible because you have no idea what may or may not happen. As a closer, if you have a 3 run lead you probably won’t care at all that there is a guy in scoring position at second base.

      Vote -1 Vote +1

      • DeQuarium Lumpkin says:

        It does not follow that pitching in a relief role where the bases are empty, by itself, gives Lincecum a better chance of throwing a greater number of scoreless innings over a given span of single-inning outings than over the equivalent number of innings as a starter. Now, the starter-to-reliever improvement in “stuff” may lead to better performance, but simply shifting the same base-out state from starting to relief does not do so.

        Vote -1 Vote +1

        • DP says:

          If you don’t understand that knowing that a he has only 1 inning to pitch is different than 5+ innings of work ahead, then you don’t get DC’s premise.

          Vote -1 Vote +1

        • Brian says:

          You can’t, in this case, separate the two. Pitching in relief will increase his performance overall, including with the bases empty. Since batters should reach less often, he will be pitching more often in the situations in which he probably excels. The hope is that this will magnify the starter-to-reliever improvement.

          You’re somewhat right in the sense that most of any improvement he sees will still be just because any pitcher pitches better in relief but it looks reasonable to hope that Lincecum will improve more than your average pitcher.

          Vote -1 Vote +1

        • Chris K says:

          It’s the Times Through the Order Penalty (TTOP). He’s less likely to concede baserunners in a relief setting than in the third time through the order as a starter, even though there may be nobody on when he starts the inning in either scenario.

          Vote -1 Vote +1

    • Yirmiyahu says:

      That’s true. But I think Dave’s article is premised on a) the fact that the Giants have announced they’re moving Lincecum to the bullpen, and b) the question of what his role should be in the bullpen.

      Dave is right that a closer comes into a game with the bases empty more often than a set-up guy, or a middle reliever, or a long-relief guy, or a ROOGY.

      I don’t think that Dave is arguing that the closer role is the solution to all his problem pitching with runners on. I think he’s saying that it’s better than the alternative bullpen roles.

      +7 Vote -1 Vote +1

    • Bip says:

      You’re right, but it’s worth pointing out not because a shift to the bullpen would cause him to pitch with fewer men on base, but because the Giants should specifically avoid using him as a fireman.

      Basically, if the Giants start putting him in mid-inning to try to clean up a mess, they may find he’s even worse that he was as a starter, so they should avoid doing that.

      Vote -1 Vote +1

  3. Yirmiyahu says:

    Really, the only pitching role that can be guaranteed to start with a clean inning and not clean up the messes of others is… a starter.

    He should start every 2nd or 3rd game and pitch until he allows a baserunner or two. Then pull him and replace him with the regularly scheduled starter.

    This will never happen, obviously, but if his problem is pitching with men on base, it’s the most sensible solution.

    Vote -1 Vote +1

    • TKDC says:

      A more reasonable (but still would never happen) way of doing this would be to have a two-man 8th/9th inning combo with him and another guy. Lincecum comes in to start the 8th, and if he pitches a clean inning, he also starts the ninth. If at any time a runner reaches, you come in with the other guy. Both guys will have long outings, but they’ll also have short outings. It works out in the wash.

      Vote -1 Vote +1

    • DP says:

      Every inning starts clean. The leverage of the game is different based on the score, but at least no one is on base. Does this severely limit his usefulness, yes, but that is what they have to live with (that or release him and find someone else).

      Vote -1 Vote +1

      • Johnston says:

        “release him and find someone else”

        Now there is a much better idea.

        Vote -1 Vote +1

        • Jason B says:

          …except they’re on the hook for what, $16MM next year? Sunk cost, sure, but you might as well find a role where he can be useful (or maybe even excel).

          Vote -1 Vote +1

    • Catoblepas says:

      Oh man, I sort of love the idea of a team using a starter that’s particularly bad with men on, or the second/third times through the order, as the beginning of a bullpen start once every five days. I think Jeff said he thought that might be the best plan for the Angels at this point, given that finding a starter will be so tough — imagine if they had Lincecum to do it with!

      Vote -1 Vote +1

    • MadMonk says:

      he can starts all the games v. the Padres.

      Vote -1 Vote +1

  4. How does Lincecum’s performance drop compare to the norm? Is it unusual for a pitcher to perform worse with men on base?

    Vote -1 Vote +1

  5. KK-Swizzle says:

    Could be a 7th or 8th inning guy as well as long as he is starting the innings off…the takeaway is that he probably should never be viewed as a “fireman” type who comes in with guys on base and works out of jams

    Vote -1 Vote +1

  6. Paul says:

    He didn’t need any new-found velocity to dominate as a reliever during the playoffs. Actually, now I’m curious as to why he dominated during the playoffs. Sample size issues acknowledged, but where there any pitch usage changes?

    Vote -1 Vote +1

    • DP says:

      I don’t think anyone has ever looked into it, we just assumed it was the ability to throw harder because he didn’t have to sustain the effort over many innings.

      Vote -1 Vote +1

      • DrBGiantsfan says:

        Having watched his playoff relief appearances, I can say if he gained velocity, it was only by 1-2 MPH at the most. I think it was mostly a sample size thing with a possible added psychological benefit of not having time to think too much.

        Vote -1 Vote +1

      • Yeah, that’s the common thought, that his velocity would go up but I remember going in to look at his numbers and there was not really any change in velocity, he was still pitching like a starter.

        The reasoning I came up with regarding how he performed well during the playoffs was tied to his performance in the season. He struggled in the first half, as he went on a massive weight-loss diet during the off-season because the year before, he went on a massive weight-gain diet (called the In N’ Out diet by some because he talked about eatting more Double Double’s) and the extra weight caused issues with this legs joints. It seemed like he needed to figure out his mechanics all over, and finally did over the ASB.

        People forget but for most of the second half, he was actually the best pitcher on the staff, until he apparently hit the wall near the end, and performed poorly, with Bochy choosing to put him in the bullpen. DrB had the best explanation for that, noting his yo-yoing weight, his stamina was probably not the best by the end of the 2012 season.

        But, as I noted above, he was in good form: when well rested.

        And that seemed to mostly work to form during the playoffs. He relieved so he didn’t pitch as many innings, so he was relatively well rested. That allowed him to keep his fine-tuned mechanics going right, so that he could go on short bursts and kick ass as a reliever. He did do poorly as a starter, but that’s small samples: few ever compile a 0.00 ERA over the playoffs, was this just regression to the mean or was this him being a poor starter? Can’t say either way.

        But I think this theory works for why Lincecum did so well as a reliever in 2012, he was already in good form, he was just too tired to repeat it deep into the game as a starter, but was able to as a reliever.

        Vote -1 Vote +1

  7. DP says:

    DC did a nice job of presenting the Lincecum problem and then chose the solution that would get the most responses. Smartly done.

    Vote -1 Vote +1

  8. Dave says:

    Maybe he just needs to figure out how to not let the guys on base distract him.

    Vote -1 Vote +1

  9. DrBGiantsfan says:

    I do not expect Timmy to gain much velocity in the bullpen and I do not expect him to have sustained success there unless/until he works out his mechanical issues. Even then he will still have to deal with adjusting to his reduced velocity.

    Vote -1 Vote +1

  10. LONNIE says:

    Last year when they re-signed him for 2 years/35 million I was wondering why they did it..now I bet they are wondering the same thing. It was a HUGE overpay at the time but at least its only for 1 more year.

    Vote -1 Vote +1

  11. This is why I argue that generalist’s analytics suffer when compared to the analysis done by specialists who follow one club. Dave’s analysis is nicely done, I would note, but he’s missing factors and flavors to this situation that a Giants fan would know, which changes how this scenario should be viewed, in my opinion.

    First of all, a month ago or so, we would not be having this conversation or blog post. His ERA on July 22nd was 3.65 and on July 25th still 3.96. Neither would warrant him being sent to the bullpen. Some might argue that his ERA was very high most of the season until his good streak of starts, but that was the result of a poor two starts to begin the season: from the 3rd start to his relief appearance on July 22nd, covering 18 starts and a relief appearance, he had a 3.09 ERA. He’s actually been pretty good for the vast majority of the season.

    What has changed so greatly over the past month to judge that he’s no longer viable as a starter long-term? Sure he needs a pause now, he has put up a string of poor starts, but the pitcher who just a month ago had a sub-4 ERA is still in there somewhere.

    Second, the author got the news probably from the headlines. Because, if you read through all the reports and quotes from Bochy, this “experiment” will be re-evaluated after Petit’s start. Thus, Lincecum could become a reliever but he could also just return to the rotation after a skipped start. It all depends on when he figures things out with his mechanics.

    The intent – if you read from here: http://blogs.mercurynews.com/giants/2014/08/25/lincecum-rotation-one-start-possibly/ – is to hit the pause button, give him a break to figure out his mechanics, and when he is ready to return to the rotation, he’ll be put back in. Meanwhile, while he’s figuring things out, they might use him as a reliever if necessary.

    So this was a nice thought experiment on the merits of Lincecum as a reliever, but that’s jumping the gun a lot as he’s not going into relief full-time, at least based on what’s been said here, he’s taking a break to figure things out.

    And this is a common thing that Bochy has done with his pitchers while with the Giants. He has previously taken out Cain and Sanchez out of the rotation to help them straighten things out, and in both cases, same situation, the pitcher’s out to figure things out and he’s back once he does. In both cases, both returned to the rotation after one skipped start, Sanchez famously throwing his no-hitter in his return (he stupidly changed his mechanics to match his idol Johan Santana, while in WBC, despite the fact that he towers over Johan and thus need different mechanics; the skipped start was because he had been refusing to change back) and Cain threw a 1-hit shutout in his first start back. I think it’s pretty clear that Lincecum is returning to the rotation at some point, just not this coming start.

    Third, as I noted, he put in a relief appearance. Basically there is a before and after: before the relief appearance, 3.67 ERA; after the relief appearance, 9.49 ERA. He saved that game, a long 14 inning game, taking a bullet for the team by pitching on his in-between start pitching day, but look at how many games the Giants have lost since, 2-4 since (the team was 13-7 to start the season with him starting), so roughly two wins. It probably would have been better to go with Colvin in relief (he was asked/offered to go next) since the Giants got a 3 run lead in the top of the inning and hopefully he could have got us out of the inning without using Lincecum. But who knows, he was already throwing in the bullpen, getting warm, perhaps the damage was done already by the time he came in.

    So I think these bits of info helps to paint a different story regarding this situation. As fun as it is to think of using Lincecum as a reliever (it should be added that he has a rubber-like arm, he still throws pole-to-pole THE NEXT DAY AFTER a start, and he still doesn’t ice his arm after starts, so he could be an almost every day reliever, for multiple innings), he’s been a very good starter for most of the year before being used in relief in-between starts, so it’s premature to push him into a relief role.

    And I think he can be a great reliever. He could be a good closer, but given his rubber arm, I think that’s selling him short. I think he can be an uber-reliever, much like he was used in the 2012 playoffs.

    He can pitch middle relief and save the rest of the bullpen by going 3-4 innings. He can pitch set-up role, even with the issues noted above, there are many situations where you can start your set-up guy at the start of the inning. He could close the game, whenever the closer has been used too much and need rest or even when the closer doesn’t have it that day and he could be brought in then to finish the close. Then he can do it all again more times than other relievers can do it, putting in more appearances and innings than usual relievers, because he has a rubber arm.

    I would also note a dichotomy regarding his record with men on base: during his career, he has only been significantly bad with men on 3B.

    Split G PA SO/W BA OBP SLG OPS
    RISP 246 1610 2.04 .231 .324 .360 .683
    — 247 3785 2.87 .236 .309 .360 .669
    Men On 248 2757 2.49 .231 .309 .364 .674
    1– 245 1147 3.52 .230 .288 .370 .659
    -2- 222 617 1.90 .213 .316 .333 .650
    –3 133 208 1.08 .297 .417 .510 .927
    12- 180 369 4.65 .227 .279 .322 .601
    1-3 123 172 2.35 .273 .343 .453 .796
    -23 88 134 1.09 .198 .388 .307 .695
    123 72 110 6.50 .215 .227 .312 .539

    So it looks more like he can handle relief except when there’s a guy on 3B. There are a lot of relief situations where a manager can avoid using Lincecum in this case and bring him in later in the game.

    Vote -1 Vote +1

    • Machaut says:

      There are a lot of logical leaps in this comment, but I’m only going to address the last point:

      Would be curious to see how his numbers with a man on third compare with league averages. Obviously, having a man on third sometimes means he’s given up a triple; more often it means someone has already reached and subsequent plate appearances have advanced the runner. Neither scenario is indicative of a shutdown pitching performance. Furthermore, is there any credibility at all to using base-specific runner splits? I’m not claiming that there is no connection whatsoever, but I’ve never read anything suggesting that a pitcher alters his mechanics significantly depending on which base a runner is on (bases empty v. runners on actually changes mechanics, which is another story entirely). Correlation is not causation, and just because a pitcher is awful with runners on third does not mean those runners are causing the drop in performance.

      My suspicion is that pitching with runners on third works much like “clutch”; some are better at it, some are worse, and in the end, everyone averages out to about zero with an even distribution of outliers, proving…nothing.

      Vote -1 Vote +1

      • Good point about the sampling rate, but that’s true for the split for bases empty vs. men on too, I would think, I mean, it takes 8 years plus for BABIP to prove out.

        I wasn’t trying to suggest that Lincecum can alter his mechanics, but approach by the pitcher does change by batter, Rob Neyer did an analysis of how Kirk Rueter’s peripherals changed depending on the runner state, empty vs. man on, and how Rueter seemed to accept more walks with men on, but gave up much less homers.

        But since the point of this article was that he was not that good with runners on, I looked at the stats and basically he has not done that well with runners on 3B (plus any other runner states), but OK with any base state when no runner on 3B, during his career. That was a bit tongue in cheek, and yet, that is what his stats have been over his career, though I don’t have them for just his latest years, so perhaps they have gotten much worse.

        Vote -1 Vote +1

  12. Julian says:

    Is there any chance he’d be more consistent with his mechanics day to day if he pitches more often? I can see that really helping him too.

    Vote -1 Vote +1

  13. Rubén Amaro, Jr. says:

    Boy oh boy, that contract extension looks even worse now.

    Vote -1 Vote +1

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>