Why Do Managers Bat Replacements in the Starter’s Spot?

There are plenty of managerial decisions that I routinely do not understand. Most of them relate to the bullpen and how the manager selects a reliever for certain situations. The rest relate to lineup construction. We know that in the macro lineup order doesn’t matter that much, but that doesn’t exactly justify the decision to use a well below-average hitter in that position. Why place a poor hitter among your best? Why not move him to the bottom, so the good hitters have a better chance to get on base and knock each other in?

This leads to another managerial tactic that I never understood: placing a replacement in the starter’s batting order spot. When this happens with a bottom of the order hitter it’s usually no big deal. It’s when the manager replaces one of his top hitters with a replacement that baffles me. There has to be some reason behind it, since so many managers routinely do it. But that doesn’t make it any less perplexing.

Ron Gardenhire not only does this, but he’s doing it while one of his starters is on the DL. Orlando Hudson should be back from his wrist injury by week’s end, but until then Matt Tolbert acts as his stand-in. If this were just a defensive move it wouldn’t be a problem. After all, few teams can absorb an injury to their starting second baseman and replace him with someone who can produce similarly. The problem is that Gardenhire has not just penciled Tolbert’s name into Hudson’s spot in the field, but also in the batting order. In 45 PA this year Tolbert has gotten on base 10 times. Last year, in 231 PA, he had a .303 OBP. He belongs nowhere near the No. 2 spot, and yet Gardenhire has used him there for most games during Hudson’s absence.

Tigers center fielder Austin Jackson is currently day-to-day with back spasms. He left Sunday’s game and hasn’t played since. That shouldn’t be a big problem for the Tigers, since they have Johnny Damon, who for most of his career was a leadoff hitter, in the No. 2 spot. But instead of moving Damon up one spot and putting a hitter like, say, Brennan Boesch or Carlos Guillen in the second spot, Leyland has chosen to use Jackson’s replacements, Ryan Raburn and Don Kelly, in the leadoff spot. Raburn, maybe, makes a degree of sense. He had a .359 OBP and .378 wOBA last season, though he’s playing horribly this season. Kelly, however, is a plain bad hitter. He’s actually brought down his .290 career OBP this year, reaching safely just 13 times in 77 chances. There is no reason to hit him first and give him more PA than hitters like Guillen and Boesch.

These are just two examples I noticed last night. Many other managers do this, too. Jerry Manuel, for instance, used to bat Alex Cora second when giving Luis Castillo a day off. Hitting Castillo second is questionable in the first place, but at least he had a .347 OBP to help justify the move. But to replace him in the field and in the batting order with Cora just seems reckless. In only three seasons of his career has Cora’s wOBA broken .300, the last time in 2008 with the Red Sox. He has a .306 OBP this season. There are plenty of hitters on the Mets, including David Wright, who would fit better in that spot when Castillo sat. With Castillo on the DL Manuel has shown a shred of sanity, hitting Angel Pagan second.

The question still remains of why these managers employ this tactic. It can’t be because the other players are comfortable in their batting order spots. Manuel has messed with his lineup numerous times this season, moving Wright from third to fifth, to fourth for a game, and now back to third. Jason Bay has hit in the Nos. 3, 4, and 5 spots. Neither has hit second even one time. Alex Cora has 16 times. Similarly, Damon has hit leadoff three times and has hit third four times. Why in the world, then, play Kelly there for even one game? Gardenhire routinely hits Justin Morneau third when Joe Mauer takes a day off. Why, then, can’t he put a hitter better than Matt Tolbert in the second spot?

Again, on a macro level, lineup order doesn’t matter a great deal. But on a game-to-game basis, when anything can happen at any moment, why wouldn’t you want to put your best hitters near the top of the order, where they can get on base and knock each other in? Why put a player who makes out more frequently than other players in the lineup near the top? Why let them take one more turn at bat than clearly superior hitters? It just doesn’t make sense. That hasn’t stopped managers from doing anything, though, so I fully expect this trend to continue. But that doesn’t make it any less of a shame.




Print This Post



Joe also writes about the Yankees at River Ave. Blues.


47 Responses to “Why Do Managers Bat Replacements in the Starter’s Spot?”

You can follow any responses to this entry through the RSS 2.0 feed.
  1. For some reason team’s think move-them-over players are good to bat second. Wasting outs in front of their best hitters, I don’t understand how they can look past how idiotic it truly is.

    Vote -1 Vote +1

  2. NM says:

    Jerry Manuel is THE KING of doing this. Just like when stuck Catalanotto at 1B and batted him cleanup the night the Mets DFA’d Jacobs, not to mention the countless times hes batted Cora 2nd….

    Vote -1 Vote +1

  3. Mark says:

    Which is exactly why Gardy doesn’t bat Mauer 2nd. In his mind, turning Mauer into a “#2 hitter” would be a waste.

    You missed when he batted Punto leadoff recently when he replaced Span in CF.

    Vote -1 Vote +1

    • Bryz says:

      Punto is a better hitter than Tolbert (still not a good hitter though), yet that’s the first time since Hudson was out that Gardy put Punto higher in the lineup than Tolbert.

      Vote -1 Vote +1

  4. RJ says:

    I don’t know if any more recent work has been done on the topic, but didn’t earlier studies/simulations indicate that the difference between the best and worst possible batting orders over a complete season was relatively small – on the order of (at the upper end) a couple wins?

    While teams obviously should strive to deploy their resources in the most effective manner, it seems doubtful that the difference between Tolbert hitting second and Tolbert hitting eighth over a two week period will make much of a difference to the Twins overall season – especially when compared to the already incurred cost to the Twins of having to play Tolbert over Hudson at all.

    Vote -1 Vote +1

    • BrettJMiller says:

      The fact that is has a minimal effect on the season still doesn’t make it smart. It may not hurt much, but if it hurts at all, it’s a bad, and easily avoidable decision.

      Vote -1 Vote +1

    • Jason B says:

      Even if the difference is relatively modest, why take the chance? Why not place a .360 OBP / .350 wOBA in the two hole and the .290 OBP / .270 wOBA in the nine hole, and get an extra 5-6 AB’s out of the good hitter over the course of a couple of weeks? There’s plenty of variability and random fluctuation that’s beyond a manager’s control already; he may as well go ahead and control the key decisions that he can (like lineup construction).

      Vote -1 Vote +1

      • Lucas says:

        Ballplayers and managers are routine- and habit-driven almost to the point of superstition. Which is sort of dumb, but it provides a pretty good defense to bat-Tolbert-second decisions by managers: if lineup order doesn’t matter all that much, why rock the boat and mess with players’ sense of order in the universe by moving them around?

        Vote -1 Vote +1

    • Jason B says:

      Yeah…yeah, what Brett said (more succinctly than I).

      Vote -1 Vote +1

    • The A Team says:

      Let’s not forget that just because 10 runs ~ a win, that doesn’t mean you shouldn’t worry about an extra 5 runs over a season. On the margin, each run could potentially represent a win.

      Vote -1 Vote +1

    • Rich says:

      “A couple of wins” is not minimal. Teams pay $15M a year for a guy worth “a couple of wins”.

      Vote -1 Vote +1

      • Jason B says:

        Agreed – I think “The A Team” was arguing the same point and in agreement with us. He was basically saying that ~10 runs scored/saved roughly equates to a win, but you never know where the marginal wins will come from so you had better score/save all that you can.

        Like batting your cover-your-eyes bad hitters lower in the order, for instance…

        Vote -1 Vote +1

  5. Schu says:

    Comfort level of the players involved. People like routine and that means penciling them into the same spot in the lineup day in and day out.

    Vote -1 Vote +1

    • freshhops says:

      This seems crazy to me. The first inning is basically the only inning where you know who will come to the plate and who will lead off the inning. After that, it’s basically random. Does it really make that much difference to Albert Pujols that Matt Holiday is hitting behind him; maybe it gives him protection, but that seems like an effect on the opposing pitcher. I’m pretty sure Albert has hit successfully regardless of who was following him in the order, so I don’t see why Rasmus-Pujols-Holiday (or whatever) is the order you need to have for Albert to hit best.

      Vote -1 Vote +1

  6. Visnovsky says:

    Twins fans have been complaining non-stop about this for the last two years.

    Vote -1 Vote +1

    • Not David says:

      Two years? It’s certainly been longer than that.

      Gardy used to do the same thing with Redmond, he’d put him in Mauer’s spot in the order.

      Vote -1 Vote +1

      • Bill@TDS says:

        This has always been the biggest of my many complaints about Gardy, not because it’s the most important, but because it’s the most obviously wrong-headed. Batting Redmond #3 when Mauer needed a rest was the worst.

        Vote -1 Vote +1

      • Erik says:

        I believe Gardenhire thinks that each spot in the batting order is for a designated position – meaning the #2 spot in the order is for the 2b, #3 is for the C (as in Redmond for Mauer), etc.

        Gardenhire just does not see things as he should – which is disappointing as a Twins fan.

        Vote -1 Vote +1

  7. JR! says:

    Gardy’s biggest atrocity was routinely replacing Mauer with Redmond in the 3-hole…

    Vote -1 Vote +1

  8. Dylan says:

    I think it’s pretty clear why they do this. The players have a comfort level at their certain spot. When guys move around, they start to attack AB’s differently, thinking they need to live up to the conventional role of that spot in the order. Jose Reyes started looking to crush everything when he batted 3rd. David Wright often becomes indecisive batting 2nd, unsure of whether to let guys steal or attack the AB. Sadly, players are human, and this is one of the spots where that actually matters. There are certain situations that should supersede this (Redmond should never have batted 3rd for example) but for the most part it makes sense. You’re better off having a slightly worse order for one game, which statistically should have an extremely minor impact on your season, than messing with a starter’s head.

    Vote -1 Vote +1

    • Jason B says:

      Sorry, but I just don’t buy this argument at all. It sounds all well and good, but is there any evidence of it in the least? All this “comfort level” and “different approaches” and “messing with heads” sounds like a weak excuse for abject laziness.

      Vote -1 Vote +1

      • Don A says:

        Actually, since lineups don’t matter that much, the burden of proof is on the other side. You better be damn sure that you’re not messing with people’s heads for the short, medium, or long term before you just go willy nilly changing their work environment.

        Not to mention, if a player hits a slump after you make a change, you WILL get blamed, justly or unjustly.

        Vote -1 Vote +1

      • Dylan says:

        First of all, no, there is no evidence because something like this would be almost impossible to measure. Asking for evidence of this is ridiculous because there is no reasonable statistical way of tracking it.

        But more important, since it has been proven that batting order has very little impact on a game, it really doesn’t matter whether there’s evidence because, like I just said, there’s very little impact. So unless there’s evidence AGAINST it, there’s no real reason not to do it.

        So let me ask you, is there evidence against it?

        Vote -1 Vote +1

      • Jason B says:

        That’s…not the way arguments work. You don’t gather proof against something. The burden of proof is on someone stating a claim.

        And you’re right – batting order has very little impact. But it *does* have (statistically provable) impact, and it’s *in favor* of moving the crappy hitters down and the better hitters up. So…yeah. This side has statistical evidence marshaled in its favor. The other side has… “messing with his head” and “creatures of habit” claptrap.

        Vote -1 Vote +1

      • Dylan says:

        There is no “claptrap.” The stats happen to back it up. Wright is a worse hitter statistically in the 2 hole as well. The only evidence at all in that situation goes against what you’re saying. And yes, when your eyes tell you something, and the statistics indicate that it pretty much doesn’t matter, using the logical argument that humans are in fact creatures that tend to thrive in situations of habit is a perfectly valid one. It would be one thing if there were actually a significant impact caused statistically by this, but there is not.

        Vote -1 Vote +1

      • Dylan says:

        And there are, BTW, plenty of statistics backing up the fact that human performance is improved by working in consistent conditions, there’s a reason companies keep working conditions for their employees fairly similar day to day. The fact that it would be basically impossible to back this up with baseball statistics (what the hell would you look for? you would need stats of a guy consistently batting in a certain spot for an extended period of time, which, to my knowledge, I don’t have the technology to do without manually scanning over box scores) doesn’t make these invalid. It would be completely wrong to ignore pretty well established facts about the human race just because it cannot be proven in a baseball atmosphere.

        Vote -1 Vote +1

      • Rich says:

        “Actually, since lineups don’t matter that much, the burden of proof is on the other side”

        You care to show some proof for that? A couple of wins matter.

        Vote -1 Vote +1

      • Jason B says:

        To your point about humans liking to work in routine or consistent conditions – perhaps. Some absolutely do crave routine and structure, some absolutely don’t.

        You jumped from point A to point Z with the Wright example; you seem to be saying because one player hits better in spot x rather than spot y, therefore let everyone hit where they want to in the batting order? Holy freaking lame argument batman!! You can find a *single* example to “prove” or “disprove” anything (and I use the term prove *very* loosely). The *body of evidence* shows that lineup construction matters slightly; not a lot, but as Rich says a win or two over the course of a season can be meaningful, as those marginal wins can mean everything to a team contending for the division or wild card. And the body of evidence shows that arranging the batting order generally from ascending to descending OBP is the best way to achieve those marginal wins. It should be pretty intuitive that you want a guy with a .415 OBP to get more plate appearances than one with a .290 OBP, no? Outs are the most precious resource in baseball – don’t waste ‘em on the Willy Tavareses of the world!

        Vote -1 Vote +1

  9. Mark says:

    Besides, batting Tolbert or Pluoffe in the 2 hole for a few games is nothing compared to not having Mauer hit leadoff on a regular basis. I’m suprised that the Gardy bashers never harp on that fact.

    Vote -1 Vote +1

  10. Glanzer says:

    It is maddening as a Twins fan to sit there and see light-hitting backup middle infielders in the 2-hole, getting more at-bats than someone like the red-hot Delmon Young. Gardenhire is absolutely set in his ways and seems to believe that Span will get on with a hit, the #2 guy will give up his at-bat to move him over, then Mauer and Morneau will try to drive him in. That rarely works out. It was even worse when Span got a day off Sunday and they threw Punto in the leadoff spot and Tolbert in the 2-spot. I will never agree with that rationalization that players are more comfortable in certain spots in the batting order.

    Vote -1 Vote +1

  11. Detroit Michael says:

    This isn’t a new practice at least. I remember Sparky Anderson one time batting Harry Spilman in the clean-up slot when Kirk Gibson was injured. If you don’t remember who Spilman was, that sort of proves my point.

    Vote -1 Vote +1

  12. Mark says:

    The way Delmon’s hitting, you don’t want him worrying about taking pitches to Span can steal. And we want him driving the ball, not bunting and hitting ground balls to the right side, as the Twins expect their #2 hitters to do.

    Vote -1 Vote +1

  13. Coby DuBose says:

    Leyland hit Clete Thomas in the #3 hole last year when he filled in for Ordonez.

    It’s routine for him.

    :|

    Vote -1 Vote +1

  14. Dwight S. says:

    As far as Leyland goes he likes to put the guy that fills in in whoever he is filling in for spot. So if Raburn is filling in for Jackson he will bat him leadoff, if Kelly is in for Damon he’ll bat him 2nd and so on. His reasoning is he doesn’t like to shuffle up his lineup because he thinks certain players get accustomed to where they hit in the lineup and won’t hit as well if you move them. Personally I find it laughable, I have a hard time believing Damon is going to hit any different whether he is 1st or 2nd or Inge 6th or 7th. It’s just another questionable move he does.

    Vote -1 Vote +1

    • TJ says:

      And yet Ramon Santiago has hit 2nd, 7th, 8th, and 9th this season. So I guess sometimes traditional baseball wisdom (contact #2 hitter) takes precedence over being accustom to where you hit in Leyland’s book too.

      Vote -1 Vote +1

  15. Jon Sciambi says:

    Completely agree that it’s dumb. But the idea that these guys want roles and don’t want to be taken out of their comfort zone is accurate; they have been brainwashed accordingly. Look at what Joe Poz wrote the other day about Daric Barton bunting…I saw him do it in the 1st inning against Dontrelle Willis and was baffled. However, I think every guy in that clubhouse thinks it’s a good play, a selfless play. And Billy could go down there and give them run expectation tables for hours explaining why it’s dumb and they aren’t gonna change their minds. I’ve had the conversation with coaches and players too many times. Even if you go “good process” as a manager every time (correct batting orders, more efficient closer usage, etc.), as soon as you get a bad result, you re going to have problems.

    Now, does it matter if the players are uncomfortable with their roles? Does it matter if they’re pissed? I lean towards “not really.”

    But here’s my point…It’s not quite as easy to make these changes as it seems because it takes some courage. When you are consistently making unpopular choices it’s not so easy taking the criticism from the media and players. The smartest thing to do isn’t always as easy as we make it seem. And yes, I put myself in the “we” category as I’d like to see this stuff change but I get to see just how difficult it is to do from the inside.

    Vote -1 Vote +1

    • Chet says:

      “But here’s my point…It’s not quite as easy to make these changes as it seems because it takes some courage. When you are consistently making unpopular choices it’s not so easy taking the criticism from the media and players. The smartest thing to do isn’t always as easy as we make it seem. And yes, I put myself in the “we” category as I’d like to see this stuff change but I get to see just how difficult it is to do from the inside.”

      Is this an excerpt from the Obama Biography?

      Vote -1 Vote +1

  16. Damnedifino says:

    A few years ago Gardy used a bad knee hobbled rookie, Jason Kubel, as his lead-off hitter for several games because he was replacing the regular lead-off hitter. Go figure. Gardy is usually a pretty good tactician, but I’ve never understood his insistence on this wasteful batting order strategy.

    Vote -1 Vote +1

    • Erik says:

      Looking at it now, I would be just fine with Kubel in the leadoff spot for the Twins – assuming Span needed a day off.

      What is wrong with the Twins is this – they have the wrong guys on their 25-man roster. If they had the right guys on their active roster, these decisions would be a lot smarter.

      Vote -1 Vote +1

      • TFINY says:

        Who would you have on? I’m guessing no Casilla, or Tolbert, or maybe Harris. Slama on, Crain off?

        What do you see as the biggest problems?

        Vote -1 Vote +1

  17. Damnedifino says:

    Focus Erik. The operative words are “rookie”, “lead-off” and “bad knee”. Nevertheless, Gardy put Kubel in the #1 spot for several games simply because he was the usual lead-off hitter’s defensive replacement.

    Vote -1 Vote +1

  18. Cora hits second because he’s Gangsta.

    *Face meet palm*

    Good article. It’s a subject that routinely blows my mind. Jeez, I put together more logical batting orders on MLB 2K10.

    Vote -1 Vote +1

    • Gilbert says:

      In video games you don’t have to have your bench coach tell one of your $5 million players to grab a bat and get on deck when they are following a different person.

      Vote -1 Vote +1

  19. 81 says:

    Joe Girardi has a decent track record of not doing this

    Vote -1 Vote +1

  20. WhatLeylandNooooooo says:

    This isn’t directly related to this post but a recent decision made by Jim Leyland made me think of this post. Last night Jim Leyland batted Gerald Laird second which is a move that I feel deserves an entire fangraphs post vilifying the reasons why it was such a horrible decision.

    I’ll get it started. These are Gerald Laird’s 2010 offensive statistics:

    OBP: .255
    SLG: .239
    OPS: .494
    wOBA: .230
    wRC+: 35

    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>