Mar
17

By The Book, a look at an ideal Yanks batting order

By

The Yanks batting order kind of writes itself, right? The top four guys are basically set in stone, there’s an obvious spot at the bottom of the order for Gardner or Cabrera, and the five through eight spots can be filled capably in almost any order. So why am I revisiting the topic? Because I love reading Beyond the Boxscore, and Sky Kalkman wrote a post today about lineup arrangement. It’s based on The Book, which is next on my stack of books (after I finish The Fielding Bible Volume II). Sky walks through the reasoning behind the importance of each batting order spot, and concludes with the order of on-base importance. We’ll jump to that part, and then get into what this means for a potential Yanks lineup.

As an exercise in the ability to not make an out, this is the order of lineup spot importance: #1, #4, #2, #5, #3, #6, #7, #8, #9. Power changes things around here — for instance, A-Rod is the best OBP guy on the team, but he’s not going to hit leadoff because he has a ton of pop. That means he’s best suited for the fourth slot. Make sure to click over and read Sky’s rationale for each position. Things might make a bit more sense after doing so.

With all this in mind, here’s how I would arrange the Yanks:

1. Derek Jeter
2. Mark Teixeira
3. Johnny Damon
4. Alex Rodriguez
5. Hideki Matsui
6. Jorge Posada
7. Nick Swisher
8. Robinson Cano
9. Brett Gardner

If Xavier Nady wins the right field job, I’d just as easily slot him into the seven spot. Ditto Cabrera and the CF job/ninth spot.

This actually isn’t that absurd when you think about it for a bit. Plenty of fans want to see Derek Jeter bat leadoff because he’s well-suited for the job. He gets on base at a good clip and doesn’t hit for much power these days. In fact, Damon hits for considerably more power, so Derek is arguably the best candidate for leadoff hitter on the team. Since this is an idealistic scenario and not based on the reality of the club, it was a no-brainer to hit him there.

One strange aspect you might see is Mark Teixeira hitting second. As Sky notes, “the #2 hitter comes to bat in situations about as important as the #3 hitter, but more often.” So you want your #2 hitter to be better than your #3, since he will see more at bats, even if it’s just a few more per season. Teixeira is the Yanks second best hitter. He has power, which keeps him out of the leadoff role, but not as much as A-Rod, which keeps him from the fourth slot. The next best place for him, then is the #2 hole.

Having Johnny Damon hit near the top of the order is an advantage. He consistently has an above average OBP, has some speed on the basepaths, and has gap power. As a lefty at Yankee Stadium, he can also knock some balls out of the park. Under these guidelines he seems like the ideal #3 hitter — good average, good OBP, decent power, not one of the top three hitters on the team.

As for #5, I had a bit of trouble picking someone for here. As Sky notes, the old-school thought is that this is your second-best power hitter. However, that’s not exactly the case. Because of the guys hitting ahead of him, he can be valuable with all sorts of hits. So you want a high-average type player. That’s Matsui, who hasn’t hit below .285 in his six big league seasons. he has decent power, though we saw that come down a tick last year. Still, if he can manage a full season of his 2008 line — .294/.370/.424 — I don’t think any manager in the game would hesitate to hit him fifth.

The only notable changes to the probably lineup are in the top three spots. When the Yanks open on April 6 the order will almost certainly go Damon, Jeter, then Tex. Does that make a huge difference? I don’t know. Intuitively, I don’t think it makes much, if any difference in how the team scores its runs. Maybe Jeter hits into fewer double plays out of the leadoff spot and that creates a few runs for the Yanks. Maybe Johnny hits his home runs with a few more guys on base. Maybe Teixeira is in a greater position to score runs out of the #2 spot than the #3. I think that more than anything, this shows how blessed the Yanks are at the top of the order.

Categories : Analysis
  • radnom

    Damon’s OBP last season was about 20 points above Jeters.

    Sure, I think that was just a combination of Damon on an up year and Jeter on a (hurt?) down year, but you never know with players on the backside of their career.

    Also, I’m not sure I buy that Damon has “considerably” more power than Jeter. Jeter averages 17 hrs a year, and Damon 15..they both have had a few seasons where they touched the low 20’s but generally are in the mid teens, and both hit 12 in 2007. I agree that Damon has a bit more pop than Jeter at this point, but it is negligible. It certainly doesn’t warrant taking Damon out of the lead off spot considering the OBP. (btw, just considering ‘pop’ in general, Jeter has a much higher career slg than Damon, further making the point)

    • http://theyankeeuniverse.com Moshe Mandel

      There are plenty of reasons to flip Jeter and Damon in the lineup, without all of the advanced stats- Jeter hits a lot more grounders, at this point in their careers they are similar in terms of OBP but Damon has more power, and Damon makes more contact.

      • A.D.

        Going off slugging %, Damon had more power last year, but not the previous 3. I’d say chances are power is essentially a wash.

        • http://theyankeeuniverse.com Moshe Mandel

          Fair enough- so Ill go with the Damon making more contact and hitting much fewer grounders/less double plays as my main arguments.

          • radnom

            Do you have numbers for that assumption?

            • http://theyankeeuniverse.com Moshe Mandel

              Damon has a higher contact rate, a lower K%, and a GB/FB ratio often half that of Jeter.

    • http://www.beyondtheboxscore.com Sky

      CHONE projections:

      Jeter: .294/.366/.415
      Damon: .276/.351/.417

      Similar. Damon’s 2008 was on the good side of his talent level and Jeter’s 2008 was on the down side of his talent level.

      • radnom

        Yes I said I believe that as well, and I’m sure those projections make the same assumption, I’m just pointing out it is not a “no-brainer” because Damon was better suited to lead off last season.

  • http://theyankeeuniverse.com Moshe Mandel

    It’s funny, I jus did a post on this a few hours ago, and using the same criteria came up with a different lineup, with some similarities.
    Derek Jeter (R) – High OBP, not much power, decent speed and good baserunning makes for a good fit. Do not want to waste Damon’s better power here.
    Jorge Posada (S) – Good bet to be third best hitter on the team, high OBP guy. Thought about Teixeira, but did not want to waste his power on empty base situations.
    Hideki Matsui (L) – Thought about Swisher, but a healthy Matsui is the better player.
    Alex Rodriguez (R) – This is where you put your best combination of power and OBP ability.
    Mark Teixeira (S) – Is the team’s second best hitter, but his power plays better here than up top at 2.
    Johnny Damon (L) – could have fit in a number of spots, but his combination of solid power and speed fit here.
    Nick Swisher (S)- Swisher and Cano were fairly interchangeable, so OBP went first.
    Robinson Cano (L)- 8th is a nice spot for him to work on his confidence.
    Brett Gardner (L)- Provides speed to turn over the lineup, but is likely to be the teams least productive hitter.

    • BJ

      This would be horrible. Tex at 5 and Damon at 6? I mean, I guess It would kinda make sense if Arod is out and it is Tex 4 Damon 5, but the way you have it you are wasting a lot of high OBP PAs by having them so late.

      • http://theyankeeuniverse.com Moshe Mandel

        All of the guys ahead of Damon are better OBP guys. Tex would be either 5th or 2nd, so I prefered 5th to sacrifice a little OBP for more slugging at the lower slot.

        • BJ

          I just think you are being too optimistic about Jorge, and also that lineup would require some big switches every time Molina gives him a day off, unless you want him DHing and not resting. I know there are no numbers to back it up, but players always claim they like to have a spot in the order and know when it will be, and not be shuffled arround each time you give the catcher a day off.

  • http://www.beyondtheboxscore.com Sky

    I pretty much agree with your lineup, Joe. ARod and Tex are pretty interchangeable, depending where you want a righty.

    My one disagreement is with Damon. He’s not one of the best five hitters on the team (Alex Rodriguez, Mark Teixeira , Nick Swisher , Hideki Matsui , Jorge Posada based on CHONE) and he’s not a HR guy (which is what you want in the third spot. I’d instead Matsui/Swisher/Posada there (probably Matsui to start, considering he’s less of a question mark) and bat Damon sixth or seventh, where he speed is valuable in front of some combination of Cano, Swisher, Posada, Swisher, and Gardner. I posted this elsewhere:

    Jeter
    Tex/ARod
    Posada/Matsui/Swisher
    Tex/ARod
    Posada/Matsui/Swisher
    Posada/Matsui/Swisher
    Damon
    Cano
    Gardner

    • http://www.spartacus.schoolnet.co.uk/CRsmithT1.jpg tommiesmithjohncarlos a/k/a Ridiculous Upside

      Here’s the question: You say that the #2 spot is more valuable than the #3 spot. How much more valuable? More valuable than the value gained by using handedness toggling to force opposing managers to burn through their bullpen faster in search of favorable platoon matchups?

      For example, given that the lineup importance order is #1, #4, #2, #5, #3, if we assume (for argument’s sake) that our top 5 hitters are ARod > Tex > Damon = Matsui = Jeter, then the optimal lineup is probably:

      1-Damon (L)
      2-Tex (S)
      3-Jeter (R)
      4-ARod (R)
      5-Matsui (L)

      But, if we wanted to strategically break up those two righties to protect against, say, a submarining righty specialist like a Chad Bradford, we’d instead go with:

      1-Damon (L)
      2-Jeter (R)
      3-Tex (S)
      4-ARod (R)
      5-Matsui (L)

      … which is our actual lineup. How many runs/wins per season do you think we’d be forfeiting by batting the inferior Jeter second and the superior Tex third? A lot? A little? I’m curious.

      • http://theyankeeuniverse.com Moshe Mandel

        If Damon=Jeter, wouldnt you just have Jeter, Tex, Damon, A-Rod, Matsui? That is RSLRL.

        • http://www.spartacus.schoolnet.co.uk/CRsmithT1.jpg tommiesmithjohncarlos a/k/a Ridiculous Upside

          Perhaps. (But it allows an alternative solution negating my false dichotomy.)

          SHUT YOUR MOUTH, MOSHE!!! STOP CIRCUMVENTING MY KOBAYASHI MARU!!!

        • http://www.beyondtheboxscore.com Sky

          Jeter and Damon are poor choices for the #3 hitter. One, Matsui/Swisher/Swisher are likely better and Jeter/Damon get most of their value from getting on base and hitting singles and doubles. You get the most leveraged value from the #3 hitter if he’s a heavy-HR guy.

      • http://www.beyondtheboxscore.com Sky

        The platoon advantage is about .030 points of wOBA (off the top of my head). Given that at least three of the plate appearances at the top of the lineup come against the starter and not a LOOGY, the difference in talent of the players really needs to be only, say, .010 points of wOBA in order to keep the better one second. ARod/Tex are so much better than anyone else, they should be 2nd and 4th.

        • http://twitter.com/tsjc68 tommiesmithjohncarlos a/k/a Ridiculous Upside

          Thanks.

  • http://www.spartacus.schoolnet.co.uk/CRsmithT1.jpg tommiesmithjohncarlos a/k/a Ridiculous Upside

    Spots Six Through Nine

    The old-school book says the rest of the lineup should be written in based on decreasing talent. Hitting ninth is an insult.

    The Book basically agrees, with a caveat. Stolen bases are most valuable ahead of high-contact singles hitters, who are more likely to hit at the bottom of the lineup. So a base-stealing threat who doesn’t deserve a spot higher in the lineup is optimized in the #6 hole, followed by the singles hitters.

    It’s too bad that of our 6-9 hitters, the only basestealer is Gardner, who, torrid spring notwithstanding, is not ready to be considered anything more than the least-effective hitter in our lineup.

    If he manages to actually OBP, say, .350, it would be fun to bat him 6th behind Matsui and let him get on base and steal second and get chipped in on a single by Posada, Swish-Nady, or Cano.

    I wonder, though, the wisdom of putting a stolen base specialst at #6 if he’s right behind the high-OBP/slow running #4 and #5 hitters. Nails and Matsui are good at clearing the bases with homers. They’re also good at getting on base with singles and doubles. If Gardner works a walk or slaps a weak grounder to the left side to reach first behind them, it limits his ability to steal second base (since someone’s standing on it already.)

    • whozat

      “Nails and Matsui are good at clearing the bases with homers. They’re also good at getting on base with singles and doubles. If Gardner works a walk or slaps a weak grounder to the left side to reach first behind them, it limits his ability to steal second base (since someone’s standing on it already.)”

      F-in ARod. He’s so me-first, depriving Grittner of SBs by selfishly putting himself on second base.

      • BJ

        Actually, if Jeter was batting first, Brett’s stealing ability would actually be quite valuable in the 9 spot with all the grounders from the captain.

    • http://evilempire20.com/ Ryan S.

      We need a combination name for Swisher/Nady already. Swady? Nadsher? X-Swish?

      • steve (different one)

        Xick Nasher

        • http://www.spartacus.schoolnet.co.uk/CRsmithT1.jpg tommiesmithjohncarlos a/k/a Ridiculous Upside

          That sounds like a male pornstar.

          • whozat

            And, as such, is perfect.

    • http://www.beyondtheboxscore.com Sky

      If Gardner’s a .350 OBP, .380 SLG guy, you don’t want him ahead of better hitters like Swisher, Cano (probably), etc. His speed and OBP are a nice combination at the bottom of the order. Hitting in front of Jeter (when the lineup loops), his speed is important since Jeter’s a singles/doubles hitter.

    • http://www.beyondtheboxscore.com Sky

      Posada’s not really a singles hitter, right? BBs, HRs, 2Bs seem to be his skillset. I’d put Posada 5th or 6th with Swisher at the other one, then finish with Damon, Cano, Gardner.

  • Andy In Sunny Daytona

    I like your line-up. I would move Damon to the leadoff spot and move Jeter and Teixeira down one.

    Sincerely,

    Conventional Wisdom

    P.S. j/k

    • whozat

      IETC

      • BJ

        can i get a definition for IETC? I missed that one.

        • http://www.riveraveblues.com Joseph P.

          I enjoyed this comment.

          • BJ

            Thanks, i guess, but that does not really… oh, funny

  • http://actyankee.blogspot.com Matt

    Joe, you’re forgetting the game has a heartbeat! This well-reasoned, highly-rational lineup totally kills that heartbeat! HOW DARE YOU!

  • Giuseppe Franco

    Sounds to me like sabermetrics gone awry.

    Sometimes there can be too much thinking involved when creating a lineup.

    • frits

      Ya i know. like if they would put joba in teh 8th already then why would u worry about the lineup cuz the game is already won!!!!1 lol

    • whozat

      Given that the only real difference is that Damon has been moved down two spots, that seems to be a ridiculous assertion.

      • Giuseppe Franco

        No, not really.

        You forgot about the second best run producer on the team batting second.

        Perhaps you were too quick with your ridiculous response to notice.

        • http://actyankee.blogspot.com Matt

          There was also a good rationale for putting Tex second, considering he’s the second best on-base guy in the lineup. With the amount of times Tex’d probably get on base, along with Jeter, Damon could be a 100+ RBI guy and be a great run producer.

        • whozat

          Yes. That is an obvious byproduct of Damon moving down two slots.

    • http://theyankeeuniverse.com Moshe Mandel

      How dare we think about baseball!!! For shame!!

      Seriously though, this just makes you reconsider the concerns that a typical lineup has- ie basestealing from your leadoff man, contact from your 2 hitter, best hitter at 3. If those reasons do not make sense, why not come up with something that does?

      • Giuseppe Franco

        Why?

        Because the lineup is already set no matter what we think.

        • http://actyankee.blogspot.com Matt

          Why?

          Because re-thinking that “fashion” of lineup and changing it to something else could lead to better run production and an eventual shift of typical lineup construction in the future of baseball.

          • Giuseppe Franco

            Perhaps, but you’ll have to weed out most of the old school guys for that to happen because they don’t always put much stock in sabermetrics.

            Like I said, there’s certainly a place for it – no question about it – but the game isn’t always played with statistical formulas.

            A good mix of both sabermetrics and old school thinking is probably ideal.

            • whozat

              “A good mix of both sabermetrics and old school thinking is probably ideal.”

              Perhaps. Why are you so pissed about this particular sabremetric argument? Does it challenge your preconceived notions too much?

    • http://www.spartacus.schoolnet.co.uk/CRsmithT1.jpg tommiesmithjohncarlos a/k/a Ridiculous Upside

      Sometimes there can be too much thinking involved when creating a lineup.

      No, not really, there can’t.

      There should be no unquestioned ideas. There are no intellectual sacred cows.

    • http://www.riveraveblues.com Joseph P.

      You’re right. Talking about an A-Rod spread in Details magazine is so much more fun.

      • Giuseppe Franco

        Huh?

        Who said anything about A-Rod? Don’t be so defensive.

        People are debating the merits of a lineup that will never happen in a thousand years.

        • http://www.riveraveblues.com Joseph P.

          And you’re not debating anything. You’re writing it off without second thought. You can do that, and it’s fine. Just stay out of the discussion if you don’t like talking baseball.

          • Giuseppe Franco

            Thanks but perhaps you need to grow some thicker skin.

            I read every word of this post but it won’t change the order of the lineup that Girardi writes on his lineup card everyday.

            Didn’t realize disagreeing with the author of this piece was such a criminal offense.

            There is certainly a use for sabermetrics in the game but the specific lineup order really isn’t going to make any real significant difference in the grand scheme of things.

            Even Sky himself acknowledges that.

            • whozat

              Didn’t realize disagreeing with the author of this piece was such a criminal offense.

              That’s not what you’re doing. You’re saying “This is pointless, stop talking about it.” Which is, in itself, completely pointless. No one is making you read this. You wasted your OWN time by commenting. We’re talking about this because it is an interesting metal exercise. Furthermore, you’re not offering any thoughts about something else that you’d RATHER talk about. You’re just telling us to shut up because “it won’t change the order of the lineup that Girardi writes on his lineup card everyday,” when in reality NOTHING WRITTEN ON ANY BASEBALL BLOG has very much impact on actual baseball games at all. So why are you here?

              • http://actyankee.blogspot.com Matt

                when in reality NOTHING WRITTEN ON ANY BASEBALL BLOG has very much impact on actual baseball games at all.

                Except for getting Nick Swisher ;)

              • Giuseppe Franco

                If you don’t like what I have to say, don’t read it.

                It’s pretty simple.

                And I offered an opinion, sometimes we think way too much when constructing a lineup.

                It’s not really all that complicated and doesn’t always need to be analyzed down to a science, especially when the current lineup is already etched in stone – specifically the top 4-5 slots.

                My intention was not to incite some riot by the likes of yourself.

                But you go ahead and continue to rant and rave over an innocuous comment.

                • whozat

                  If you don’t like what I have to say, don’t read it.

                  Try taking your own advice some time. If you didn’t like what the author had to say, try ignoring it instead of snarking at someone who spending his time generating free entertainment for you.

                • Giuseppe Franco

                  Dude, take some xanax and call it a day.

                  It’s really not that big of a deal to get all crazy.

  • steve (different one)

    this is going to sound like a Grantbo post, but i simply don’t understand why so many people spend so much time quibbling over the batting order.

    as long as you put your better hitters up top and your worse hitters near the bottom, the actual order just isn’t going to make much difference.

    • http://www.spartacus.schoolnet.co.uk/CRsmithT1.jpg tommiesmithjohncarlos a/k/a Ridiculous Upside

      Because the only way to fix an outmoded, broken paradigm is to question it and upend it.

      I don’t forsee this moving Damon, Jeter, Tex, or ARod out of the slots they’ve become psychologically attached to, they’re veterans who grew up in a less-sabermetric era. But, if Girardi/Cashman read, understand, and agree with this logic, I’d love to see them make decisions about the futures of Montero, AJax, and Romine with an enlightened eye for optimal resource allocation.

      • http://evilempire20.com/ Ryan S.

        Either The Cardinals or The Red Sox will do this shit first, I can almost promise you guys.

        • http://actyankee.blogspot.com Matt

          I thought LaRussa hated statistical analysis.

          • http://evilempire20.com/ Ryan S.

            Does he? He’s the guy who started batting the pitcher 8th, and he always has a knack for thinking outside the box it seems – he might do that shit without even realizing it was a sabr-friendly move. Oh, and whatever team Joe Maddon is coaching in the next few years might do something like this too.

            • BJ

              Yeah, its kind of like a corollary to that whole BS Jim Rice HOF argument that “OBP didn’t matter back then”. Just because you aren’t looking at the actual numbers doesn’t mean you can’t analyze the situation and see that a certain strategy makes sense. If you have an automatic out, don’t put it before the best hitters in your lineup. Without the numbers you can’t prove anything, but you can still come to the right decision.

    • http://www.beyondtheboxscore.com Sky

      I agree:

      “Because we care, BtB would like to remind everyone that lineups are pretty overrated. Believe it or not, the difference between an optimized lineup and a typical, mildly foolish one you’ll see MLB teams use is only about one win over 162 games. It’s obviously worth getting right, but not any more than realizing Troy Percival shouldn’t be your closer or Joba Chamberlain belongs in the rotation.”

      • A.D.

        Hmm but what about totally optimized vs the antithesis…tops 3 wins? more?

    • http://www.beyondtheboxscore.com Sky

      I’ve already posted that the difference between a typical major league lineup and an optimized lineup tends to be about one win per season. That’s not a lot and shouldn’t be as important as other decisions.

      But one win is pretty useful, especially for a team expected to win between 90 and 95 games. I’d guess it’s about a 10-15% greater chance of making the playoffs.

      And one win on the free agent market has cost about $4.5M recently. That’s not cheap.

  • A.D.

    This is with Nails… but what about without him?

    Jeter
    Posada
    Damon
    Tex
    Matsui
    Swisher
    Cano
    Gardner
    Ransom

    or for Sky’s

    Jeter
    Posada/Matsui/Swisher
    Posada/Matsui/Swisher
    Tex
    Posada/Matsui/Swisher
    Damon
    Cano
    Gardner
    Ransom

    • http://theyankeeuniverse.com Moshe Mandel

      I go with the first one on Opening Day.

  • Elle

    They’ve been batting Damon second a bit this spring training–I was wondering if they were trying to get him used to that, and wanted to put Gardner in first. If his OBP is better than last year, isn’t he a textbook lead off hitter because of his speed and base-stealing, especially with Damon getting older and (presumeably) slower?

    Then again, assuming Gardner gets his OBP up is a big assumption, and Damon bat second when Posada had his first catching start so Posada could bat first and get the most at-bats in the fewest innings, so that may be all that is.

    • http://evilempire20.com/ Ryan S.

      They’re doing that so Gardner gets a feel for hitting in front of Damon. It’ll assuredly be Gardner 9th, Damon leadoff come opening day.

      • Elle

        Thanks, that makes sense. The line-up, it goes round and round.

    • anonymous

      They are batting Damon second alot to get Gardner/Damon used to hitting in that order. Damon will hopefully be batting alot while Gardner is on 1st distracting the pitcher.

    • A.D.

      They want to get Gardner more at-bats than Leone, or whomever is batting 9th, and while Gardner may be the worst bat during the season, he isn’t during spring games with players gone for the WBC & Split Squad.

      Thus it makes the most conventional sense to go that way.

  • mikef

    Hey Donnie Baseball was often parked at #2 in the early years, and I think his numbers in that spot ( with Ricky in front of him) were sick.

    I see Tex as a doubles-machine at the staduim, so batting him second would not be a bad idea – except that he would get pitched around to get to Damon, who is both quite streaky and whose “speed game” would be further limited with either Tex on base in front of him and/or Arod up behind him.

    Going back to mid-80’s, even though Donnie’s numbers suffereed a tad when moved down to 3rd in the order, Randolph became the prototypical #2 hitter and lengthened the lineup with his ability to get on base and patience to let Ricky run.

    So, if you could find me a #3 hitter instead of Damon (Matsui?) I would be all for Tesx batting 2nd.

    • whozat

      except that he would get pitched around to get to Damon, who is both quite streaky and whose “speed game” would be further limited with either Tex on base in front of him and/or Arod up behind him.

      The point is that the traditional belief in the “speed game” is far less important over the long-haul than the “getting on base game”

      His numbers are generally pretty stable month-to-month. And, frankly, if he’s so streaky, then we’d go through long periods where our leadoff hitter isn’t getting on base, which will damage the lineup more than the #3 hitter. He comes up fairly often with two outs and no one on base anyway.

  • BJ

    This was a very thought provoking article, but I think it left one aspect out of the question. It is great at analyzing each position based on what kind of scenario the batter is most likely to face, but it does not deal with the fact that the person coming up to bat after you has an affect on the kind of pitches you will get, which in turn may affect your ability to produce. For example, if you put Tex second, aren’t you giving pitchers more of an opportunity to go after the corners and not worry about allowing a baserunner than if he is right in front of Arod. I remember someone recently saying that a player’s batting improved ahead of Bonds or roid-Giambi because he knew he was going to get fastballs and could sit on them. Even if you walk him you have a chance at a double play with Damon at 3.

    I don’t know, has the idea of “protecting” your top hitters been proven to be just convention, or is the argument missing an important component.

    • whozat

      “has the idea of “protecting” your top hitters been proven to be just convention”

      Mostly, yes.

      • BJ

        Hmm, weird. Fangraphs is pretty much my homepage and I don’t remember reading about it. I guess even if there was a factor it would be hard to determine if it is random variation or legit.But like the article said, even this lineup adjusting would only really be worth 1 win, and that’s assuming that the people you put at each spot perform as you think they will. Not that big of a deal, but still fun to talk about.

        • whozat

          Pretty sure this has been commonly accepted sabremetric thinking for a long time now, so fangraphs would have no reason to be publishing articles about it now.

          • BJ

            So apparently the data does point towards batters performing independently of who is on deck, but the problem is that the sample size is so small and that you can’t really make a meaningful claim either way. Basicallly, people have published that they can not prove that there is an effect, but they have not proven that there is no effect. Bill James talks about it in Underestimating the Fog

      • BJ

        What I would really be interested in is ideal lineup by wOBA. I wonder if it would be at all different.

        My other problem with the article is that it seems to use the analysis of: at spot A, the batter comes up with men on base only x percent of the time. Well, if you start moving your high OBP guys decent power guys to #5 and your big homerun low OBP guys to #3, wouldn’t that affect the percent of time that the #4 hitter comes up with runners on base. I just feel like they are analyzing the pieces without looking at the big picture.

        • whozat

          Maybe, but is there really a lot of value in the 4 hole hitter coming up with one man on and two outs? I dunno. But I think that no one’s saying that Shelley Duncan is the ideal 3 hitter…but Adam Dunn might be. If there’s someone on base when Dunn gets to the plate, he’ll probably get driven home if he manages to make contact. If he makes an out…well, then the cleanup hitter gets to start off with a clean slate and a couple good hitters guaranteed to get a plate appearance after him.

          • BJ

            Really, the thing I think you are misunderstanding the HR thing.

            “The Book says the #3 hitter comes to the plate with, on average, fewer runners on base than the #4 or #5 hitters.”

            “The Book says the #5 guy can provide more value than the #3 guy with singles, doubles, triples, and walks, and avoiding outs, although the #3 guy holds an advantage with homeruns.”

            If you take into account that the #3 guy gets less PAs with runners on base than #4 or #5, the reason you put him #3 is not so he can knock in the runners, it is because you want to get him more PAs period so he can hit more HRs over the course of the season. That may be fine looking at each spot individually, but if you look at the big picture, a low OBP guy at #3 would take away from the value of the #4 and #5 hitter.

            • Am I the only Kevin?

              This is confusing. Isn’t the reason why the 4 and 5 guys have more baserunners on because they have both the 1 and 3 hitters fairly close in front of them? Conversely, the 3 hitter has the 9 hitter three spots in front of him? IF the 1-3 hitters are your best OBP guys (which seems to be conventional wisdom), then it makes sense that the 4-5 guys see a lot of guys on base.

              • BJ

                Actually, I think you are getting it. The author is making an argument for change based on results you get from the way things are currently done. The problem is that once you make the change, the same rules do not apply and you cannot assume that results will be based on the same relationships as the old way. Maybe the author’s methodology accounted for this somehow, but I didn’t see it. Underestimating the Fog

        • http://www.beyondtheboxscore.com Sky

          This analysis is addressing lineup by using wOBA. The whole approach in the book is done using linear weights adjusted by game situation. wOBA is just the rate form of linear weights.

          Were you looking for something specific presented in terms of linear weights?

          • BJ

            I haven’t seen the original article in the Book, so I guess it was kind of hazy where the numbers were coming from. The only stat mentioned in the BtB article was % of PA with runners on base.To me it seemed like total # of PA with runners in scoring position would be a better number to reference, as even if the the 3 spot is a few percentage points lower than the 5, if the total high leverage PA is higher at 3 that is where you stick the better hitter. Ok, so if you are saying it is based on linear weights and not the % thing, then I guess they look at the value in runs of each outcome at each spot. That makes more sense, but there is still the problem that when you change the way you make a lineup from the pretty much universal, low variance way it is done now, you can’t assume the values in runs for each outcome would be the same.

            I think one way you could do it, which is less generalizable but would work for this problem is that instead of redoing the weights for each spot on the lineup, you use weights for each outcome that include that include each baserunner situation, and use the same weights for each spot. Then for each lineup you simulate what would happen to find number of PA with each situation for each player, and see if they create runs in situations they are more likely to be in. For example starting at the beginning of the lineup, in 100% of Player A’s 1st PA, he is up with no outs and no men on base. Use Player A’s stats, to find what percentage of Player B’s 1st PA is in each situation. Use Player B’s stats (not situational, just how often they get a given outcome) to determine player C. When you do this again you get a 3rd out X% of the time, so for those you give Player D no men on base and no outs. You continue this until you get 27 outs for all possibilities, rounding percentages to 0 once they get low enough. You record what percentage of the time each player gets a PA in each situation per game, then multiply by 162. From here you multiply by the weight for each outcome for each situation and then by how often each player get that outcome to find total runs. You would have to do the whole thing again for each lineup, but if you write the algorithm basically you just have to run it again. This way it does not rely on run value data that came from the conventional wisdom system that you are trying to change.

            • http://www.beyondtheboxscore.com Sky

              Read the book. Then criticize it.

  • Pingback: A weaker lineup with Matsui limited | River Avenue Blues

  • Pingback: AL East chat at Beyond the Boxscore | River Avenue Blues

  • Pingback: BtB’s optimized Yanks lineup | River Avenue Blues

  • Pingback: Not another post on the Yanks lineup! | River Avenue Blues

  • Pingback: Led Zeppelin’s Houses of the Holy as a baseball lineup | River Avenue Blues