Jump to content


Yo! You're not logged in. Why am I seeing this ad?

Photo

Lineup Construction


This topic has been archived. This means that you cannot reply to this topic.
124 replies to this topic

#51 OttoC


  • SoSH Member


  • 7354 posts

Posted 10 December 2010 - 08:14 AM

Bill James says that things that are persistent are real...Drew Batting 5th has producewd persistently good numbers...if batting him 5th isn't in the cards [for the other spots he is not so good]...Trade him...

And his numbers while batting in the third spot are almost identical (.414/.551 vs. .414/.558) with nearly twice as many plate appearances, so bat him third.

#52 geoduck no quahog


  • SoSH Member


  • 5605 posts

Posted 10 December 2010 - 08:45 AM

Any lineup that has base cloggers like Gonzalez or Youkilis in front of Crawford is insane.

Ellsbury and/or Pedroia have to be ahead of Crawford.

It's bad enough having the catcher ahead of Ellsbury (product of losing a middle-of-the-order catcher), but don't create artheriosclorosisbasepathitis twice in front of two of the best base runners in baseball.

#53 redsoxstiff


  • hip-tossed Yogi in a bar fight


  • 6657 posts

Posted 10 December 2010 - 09:24 AM

And his numbers while batting in the third spot are almost identical (.414/.551 vs. .414/.558) with nearly twice as many plate appearances, so bat him third.


I could see this...decent base runner ...I don't think it will happen but this is my 14 cents worth...

#54 rglenmt

  • Pip
  • 526 posts

Posted 10 December 2010 - 10:00 AM

I like Ellsbury continuing to develop as the leadoff hitter as Crawford does not like to do so and should bat 3rd. Scutaro or Lowrie, not Saltlalamacchia or Varitek should bat eighth not 9th, as particularly Scutaro can do a good hitter back to back with Ellsbury and Jed can adapt. With Ellsbury, Pedroia and Crawford in front of Gonzalez and Youkilis in either order should give Adrian and Kevin an opportunity to drive in a lot of runs and Big Papi to clean up the players who may remain on base.

#55 Eric Van


  • Kid-tested, mother-approved


  • 10990 posts

Posted 10 December 2010 - 01:29 PM

Any lineup that has base cloggers like Gonzalez or Youkilis in front of Crawford is insane.

Youkilis is not a base clogger. In fact, he's consistently been one of the very best baserunners on the team.

According to BP, in 2008 he was second to Ellsbury in total Baserunning runs (excluding SB runs), in 2009 he was second to Nick Green, last year he was just a tad above average. Part of this is that he very seldom gets thrown out, but in his career he's gone 1st to 3rd on a single 37% of the time despite playing half his games in Fenway (and is 44% his last three years); the league average is 27%. Fenway also kills scoring from 2B on a single, and he's 61% career and 62% last three years versus 58% average. He's gone 1st to home on a double 43% career, 46% the last three years; average is 43%, again despite a huge park disadvantage.

#56 Eric Van


  • Kid-tested, mother-approved


  • 10990 posts

Posted 10 December 2010 - 01:56 PM

Youkilis is not a base clogger. In fact, he's consistently been one of the very best baserunners on the team.

He's been the best baserunner on the team the last three years in terms of percentage advances on hits. Better than Crawford. He's mediocre getting a lead from 2B but ferocious leaving 1B, 3rd on the team to Crawford and Cameron for scoring on a double and wiping the field for going to 3B on a single.

The "Ave" column weights the first three columns 53-32-15, which is how often (out of 100) these chances occur. "Cam-prev" is Cameron 2007-9, everyone else is last three years except for Ellsbury and Salty who are last four (= career) to boost the sample size.

Take That Base!
Name 1 to 3, 1B 2 to H, 1B 1 to H, 2B Ave
Cam-prev 46% 60% 58% 52%
Youkilis 44% 62% 46% 50%
Crawford 26% 76% 53% 46%
Cameron 32% 59% 60% 45%
Lowrie 26% 71% 40% 42%
Pedroia 23% 78% 31% 42%
Drew 25% 70% 37% 41%
MLB 27% 58% 43% 39%
Ellsbury 31% 52% 38% 39%
Salty 36% 52% 0% 36%
Scutaro 18% 60% 41% 35%
Gonzalez 16% 44% 42% 29%
Ortiz 9% 31% 13% 17%
Varitek 7% 23% 20% 14%

Edited by Eric Van, 10 December 2010 - 02:00 PM.


#57 Eric Van


  • Kid-tested, mother-approved


  • 10990 posts

Posted 10 December 2010 - 07:14 PM

Assuming that the only upcoming changes will be in the bullpen, how does Tito go about setting the lineup?

Ellsbury, CF, LH
Crawford, LF, LH
Youkilis, 3B, RH
Gonzalez, 1B, LH
Pedroia, 2B, RH
Ortiz, DH, LH
Drew, RF, LH
Salty/Tek C, S/S
Lowrie/Scutaro SS, S/RH

I personally don't think Crawford will be leading off. His disdain for the spot is well documented.

Three pages of suggestions and the first remains the best. I think 1-6 above are correct; you just go Lowrie-Drew-C at the end to break up the Ortiz / Drew pair.

The logic is pretty simple. Two LHB have to hit in a row. Which two? Ellsbury has the mildest split, then Gonzalez, but you can't bat them back-to-back in any sane lineup. But you can bat Ellsbury and Crawford back-to-back while keeping them in the slots they are comfortable in, which is important.

It makes no sense to put extra pressure on Crawford by hitting him 3rd, which forces Youkilis, the 2nd best hitter on the team, all the way down to 5. Youk and Gonzalez will hit 3-4 or 4-3 just like Manny and Papi did.

Tito is a firm believer in the disruptive power of a base-stealing threat. While it's probably better to split them up and spread that disruption around, putting them back-to-back and hence creating DS opportunities is probably next best. With Ellsbury-Crawford, there's a 60% chance that Youkilis is up with at least one of them on base.

What opposing managers will do with their LOOGYs is bring them in to face Ortiz-Lowrie-Drew-Saltalamacchia-Ellsbury-Crawford. But with lefty-killing pinch-hitters on the bench (Cameron, Varitek, McDonald or Martin) they can be chased out of the game before they ever get to Ellsbury. Tito doesn't like to pinch-hit but if that's his only garlic for the LOOGY vampires, he'll hold his nose and use it.

(Whereas if 3-4-5-6 are Crawford-Gonzalez-Youkilis-Ortiz, the first two will always see a LOOGY once a game.)

Crawford won't benefit from seeing more fastballs with Ellsbury on 1B the way Pedroia might, but he shouldn't be hurt by it. Pedroia has the slash line of a great teammate-RBI guy and doesn't have significant bases empty / men on splits, and should be an excellent fit in the 5 hole.

Next year if they replace Papi with a RH bat, they can swap Pedroia and Crawford, swap Youkilis and Gonzalez, install Kalish 7 or 8 and have a perfectly balanced lineup.
.

#58 SoxFanSince57


  • Carrie Nation


  • 10048 posts

Posted 10 December 2010 - 07:41 PM

Very interesting post EV.

Help me out with the rationale of putting a solid OBP-hitter like Drew so low in the lineup. I am not comfortable with JD so low in the order, but I see most are recommending that.

#59 Savin Hillbilly


  • SoSH Member


  • 11436 posts

Posted 10 December 2010 - 09:56 PM

There's a very interesting piece at FG on this very question that I'm surprised no one's referenced yet. They ran an optimization tool on our presumed most frequent lineup and got this:

Youkilis 3b
Gonzalez 1b
Drew rf
Ortiz dh
Pedroia 2b
Crawford lf
Lowrie ss
Saltalamacchia c
Ellsbury cf

So the tool agrees with SoxFanSince57 about Drew....however, they do point out that baserunning is not factored into the algorithm that produces this, only batting outcomes. They also point out that the projected difference between this lineup and a more conventional one posited by Buster Olney is about one win per year.

I have to admit I would love for some team (wouldn't have to be the Sox, just anybody) to break out of the "fast guy at the top, HR hitters in the middle" mold and try something like this.

#60 mfried

  • 991 posts

Posted 10 December 2010 - 10:35 PM

There's a very interesting piece at FG on this very question that I'm surprised no one's referenced yet. They ran an optimization tool on our presumed most frequent lineup and got this:

Youkilis 3b
Gonzalez 1b
Drew rf
Ortiz dh
Pedroia 2b
Crawford lf
Lowrie ss
Saltalamacchia c
Ellsbury cf

So the tool agrees with SoxFanSince57 about Drew....however, they do point out that baserunning is not factored into the algorithm that produces this, only batting outcomes. They also point out that the projected difference between this lineup and a more conventional one posited by Buster Olney is about one win per year.

I have to admit I would love for some team (wouldn't have to be the Sox, just anybody) to break out of the "fast guy at the top, HR hitters in the middle" mold and try something like this.


The conventional wisdom would seem to have the best OBP leading off or at least high in the order, With this logic one might conceive of the lineup in two shifts of high-low OBP. However, this would only work if the low-OBP guys were high-RBI guys.

#61 OttoC


  • SoSH Member


  • 7354 posts

Posted 10 December 2010 - 11:10 PM

Some time back, Mark Pankin (SABR, Retrosheet) did Markov processing and regression analyze in order to analyze lineup construction quantitatively. His conclusions were:

The model equations, which are not shown, can be interpreted to characterize the desirable abilities for each batting order position:

1) Getting on base is everything. To much lesser extent, home run hitters should not lead off. Stolen base ability is irrelevant.

2) Similar to the leadoff hitter, but not quite as crucial to get on base; some power is also desirable.

3) Should have fair power, be able to draw walks, and not strike out much.

4) Highest slugging average; also has a good on base percentage and is not necessarily the best home run hitter.

5) Good power; secondarily puts ball in play (i.e. does not walk or strike out a lot).

6) Hardest spot to characterize and probably least critical. Probably want to use player who doesn't fit well in other positions. Base stealing ability is a small plus.

7-9) Decreasing overall abilities as hitters as characterized by on base percentage and measures of power hitting.

"One clear result from this and prior studies is the importance of having the right batters at the top of the order. This follows from the finding that most of the difference in expected runs between high and low scoring lineups using the same players occurs in the first inning. In particular, the leadoff batter must have a high on base percentage. Also, the second hitter must be good. The practice of leading off a fast runner who can steal bases, but doesn't get on base much, and putting a weak hitter with good bat control who can bunt or hit behind the runner second is a perfect prescription for a lower scoring batting order."

His paper is worth taking a look at if you are interested in the subject and haven't already seen it.

#62 Eric Van


  • Kid-tested, mother-approved


  • 10990 posts

Posted 10 December 2010 - 11:36 PM

Very interesting post EV.

Help me out with the rationale of putting a solid OBP-hitter like Drew so low in the lineup. I am not comfortable with JD so low in the order, but I see most are recommending that.

There's no rationale. He's your seventh best hitter so he's not in the top 6, and if you hit him 7th instead of 8th you negate what you've gained by altering the LHR / RHR proportion that he and Lowrie see. Some of his OBP is wasted but what's the alternative? Having a guy that good hitting 8th is a boon.

There's something else, too. When we got Drew, there was a rough backwards relationship (inverse correaltion) between the year he had and the strength of the lineup he was in. The weaker the lineup, the more he thrived. And he also had a track record of hitting better 5th than 2nd. What do these two things have in common? The degree to which he might be pitched around. In a weak lineup, the better hitters get pitched around more, and this is also true of a 5 hitter versus a 2 hitter: you can pitch around a 5 hitter because the 6, 7, 8 guys are coming up, but if you pitch around a 2 hitter the 3, 4, 5, guys are coming up -- you can see it's the same thing as the overall lineup being stronger or weaker. I have puzzled for a while why a guy would be better when they are more willing to walk him (not just more walks, but a higher SA), but it really seems to be true of Drew -- I have even noticed him getting hot when the guy up after him was slumping. It may be as simple as the pitch-calling patterns being more predictable in these pitch-around situations, whereas when they are forced to pitch to him they are also forced to be more creative in their approach.

So he might really thrive hitting that low, where they will take the attitude that they shouldn't let him do the damage when the catcher or SS is up next. That attitude on the part of opposing pitchers has seemed to help him.

If all this is true, it is illustrative of a key principle that been ignored by work as good as Pankin's and as questionable as the Baseball Musings calculator: do not assume that guys hit the same no matter where they are in the lineup, and, in fact, within the strictures found by Pankin (which are close to the conventional educated wisdom), try to get guys into spots where they will hit the best. For whatever reason Bill Mueller was insanely good hitting 8th, with splits hugely unlikely to be random*, and you lost more than you gained if you moved him to a more theoretically useful spot.

*Sometimes these differences are caused by the hitter's own sense of what is required of him in a given lineup spot. Until mid-2009 Ellsbury struggled whenever he thought it was his job to get on base; when he just tried to hit the ball with authority he was much better (you could see this in any split where he might opt to emphasize one over the other). Folks have noticed his defensive slap-hitting approach as distinct from his look-for-your-pitch, hit-it-hard one. Evidence suggests he has been learning to use the latter much more appropriately since they demoted him temporarily in mid-'09, and it's one of the reasons I think he can blossom hugely as a hitter.

#63 Eric Van


  • Kid-tested, mother-approved


  • 10990 posts

Posted 10 December 2010 - 11:41 PM

There's a very interesting piece at FG on this very question that I'm surprised no one's referenced yet. They ran an optimization tool on our presumed most frequent lineup and got this:
They also point out that the projected difference between this lineup and a more conventional one posited by Buster Olney is about one win per year.

And you stand a very good chance of giving back that win, and maybe more, if guys were uncomfortable in their roles or started altering their approach because of being in an unfamiliar spot.

Again, within the better-to-worst, OBP before SA general principles, find the spots where people will thrive.

#64 ponch73

  • 111 posts

Posted 11 December 2010 - 12:17 AM

Three pages of suggestions and the first remains the best. I think 1-6 above are correct; you just go Lowrie-Drew-C at the end to break up the Ortiz / Drew pair.

Why is it fair to characterize a lineup where a guy with a lifetime .330 OBP as a lead-off hitter bats first as the best? He also happens to be a guy coming off significant injuries and who may or may not have lingering pain.

I can understand that batting Ellsbury leadoff may in fact be the most likely choice (given Tito's proclivity to start him there), but I'm having trouble understanding why this is an optimal decision.

1. The numbers (OBP) don't support it.
2. Ellsbury himself doesn't appear that comfortable in the role based on his past approach and splits.
3. It might make sense to ease him back in a less conspicuous batting slot.
4. There appear to be other, higher OBP options for the Red Sox to consider (Lowrie or Scutaro, for example).

#65 Eric Van


  • Kid-tested, mother-approved


  • 10990 posts

Posted 11 December 2010 - 02:57 AM

Why is it fair to characterize a lineup where a guy with a lifetime .330 OBP as a lead-off hitter bats first as the best? He also happens to be a guy coming off significant injuries and who may or may not have lingering pain.

I can understand that batting Ellsbury leadoff may in fact be the most likely choice (given Tito's proclivity to start him there), but I'm having trouble understanding why this is an optimal decision.

1. The numbers (OBP) don't support it.
2. Ellsbury himself doesn't appear that comfortable in the role based on his past approach and splits.
3. It might make sense to ease him back in a less conspicuous batting slot.
4. There appear to be other, higher OBP options for the Red Sox to consider (Lowrie or Scutaro, for example).

He has a .364 OBP in 499 PA since he finally got a clue how to hit, dating from when they dropped him down in the lineup at the end of May 2009. Given his lack of power and his baserunning, that makes him the best leadoff candidate if that's for real and he's healthy (the alternative, Pedroia, is a .374 OBP guy whose power would be wasted there). The presumption is it's real (and a lot of evidence backs that up) and we'll find out in ST if he's healthy.

#66 Savin Hillbilly


  • SoSH Member


  • 11436 posts

Posted 11 December 2010 - 09:20 AM

Why is it fair to characterize a lineup where a guy with a lifetime .330 OBP as a lead-off hitter bats first as the best?

<snip>

4. There appear to be other, higher OBP options for the Red Sox to consider (Lowrie or Scutaro, for example).

Huh?

Lowrie's lifetime OBP: .336
Scutaro's lifetime OBP: .336
Ellsbury's lifetime OBP: .344

Granted, the lifetime number is a bit misleading in Lowrie's case, but it's misleading in Ellsbury's case too, for the same reasons. It's not the least bit misleading in Scutaro's case.

#67 BucketOBalls


  • SoSH Member


  • 5644 posts

Posted 11 December 2010 - 12:34 PM

Granted, the lifetime number is a bit misleading in Lowrie's case, but it's misleading in Ellsbury's case too, for the same reasons. It's not the least bit misleading in Scutaro's case.


Lifetime probably is misleading for Scutaro. His is .348 over the past five years(he was bad early in his career). And it was .348 in the first half last year before he got hurt.

#68 JimBoSox9


  • will you be my friend?


  • 12465 posts

Posted 11 December 2010 - 12:54 PM

Crawford
Pedroia
Gonzalez
Youkilis
Ortiz
Lowrie
Drew
Salty
Ellsbury

I don't know if I really have an opinion on the optimal lineup (it turns out that my brain is more slavishly wedded to the L-R-L idea than I thought), but I'm 75% sure that the above is going to be what Tito goes with against RHP at least.

Logic says that if you're going to have to hit two lefties in a row, the two lefties you want to group together are the two with the least radical platoon splits, or sandwhich them with lefty mashers. Here's a quick summary of production against LHP by our lefties:

Player 2008 2009 2010 3-yr avg

Gonzalez .213/.287/.387 .234/.339/.431 .337/.424/.513 .258/.347/.440
Crawford .248/.293/.348 .269/.325/.378 .256/.312/.384 .259/.312/.372
Ellsbury .295/.347/.381 .318/.374/.411 .235/.235/.235 .305/.357/.391
Ortiz .221/.308/.433 .212/.298/.418 .222/.275/.324 .218/.291/.383
Drew .284/.426/.500 .272/.381/.482 .208/.302/.309 .246/.358/.409

So this shows us two players who are very vulnerable to reduced production against lefties (Ortiz and Crawford), one player who hits lefties as well as righties (Ellsbury), and two players who had very outlier 2010s (Gonzalez and Drew). I'm going to make the assumption that the 2010 from the 28-year-old Gonzalez has more of a chance of being repeated than the 35-year-old Drew has of bouncing back from his 2010.

And here are the righties, vs LHP:

Player 2008 2009 2010 3-yr avg
Youkilis .288/.403/.606 .309/.435/.518 .404/.513/.798 .328/.447/.620
Pedroia .313/.376/.528 .277/.366/.399 .236/.367/.333 .284/.370/.439
Lowrie .338/.409/.525 .211/.318/.368 .338/.420/.606 .324/.403/.541
Salty .158/.220/.224 .229/.282/.375 .143/.368/.286 .194/.265/.306

That's two players who mash lefties (Youks and Lowrie), one who has head-scratchingly been trending towards a reverse split (Pedroia), and one player who seems worse against lefties at the MLB level but only has 1 season worth of PA (Salty).

Looking this over gives me a couple of thoughts:

-Ortiz must be the most well-protected. He's going to be the hardest one of the LHH for Tito to pinch-hit for late. The best way to protect him is to hit him low and back him up with someone who kills lefties - hello Lowrie.

-Ellsbury should be one of the two lefties hitting in a row. He's had enough PA to believe he really is just about a split-neutral player, and he's also the easiest player to pinch hit for (which is kind of stupid given the splits, but that's perception for ya).

-Two of these three really need to happen: Gonzalez's 2010 to be real, Pedroia to figure out what's wrong versus lefties, or JD Drew to bounce back. I don't need a clean sweep, but 2 out of 3 would make me feel much better.

-Salty really is the last piece that's tough to fit in.

This is what I came up with based on the above.

Pedroia
Crawford
Gonzalez
Youkilis
Ortiz
Lowrie
Drew
Salty
Ellsbury

Don't really love it, but I don't think there is such a thing as an ideal solution to this question. I didn't think it was possible to be so hard to make a lineup I like with so many hitters I love.

Edited by JimBoSox9, 11 December 2010 - 12:55 PM.


#69 sachilles


  • Rudy-in-training


  • 630 posts

Posted 11 December 2010 - 01:49 PM

Center fielder Jacoby Ellsbury(notes), who missed most of the 2010 season with rib injuries, likely will lead off once he fully recovers.

“Our best team is when Jacoby’s hitting first,” Francona said.

yahoo AP link.
Taken from an article about the Crawford signing.

#70 ponch73

  • 111 posts

Posted 11 December 2010 - 02:07 PM

Huh?

Lowrie's lifetime OBP: .336
Scutaro's lifetime OBP: .336
Ellsbury's lifetime OBP: .344

Granted, the lifetime number is a bit misleading in Lowrie's case, but it's misleading in Ellsbury's case too, for the same reasons. It's not the least bit misleading in Scutaro's case.

Now try rerunning those lifetime OBP when batting in the leadoff spot.

Lowrie: .396 (96 PA batting 2nd since he's apparently never led off)
Scutaro: .357
Ellsbury: .330

#71 Buzzkill Pauley


  • SoSH Member


  • 5246 posts

Posted 11 December 2010 - 02:14 PM

Without picking up another player, I would structure the lineups this way (with 3-year splits):

vs. RHP:
1. LF - Crawford (.316/.367/.493)
2. 2B - Pedroia (.315/.373/.488)
3. 1B - Gonzalez (.301/.409/.572)
4. 3B - Youkilis (.302/.390/.541)
5. DH - Ortiz (.274/.383/.549)
6. RF - Drew (.279/.386/.525)
7. CF - Ellsbury (.278/.333/.396)
8. SS - Scutaro (.274/.357/.404)
9. C - Saltalamacchia (.263/.347/.398)
B - Cameron (OF), McDonald (OF), Lowrie (IF), Varitek ( C)

vs. LHP:
1. LF - Crawford (.259/.312/.372)
2. 2B - Pedroia (.284/.370/.439)
3. 1B - Gonzalez (.258/.347/.440)
4. 3B - Youkilis (.328/.447/.620)
5. RF - Cameron (.289/.413/.567)
6. SS - Lowrie (.324/.403/.541)
7. CF - Ellsbury (.305/.357/.391)
8. DH - Ortiz (.218/.291/.383)
9. C - Varitek (.252/.347/.491)
B - Drew (OF), McDonald (OF), Scutaro (IF), Saltalamacchia ( C)



My assumptions are:
  • as creatures of habit, baseball players are more comfortable hitting from a stable spot in the lineup,
  • Crawford is actually willing to bat anywhere in the Sox lineup -- including leadoff,
  • Tito wants to make his players as comfortable as possible so they can perform at their best ability,
  • he wasn't blowing smoke when he said that he planned to have Scutaro as his regular shortstop, and
  • if Tito can't platoon Ortiz without poisoning the clubhouse, then he needs to try pressuring the pitcher.
I also think that, in theory, stacking hitters is best done having the batter with the worse split hit first, because you can bring in a PH for the one batter the specialist must face.

#72 ponch73

  • 111 posts

Posted 11 December 2010 - 02:19 PM

He has a .364 OBP in 499 PA since he finally got a clue how to hit, dating from when they dropped him down in the lineup at the end of May 2009. Given his lack of power and his baserunning, that makes him the best leadoff candidate if that's for real and he's healthy (the alternative, Pedroia, is a .374 OBP guy whose power would be wasted there). The presumption is it's real (and a lot of evidence backs that up) and we'll find out in ST if he's healthy.

Can you elaborate more on this evidence?

Because the argument above seems to suggest that Ellsbury is a heckuva lot more comfortable batting lower in the order. And that is something that is borne out in the stats.

Ellsbury batting first: .330 OBP in 1,229 PA

Ellsbury batting anywhere else: .405 OBP in 281 PA

Frankly, when I hear Tito say that "our best lineup is when Jacoby is batting first," I am very, very surprised.

#73 koufax32


  • He'll cry if he wants to...


  • 3370 posts

Posted 11 December 2010 - 02:27 PM

yahoo AP link.
Taken from an article about the Crawford signing.


So we have a few apparent certainties. If healthy it appears Dreamboat will lead off. Additionally it wouldn't make much sense for Crawford to be late in the order. Also IIRC Tito has stated his preference to start Scutaro while using Lowrie as a supersub. One last thing is the probability of Youk and Gonzalez in the 3-4 spots in some order. Dropping one of them to five costs the team too many lost at bats from one of their top two hitters. That would mean a top 5 that is either

Ellsbury
Pedroia
Gonzalez
Youkilis
Crawford

or

Ellsburt
Crawford
Youkilis
Gonzalez
Pedroia

The first completely blows up the back half of the lineup. The second becomes more balanced.

#74 sachilles


  • Rudy-in-training


  • 630 posts

Posted 11 December 2010 - 02:37 PM

Crawford was quick to say that he would go wherever Terry Francona wanted him to. While Tito acknowledged Crawford's willingness, it doesn't seem like he has any plans to put him in the leadoff spot, saying that the Red Sox were at their best with Jacoby hitting leadoff. As far as Crawford was concerned, Francona said that he was a top-of-the-order guy, and since Tito expects to talk with both Crawford and Pedroia in the coming days, it seems like his spot at second might be Crawford's destination.

From Ben Buchanan link

#75 SumnerH


  • Malt Liquor Picker


  • 14660 posts

Posted 11 December 2010 - 03:01 PM

Now try rerunning those lifetime OBP when batting in the leadoff spot.

Lowrie: .396 (96 PA batting 2nd since he's apparently never led off)
Scutaro: .357
Ellsbury: .330


Do you have any evidence that this is more predictive than using the larger sample?

#76 geoduck no quahog


  • SoSH Member


  • 5605 posts

Posted 11 December 2010 - 03:34 PM

OK, aside from theory and optimal lineups. Here's the actual lineup and why it works.

Ellsbury
Pedroia
Crawford
Gonzalez
Youkilis
Ortiz
Salty
Drew
Scutaro

First of all, accept the fact that the Red Sox are going to destroy pitching staffs. All the hitters will benefit. Only injury will prevent this lineup from having better-than-average years.

When Ellsbury get on, he's going to be at 2nd base, either by a double, a steal or a Pedroia at bat.

If first base is open, Crawford will be walked. That will put two speedsters on the basepaths.

Teams will want to shift against Gonzalez, but how can they with one or both of Ellsbury/Crawford on base? Advantage Gonzalez.

After Gonzalez does his thing, Youkilis comes up with at least someone in scoring position.

Now the shift re-appears for Ortiz, but what are the slim odds that Ortiz comes to the plate without no one on base? Particularly Youkilis - who may have been pitched around to get to Ortiz.

Salty will probably be the spot for pitcher-relaxation. That may prove to be a huge mistake. It's also is an ideal pinch-hitting spot with Varitek ready to close the game out behind the plate.

Now Drew/Scutaro act like a 1-2 in any other lineup, except they're 8-9 in the Red Sox lineup...and they put adequate speed on the basepaths in front of the turning over lineup.

There aren't enough quality left-handed pitchers in the universe that can survive a series against this team. Opponents will just have to pick their spots.

#77 ookami7m

  • 4194 posts

Posted 11 December 2010 - 03:58 PM

I agree with the lineup and some of the reasoning by why would anyone walk Crawford to get to gonzo or
R youks?

#78 sachilles


  • Rudy-in-training


  • 630 posts

Posted 11 December 2010 - 04:00 PM

If first base is open, Crawford will be walked. That will put two speedsters on the basepaths.

Teams will want to shift against Gonzalez, but how can they with one or both of Ellsbury/Crawford on base? Advantage Gonzalez.


Why would you walk Crawford to get to Gonzo?

#79 Al Zarilla


  • SoSH Member


  • 21107 posts

Posted 11 December 2010 - 04:01 PM

OK, aside from theory and optimal lineups. Here's the actual lineup and why it works.

Ellsbury
Pedroia
Crawford
Gonzalez
Youkilis
Ortiz
Salty
Drew
Scutaro

First of all, accept the fact that the Red Sox are going to destroy pitching staffs. All the hitters will benefit. Only injury will prevent this lineup from having better-than-average years.

When Ellsbury get on, he's going to be at 2nd base, either by a double, a steal or a Pedroia at bat.

If first base is open, Crawford will be walked. That will put two speedsters on the basepaths.

Why would you walk Crawford to get to Gonzalez? Holy crap, just typing their names and thinking of them playing for the Red Sox is a rush!

Edited by Al Zarilla, 11 December 2010 - 04:06 PM.


#80 OttoC


  • SoSH Member


  • 7354 posts

Posted 11 December 2010 - 04:06 PM

...All the hitters will benefit. Only injury will prevent this lineup from having better-than-average years...

The law of averages might suggest otherwise.

#81 Savin Hillbilly


  • SoSH Member


  • 11436 posts

Posted 11 December 2010 - 04:16 PM

Lifetime probably is misleading for Scutaro. His is .348 over the past five years(he was bad early in his career). And it was .348 in the first half last year before he got hurt.

His OBPs for the past five years are .350, .332, .341, .379, and .333. Can you say "outlier"?

He's not an outright hacker or anything, but there will usually be at least 7 guys in our lineup next year who are better at getting on base than he is.

#82 BucketOBalls


  • SoSH Member


  • 5644 posts

Posted 11 December 2010 - 04:34 PM

His OBPs for the past five years are .350, .332, .341, .379, and .333. Can you say "outlier"?

He's not an outright hacker or anything, but there will usually be at least 7 guys in our lineup next year who are better at getting on base than he is.


How can a 5 year average be an outlier?

As for the 7 guys...none of them are used to batting leadoff and none of them are Ellsbury. I wouldn't be adverse to trying Lowrie there, but he's never done it.

#83 TheGoldenGreek33

  • 1934 posts

Posted 11 December 2010 - 04:46 PM

Have not seen this mentioned, yet. Carson Cistulli over at FanGraphs ran Bill James' 2011 projections into Baseball Musings' "line-up optimizer tool" for the Sox. It ends up looking like this:

1. Kevin Youkilis
2. Adrian Gonzalez
3. J.D. Drew
4. David Ortiz
5. Dustin Pedroia
6. Carl Crawford
7. Jed Lowrie
8. Jarrod Saltalamacchia
9. Jacoby Ellsbury

Edited by TheGoldenGreek33, 11 December 2010 - 04:46 PM.


#84 glennhoffmania


  • Miracle Whipper


  • 8383584 posts

Posted 11 December 2010 - 06:39 PM

yahoo AP link.
Taken from an article about the Crawford signing.


Yup.

Manager Terry Francona said Jacoby Ellsbury will be their leadoff hitter when he is deemed fully healthy, and Crawford didn’t seem to mind. “Whatever he [Francona] wants to do with me is fine,” he said.


Link

Not sure what the point of making that statement today is. I'm still confused, as I was last year, about why a high OBP guy like Drew can't seem to crack the top of the lineup.

#85 Savin Hillbilly


  • SoSH Member


  • 11436 posts

Posted 11 December 2010 - 06:40 PM

How can a 5 year average be an outlier?

A 5-year average can be goosed by an outlier. Pull out the .379 year and the average for the remaining 4 years is .336.

Maybe the easiest way to make the point is with a graph. He's a league average OBP guy.

As for the 7 guys...none of them are used to batting leadoff and none of them are Ellsbury. I wouldn't be adverse to trying Lowrie there, but he's never done it.

Actually one of them is Ellsbury. The guy I was assuming might have a worse OBP than Scutaro is whoever's playing catcher.

#86 redsox2020

  • 252 posts

Posted 11 December 2010 - 10:34 PM

Have not seen this mentioned, yet. Carson Cistulli over at FanGraphs ran Bill James' 2011 projections into Baseball Musings' "line-up optimizer tool" for the Sox. It ends up looking like this:

1. Kevin Youkilis
2. Adrian Gonzalez
3. J.D. Drew
4. David Ortiz
5. Dustin Pedroia
6. Carl Crawford
7. Jed Lowrie
8. Jarrod Saltalamacchia
9. Jacoby Ellsbury

It's been mentioned in this thread a few times already, but all it appears to do is stack your studs at the top of the lineup. I remember either this tool or one similar too it used to always say Manny Ramirez was the team's ideal leadoff hitter.

#87 radsoxfan


  • SoSH Member


  • 7684 posts

Posted 11 December 2010 - 10:54 PM

It's been mentioned in this thread a few times already, but all it appears to do is stack your studs at the top of the lineup. I remember either this tool or one similar too it used to always say Manny Ramirez was the team's ideal leadoff hitter.


Clearly it values OBP far above anything else. I do think it's worth questioning Ellsbury in the leadoff hole, especially to start the season until we see how he plays after virtually an entire season off.

In an ideal world I'd prefer:

vs RHP

Crawford
Pedroia
Gonzalez
Youkilis
Ortiz
Drew
Salty
Lowrie/Scutaro
Ellsbury

vs. LHP

Lowrie
Pedroia
Gonzalez
Youkilis
Cameron
Varitek
Crawford
Mcdonald
Ellsbury

Edited by radsoxfan, 11 December 2010 - 10:59 PM.


#88 Eric Van


  • Kid-tested, mother-approved


  • 10990 posts

Posted 12 December 2010 - 01:18 AM

Can you elaborate more on this evidence?

Because the argument above seems to suggest that Ellsbury is a heckuva lot more comfortable batting lower in the order. And that is something that is borne out in the stats.

Ellsbury batting first: .330 OBP in 1,229 PA

Ellsbury batting anywhere else: .405 OBP in 281 PA

Frankly, when I hear Tito say that "our best lineup is when Jacoby is batting first," I am very, very surprised.

It's not that he's more comfortable lower in the order, it's that he had a better approach. Since his last demotion, he is now using that approach at leadoff, with much better results.

This was already discussed at great length before it happened. But I think it's worth giving a clear summary, including some info I haven't talked about yet..

-- All hitters vary their approach depending on count. Behind 0-2, the pitcher can throw anything in his arsenal, so you can't guess, so you must defensively cover the whole zone. Ahead 2-0, you can guess, hammering your pitch of you get it, taking a strike if you guess wrong. That hitters do a lot of such guessing is indicated by the surprising percentage of pitches right down the middle that are taken.

-- Ellsbury was one of the best 0-2 hitters in baseball and one of the worst 2-0. And what's more, watching him you could not just see a defensive vs. an aggressive approach, you could actually see two different swings, the opposite-field slap and, once in a blue moon, a sweet swing that would shockingly send balls over the bullpen.

-- The terrible 2-0 numbers suggested that far too often he was not being aggressive when ahead in the count, i.e., he was being defensive instead. Sure enough, pitch/fx data showed that he led the club in how often he swung at pitches down the middle but was last in how hard he hit them. The only reason you take a pitch down the middle is if you are guessing something else, and Ellsbury was doing less of that than anybody.

-- His splits suggested that he was switching between defensive and aggressive depending on game situation in a way that seemed completely logical, but was actually counter-productive because the defensive approach yielded much worse results. The huge top of the order / bottom of the order split was not that -- it was actually a split by the quality of opposing hitter. His numbers hitting leadoff with a weak hitter (basically, someone other than Pedroia) behind him were great, resembling his numbers at the bottom of the order. Now, what would the difference be there? With a great hitter up next, he would think it was his job to get on base. With a weak hitter up next, he might be more likely to want to drive the ball and be aggressive.

-- Furthermore, he was much worse with 0 outs (a get-on-base situation) than with 2 (much more a drive-the-ball situation). He was much worse with the bases empty than with RISP (same difference). He was much worse with the Sox tied or trailing than with them ahead (ditto). In fact, you could put all four of these splits together and come up with a logical formula for when he would look for his pitch when ahead in the count and unleash the hard, aggressive swing, depending on which of the four situational criteria (following hitter, score, men on base, outs) were met.

-- The Sox knew all of this and I have reason to believe that when they dropped him in the order at the end of May 2009, it was to work on using the aggressive approach the way other hitters do, i.e., whenever he was ahead in the count regardless of game situation. I have not re-run his splits or pitch/fx data since they returned him to the lineup, but I have noticed all the aforementioned splits getting smaller, have noticed him sitting on pitches and hitting the ball hard in what used to be his defensive game situations, and of course the overall results are consistent with a hitter who has a much better clue as to how to work his way through a PA.

-- Finally, the reason why his numbers hitting low in the lineup are so outrageously great (too good to be explained completely by this rationale) appears to be the lack of respect given to him by opposing pitchers. That was the situation that had the most influence in how often he was aggressive. They were still pitching him like he was a slap hitter and he was no longer just trying to set the table, he was looking for his pitch when ahead in the count much more often than any scouting report said he would. I would expect that the opposing book on him now acknowledges that he is capable of doing some damage and can't be thrown fat pitches when he's ahead in the count.

It appears as if he got all the way to MLB doing far less guessing and far more defensive plate coverage than most hitters, because he was so gifted at the latter that the results were always good enough. It's quite possible that he still has more to learn about working the count and looking for his pitch than most hitters his age, which is why I think he still has significant upside left.

Edited by Eric Van, 12 December 2010 - 01:24 AM.


#89 HriniakPosterChild

  • 3501 posts

Posted 12 December 2010 - 01:23 AM

After Gonzalez does his thing, Youkilis comes up with at least someone in scoring position.


Aren't the bases traditionally empty after a home run?

#90 reggiecleveland


  • sublime


  • 13654 posts

Posted 12 December 2010 - 01:38 AM

It's been mentioned in this thread a few times already, but all it appears to do is stack your studs at the top of the lineup. I remember either this tool or one similar too it used to always say Manny Ramirez was the team's ideal leadoff hitter.


But it does point out the folly of the worst hitter, or 2nd worst (Ellsbury)getting the most plate appearances.

#91 Savin Hillbilly


  • SoSH Member


  • 11436 posts

Posted 12 December 2010 - 10:10 AM

To expand on EV's last post, here's a breakdown of Ellsbury's slash numbers before, during, and after the 2009 batting order shift:

<5/31 (leadoff, 224 PA): .299/.335/.369
<5/31-7/19 (mostly bottom third, 147 PA): .286/.361/.437
>7/19 (leadoff again, 326 PA): .306/.363/.433

So the offensive improvement prompted by his move to the lower third of the order carried forward after he was moved back to the leadoff spot. It's not that he is, in some inherent and permanent sense, worse at hitting leadoff than at other spots in the order; it's that hitting lower in the order for a while, for whatever reason, enabled him to fix things about his offensive approach that had prevented him from being as effective as he could be in the leadoff spot.

#92 PrometheusWakefield


  • SoSH Member


  • 6590 posts

Posted 12 December 2010 - 06:17 PM

It's been mentioned in this thread a few times already, but all it appears to do is stack your studs at the top of the lineup. I remember either this tool or one similar too it used to always say Manny Ramirez was the team's ideal leadoff hitter.

I'd like to see a team take this advice. People exaggerate the importance of lining hitters up under a theory of putting guys who get on base ahead of guys who drive runners in, and underestimate the importance of giving your best hitter the most at bats. It's probably too radical to ever be tried, but it should be.

#93 glennhoffmania


  • Miracle Whipper


  • 8383584 posts

Posted 12 December 2010 - 06:23 PM

I'd like to see a team take this advice. People exaggerate the importance of lining hitters up under a theory of putting guys who get on base ahead of guys who drive runners in, and underestimate the importance of giving your best hitter the most at bats. It's probably too radical to ever be tried, but it should be.


But at a minimum, you don't drop one of your best hitters (Pedroia) down in the order to keep one of your worst hitters (Ellsbury) at the top. If Tito drops Pedroia to sixth and puts Ellsbury and Crawford 1-2, I'll be pretty annoyed. Not to mention it would mean that 4 of the first 5 hitters are lefties.

Edited by glennhoffmania, 12 December 2010 - 06:24 PM.


#94 Savin Hillbilly


  • SoSH Member


  • 11436 posts

Posted 12 December 2010 - 07:19 PM

But at a minimum, you don't drop one of your best hitters (Pedroia) down in the order to keep one of your worst hitters (Ellsbury) at the top.

This is overstating a point, I think. The last time Pedroia and Ellsbury both played a full season, the former was good for 18.3 wRAA, the latter 14.3. If they're both equally comfortable in the leadoff role, I agree I'd put Pedroia there over Ellsbury, but you don't really lose that much by putting Ellsbury there instead.

#95 Adirondack jack

  • 1317 posts

Posted 12 December 2010 - 07:24 PM

I'd like to see a team take this advice. People exaggerate the importance of lining hitters up under a theory of putting guys who get on base ahead of guys who drive runners in, and underestimate the importance of giving your best hitter the most at bats. It's probably too radical to ever be tried, but it should be.


It's what many little league teams do, for whatever that's worth. It would be interesting to see at the big league level, but we shouldn't hold our breath.

#96 Harry Hooper


  • SoSH Member


  • 14265 posts

Posted 12 December 2010 - 08:13 PM

Back in the 70's, Herzog had bat Brett and McRae 1-2 for the Royals for a spell.

#97 ponch73

  • 111 posts

Posted 12 December 2010 - 08:42 PM

It's not that he's more comfortable lower in the order, it's that he had a better approach. Since his last demotion, he is now using that approach at leadoff, with much better results.

This was already discussed at great length before it happened. But I think it's worth giving a clear summary, including some info I haven't talked about yet..

This was an incredibly-insightful post ... thank you.

In the event that Ellsbury needs a stint lower in the order to reacquire his aggressiveness, who would be your second choice to bat leadoff? And is the primary drawback to Lowrie hitting first the fact that his slugging would be wasted there (a la Pedroia)?

Edited by ponch73, 12 December 2010 - 08:44 PM.


#98 Buzzkill Pauley


  • SoSH Member


  • 5246 posts

Posted 12 December 2010 - 09:01 PM

This was an incredibly-insightful post ... thank you.

In the event that Ellsbury needs a stint lower in the order to reacquire his aggressiveness, who would be your second choice to bat leadoff? And is the primary drawback to Lowrie hitting first the fact that his slugging would be wasted there (a la Pedroia)?


I would think Pedroia would be the first choice for leadoff, if Ellsbury isn't ready because of rust or ribs. A leadoff double is always good.

Frankly, the primary drawback of Lowrie is probably that Tito thinks of him as a sub at this point, with Scutaro the starter.

#99 Eric Van


  • Kid-tested, mother-approved


  • 10990 posts

Posted 12 December 2010 - 10:51 PM

I would think Pedroia would be the first choice for leadoff, if Ellsbury isn't ready because of rust or ribs. A leadoff double is always good.

Frankly, the primary drawback of Lowrie is probably that Tito thinks of him as a sub at this point, with Scutaro the starter.

Pedroia is definitely the best option. His career OBP leading off an inning is .002 higher than his career numbers; the average AL hitter last year was .005 worse. His SA is +.016 versus the AL 2010 average of +.002. IOW, having him lead off, once he can get comfortable with the idea, would add value to him rather than subtract it. (And this is how I think you should look at every lineup. First look at how batting each guy in the proposed spot affects that guy, then look at how it plays as a sequence. It has been shown the the differences between a decent sequence and the best sequence are small but by no means inconsequential; I think the differences you can get from guys hitting in less than optimum spots are larger than the sequencing differences.)

So far in his career he has had little success there because he was probably thinking too much about his role changing. I think that in short order (no pun intended, honest!) he would forget about that crap and go back to the laser show.

Of course the problem with a Pedroia-first lineup is that then you're Crawford, Youkilis, Gonzalez, Ortiz, or Crawford, Gonzalez, Youkilis, Ortiz, and you are rather more susceptible to LOOGYs then when the back-to-back LHB are Ellsbury and Crawford. Either of those lineups will cause Papi to see a lot more LHR than the lineup where Youkilis-Gonzalez-Pedroia bat 3 to 5.

(Edit: Crawford is also +.002 OBP leading off an inning, but he's -.022 SA. It's 1261 PA, so, yeah, at this point in a career the differences are probably meaningful and predictive. So batting him leadoff loses you value.)

I'm as big a Lowrie fan as there is but let's wait a while before we decide he's good enough to be the guy getting that many PA. There's no special upside to gambling on that as they have with Ellsbury, who they believe disrupts the opposing pitcher when he does get on and thus has more value in that role than you could measure with his own stats. (The one time I looked at Pedrroia he did seem to benefit from having Jacoby on base.)

Edited by Eric Van, 12 December 2010 - 10:58 PM.


#100 Al Zarilla


  • SoSH Member


  • 21107 posts

Posted 12 December 2010 - 10:57 PM

I would think Pedroia would be the first choice for leadoff, if Ellsbury isn't ready because of rust or ribs. A leadoff double is always good.

Frankly, the primary drawback of Lowrie is probably that Tito thinks of him as a sub at this point, with Scutaro the starter.

How is Pedroia's foot? Dumb question: when they put a screw in it, as I believe they did, do they take it out later?