The 2018 Lineup

Buzzkill Pauley

Member
SoSH Member
Jun 30, 2006
10,569
If that’s the standard run-expectancy chart, I don’t think that bolsters your case very much.

Statistically speaking, even if the #3 sees ~20% of plate appearances with none-on-two-outs, you would still have to account for the other ~80% of plate appearances seen by the #3, then compare that to the results for other lineup spots, before making such a general conclusion about whether the the increased amount of PA’s with none-on-two-outs outweighs the increased amount of PA’s overall.

And also, I think you may be misunderstanding what a low-leverage situation is, as opposed to a low-expectancy situation. Although, any 1st inning situation will be relatively low-leverage.
 

rotundlio

Member
SoSH Member
Jul 8, 2014
323
But then the #4 hitter will come to the plate most often with nobody out and nobody on, therefore also limiting their run-producing capabilities, while also providing less total PAs.
Let's go back to the run expectancy chart.

Nobody on, two outs: 0.11 expected runs
Batter singles: 0.23 expected runs
Batter doubles: 0.33 expected runs
Batter homers: 1.11 expected runs

Nobody on, nobody out: 0.52 expected runs
Batter singles: 0.89 expected runs
Batter doubles: 1.11 expected runs
Batter homers: 1.52 expected runs

Nobody on, nobody out is much higher-leverage. And you can see here why you might place an emphasis on home run power at #3.
 

soxeast

New Member
Aug 12, 2017
206
I believe this would be by The Book.



Yes! But there is an important caveat. The #3 hitter is uniquely positioned so as to accumulate bases empty, two-out plate appearances. Approximately 40% of the time he will come to the plate in the first inning — the only scripted portion of the ballgame — with nobody on and two out. There is nothing to be done about it. It is in the nature of being a #3 hitter.

I made a table. It contains every player to have started at least 130 games at #2 through #5, as well as the percentage of their plate appearances that came with nobody on base and two batters out.


#3 hitters are nearly twice as likely to come to the plate with nobody aboard and two out! When the difference between #3 and #4 is 15–20 plate appearances, that is quite enough reason to move them down. And they'd be gaining 15–20 higher-leverage plate appearances by moving them up.



Because it doesn't abide by "traditional tenets of lineup construction?" As far as I can tell, the traditional tenet of lineup construction is that the #3 hitter is guaranteed to bat in the first inning, and ideally, there will be runners on base. But we see that this tenet is fallacious.
1-- In the 1st inning isn't it often the best time to get an elite starter? How many times have we heard the comment /or seen the starter settle down?

2-- Isn't it better to have the number 3 hitter on base in the 1st inning for the number 4 hitter? I have not argued the number 3 hitter to be the best or 2nd best. I have argued that Hanley and XB just aren't good enough. You also want a number 3 hitter on base for the number 4 hitter, right?

3-- As I've said before- imo XB and Hanley are a tier lower than Betts, Beni, and JD. One (HanRam) doesn't get on base enough or when he is on base he is one of the worst baserunners we have ever seen. ANd the other (XB) just doesn't hit with a combo of enough power or get on base enough.

4--- The 3rd hitter is not just an RBI hitter. You also want him to get on base as much as possible for the 4 hitter. Houston had Altuve on base over 40% of the time. Turner was on base over 40% of the time (Bellinger wasn't but he hit 39 hr), and Rizzo was on base .385% of the time and a bit over 39% of the time over the last 2 years). That's higher than Hnale projections. As for XB not enough power for one thing that's been mentioned before.
 

nvalvo

Member
SoSH Member
Jul 16, 2005
21,673
Rogers Park
That expectancy chart rotundlio posted is key. Let's say you have Joey Votto on your team. Great hitter!

The Reds hit him third 162 games last season. If they were to move him down a spot, he would get about 20 fewer PA a year, which would be bad. But the value of Votto's batting line, expressed in Win Probability Added, would increase, because the mix of game situations in which he would come to the plate would be somewhat more valuable. So the question is whether the benefit is worth the cost.

The authors of The Book found in their simulations that the answer is yes. The cost in plate appearances moving a hitter like Votto from 3 to 4 is less than the value of the uptick in expected runs that results, but only by a small margin.

That this was the result in a given simulation is no guarantee. But it's something.
 
Last edited:

DanoooME

above replacement level
SoSH Member
Mar 16, 2008
19,883
Henderson, NV
Didn't Brett and McRae bat 1-2 for the Royals way back when?
I think Willie Wilson was their primary leadoff guy. He was more of an old school leadoff type .. world class speed but lousy OBP - which was very common in the sixties and seventies. Didn't the Red Sox used to bat Burleson leadoff?
The Royals did that in 1977, after Brett mostly hit 3rd and McRae 5th in 1976. One thing to remember is that Brett was not yet GEORGE BRETT. He was still only 24 in 1977. And they didn't have Willie Wilson until 1978, except for a cup of coffee in September.

One interesting thing is the Royals scored 743 runs in 1978 with Brett back in the 3 hole after 812 runs in 1977 and 713 runs in 1976 with mostly the same players (McRae played a lot more OF in 1976, Mayberry was replaced at 1B after his drug issues in 1978 by Clint Hurdle - yes, that one -, and Darrell Porter took over the majority of the starts at C in 1977). The rest of the group was pretty much the same (Frank White, Freddie Patek, Amos Otis, and Al Cowens were the others). Maybe Herzog was on to something since he didn't really have a traditional leadoff hitter in 1977, but went with Wilson in 1978 and beyond.
 

Adrian's Dome

Member
SoSH Member
Aug 6, 2010
4,424
Let's go back to the run expectancy chart.

Nobody on, two outs: 0.11 expected runs
Batter singles: 0.23 expected runs
Batter doubles: 0.33 expected runs
Batter homers: 1.11 expected runs

Nobody on, nobody out: 0.52 expected runs
Batter singles: 0.89 expected runs
Batter doubles: 1.11 expected runs
Batter homers: 1.52 expected runs

Nobody on, nobody out is much higher-leverage. And you can see here why you might place an emphasis on home run power at #3.
Does this account for batter skill?

If not, it's statistical hot garbage.
 

rotundlio

Member
SoSH Member
Jul 8, 2014
323
You're dismissing my point out of hand, so I think we're getting close.

Here is that same list of full-time #2, #3, #4, and #5 hitters sorted by their FanGraphs "leverage index" in 2017. I highlighted the #3 hitters in red. Here is that same data for 2016.

I can continue to conjure up statistical hot garbage for you, but the truth is that this is very much a moot point. There are so many variables to constructing the "optimal" lineup, from wind to weather to injuries to whether or not the opposing starting pitcher has been featuring the cutter more frequently with two strikes against right-handers in his past five starts, that any number of lineup configurations will be functionally identical on any given night. Cora has every intention of giving his players regular rest, as he'd damn well better, and his fifth-best option against Marcus Stroman will not be his fifth-best option against Max Scherzer.

You can go ahead and bat J.D. third. It's probably his decision to make.
 
Last edited:

soxeast

New Member
Aug 12, 2017
206
[QUOTE="rotundlio, post: 2693053, member: 71704"]You're dismissing my point out of hand, so I think we're getting close.

Here is that same list of full-time #2, #3, #4, and #5 hitters sorted by their FanGraphs "leverage index" in 2017. I highlighted the #3 hitters in red. Here is that same data for 2016.

I can continue to conjure up statistical hot garbage for you, but the truth is that this is very much a moot point. There are so many variables to constructing the "optimal" lineup, from wind to weather to injuries to whether or not the opposing starting pitcher has been featuring the cutter more frequently with two strikes against right-handers in his past five starts, that any number of lineup configurations will be functionally identical on any given night. Cora has every intention of giving his players regular rest, as he'd damn well better, and his fifth-best option against Marcus Stroman will not be his fifth-best option against Max Scherzer.

You can go ahead and bat J.D. third. It's probably his decision to make.[/QUOTE]

Huh?
 

rotundlio

Member
SoSH Member
Jul 8, 2014
323
Either we agree that the three-hole is less important than the two- and the four-, or we don't, but there's no use in continuing to beat it to death. I would happily trade those two runs for a peaceful resolution to this debate.
 

soxeast

New Member
Aug 12, 2017
206
Either we agree that the three-hole is less important than the two- and the four-, or we don't, but there's no use in continuing to beat it to death. I would happily trade those two runs for a peaceful resolution to this debate.
Sure.

Let's agree to disagree.
 

Reverend

for king and country
Lifetime Member
SoSH Member
Jan 20, 2007
64,448
Yes it does imo. The point I made was that the lineup doesn't have to be 100% by the book. Not every stat can be used optimally by each team. That's what I've said. Nothing here provided/ stated disproves that imo.

You are going to disregard what championship teams (Houston/Cubs) and the elite team of the Dodgers have done that are heavy users of advanced metrics? That goes against nearly every successful business strategy. A sound business strategy is that you try to emulate what the best teams/ best businesses do. It seems you think you've found a weakness in their business models - that they refuse to employ the strategy.

I say you haven't unless you can show why the best teams didn't employ "the 5th hitter batting 3rd" and then see how their strategy wouldn't work as well as yours. You're just completely closing your eyes to the championship teams strategy without knowing why they are doing what they're doing. IMO that's not a successful strategy. Until you find out why - then imo you are guessing. I think my point is further bolstered for example by the actions of the Houston manager going against the Book last year and Francona's comments that you can't always use the book. And what the Royals did a few years ago they went against the book on many issues.

So you're basically ignoring what the championship teams have done, ignoring what prior championship mangers have done (Hinch) or are saying (Francona) while recognizing that the book cannot be right all the time-- and you think that proves I'm wrong? Anyhow -- address these issues then I'm on board. Or we can just agree to disagree. No big deal.

I'll just reiterate -- I am not against a number 5 hitter - batting 3rd. You can go back and read my past posts in which I've said several times that imo Hanley and Xander are a tier below Betts, Beni and JD. I wouldn't have a problem if they were better. For example I mentioned if Hanley was like what he was in 2016, I wouldn't have a problem with him batting 3rd (but then he wouldn't be 5th best hitter, he'd be much higher.) . I do think it's wrong for the 2017 Hanley to be batting 3rd however - he must do better especially in a crucial top of the order in which he gets more at bats- than what he's shown.
I meant agreeing to disagree.

It was a form of joke. I thought it would be funny. Still do.,,
 

joe dokes

Member
SoSH Member
Jul 18, 2005
30,556
I'll just reiterate -- I am not against a number 5 hitter - batting 3rd. You can go back and read my past posts in which I've said several times that imo Hanley and Xander are a tier below Betts, Beni and JD. I wouldn't have a problem if they were better. For example I mentioned if Hanley was like what he was in 2016, I wouldn't have a problem with him batting 3rd (but then he wouldn't be 5th best hitter, he'd be much higher.) . I do think it's wrong for the 2017 Hanley to be batting 3rd however - he must do better especially in a crucial top of the order in which he gets more at bats- than what he's shown.
The bolded strikes me as the kind of circular logic that leads to "good pitching stops good hitting." Except when it doesn't. "well, then it wasn't good pitching."

How can you possibly know how 2016 Hanley would rank in a hypothetical 2018 lineup?
 

Saints Rest

Well-Known Member
Lifetime Member
SoSH Member
The Book seems to (I haven’t read it, only followed summaries here) have noted that their simulations can’t account for individual batter psychology (for lack of a better term). Some players will likely respond, for various and idiosyncratic reasons, to batting in certain spots.
Someone upthread mentioned Burleson batting leadoff (when the prevailing wisdom of the time clearly would have called for Remy, as the speed guy, to bat leadoff).
I also seem to recall that the Sox discovered that Ortiz as 3 and Manny as 4 worked better than the inverse, despite conventional wisdom and/or metrics saying the opposite.
 

BuellMiller

New Member
Mar 25, 2015
451
Of course each team is not the average team, so the best possible (assuming players are random-number generators, which they are not) batting order may not be obvious from the results presented in The Book. That said, The Book's results plus knowledge of the players is probably the best guide, short of...

why doesn't someone just grab some rate projections and code up a quick simulation using some basic guesses at probabilities for things like "scoring from 2nd on a single" and run it. 9! possible batting orders for a given line-up, times ~38 PA/game, times 162 games is only a couple trillion calculations, which is trivial. I actually did this over a decade ago, but it'd take me more time to find the code - if it even survives - than it would for someone to write it anew. (No, I'm not volunteering. Instead, I'm going to be "that guy".)
<a bit off topic>
Ha, I did this myself about 15 years ago, too (needed something to do while my "real" simulations were running for grad school). Ended up running about 1000 simulations per lineup, and compiling the average and stdev of each lineup (i think to also avoid the need of doing every single lineup, did a random sampling). For the Red Sox, the lineups weren't all that different (certainly most if not all were within statistical error). The interesting result was when I did it for my softball team, and the quirky rules of the summer league I was in (the big one being a 6 run/ inning cap), and players with a bit wider spread of skills (e.g. players with OPS of .700 at the low end, up to players with OPS at least twice that), that suggested interspersing your lineup with your lesser players, instead of putting them all at the bottom (this made intuitive sense, since you want to make sure you score 6 runs every inning you can, and not waste a good inning where you get 6 runs with 0 or 1 outs.)
(We did win the championship after implementing the new lineups, but correlation <> causation)
</a bit off topic>
 

nvalvo

Member
SoSH Member
Jul 16, 2005
21,673
Rogers Park
Not off topic at all, BuellMiller, and fascinating.

Two links about lineup construction that I offer in the hope that they will be interesting:
  • A great Travis Sawchik piece at Fangraphs on unorthodox lineup construction involving Billy Hamilton.
  • And our own SoxScout, Red Sox Stats on twitter, offers his preferred lineup, which is pretty different from what most of us have been saying:
 

Savin Hillbilly

loves the secret sauce
SoSH Member
Jul 10, 2007
18,783
The wrong side of the bridge....
Yes to that.

I do think it makes more sense to start with Benintendi, if only because it allows a neat L/R alternation with JDM in the cleanup spot where he belongs.

And I like JBJ in the 3 hole -- it would be a vote of confidence, and yet it's not really the overpromotion it looks like to the uninitiated (e.g. the "Bradley 3rd? Are you high?" comment on the tweet).
 

Hank Scorpio

Member
SoSH Member
Apr 1, 2013
6,991
Salem, NH
Bradley in the #3 spot is interesting. He rarely hits into double plays, and while he strikes out somewhat moderately, I still think his ceiling is pretty high if he can find some consistency. He's probably capable of 30 home runs, and putting up a .280ish average, and it would be interesting to see him hitting ahead of a slugger like JDM. He'd get a lot more drivable pitches.

I don't really get the point of swapping out Hanley for Moreland and calling it "optimal", even if it's to maintain a LRLR pattern. Bad Hanley is somewhat worse than Moreland, but Good Hanley is leaps and bounds better. And we all know there is no Good Moreland or Bad Moreland. There is only Mitch Moreland.
 

geoduck no quahog

not particularly consistent
Lifetime Member
SoSH Member
Nov 8, 2002
13,024
Seattle, WA
I believe (non-sabremetrically) that:

a. Slow people shouldn't be in front of fast people (everything being equal), since it impacts taking the extra base or stealing. I acknowledge you typically don't steal in front of certain hitters, regardless.

b. There's a real benefit in not stacking guys who hit the same-handedness pitching well (not necessarily straight L-R-L) so you force the hand of the opposing manager before pinch hitting.

Everything else - what you guys said - acknowledging that the lead-off hitter is kind of moot after the first inning.

This is what, to me, makes a sequence of Bradley, Betts and Benintendi logical (perhaps flipping Benintendi and Betts if that works better), with JDM #4 so #2 can steal with a hitter between him and the HR guy...game situation dictating that. Also - the hole in the infield thing is kind of important.

Are there lineup analyses that address the lineup positioning of guys based partially on speed?
 

Sampo Gida

Member
SoSH Member
Aug 7, 2010
5,044
My preferred lineup

Betts
Benintendi
JDM
Devers
Hanley
Boegarts
JBJ
Nunez
Vazquez

I'm not putting Boegarts or JBJ in a middle of the order spot. Need more power from XB and JBJ needs to be more consistent. If they can improve on that then I can reconsider
 

Dewey'sCannon

Member
SoSH Member
Jul 18, 2005
870
Maryland
Cora has said he like a power/speed guy batting leadoff to put pressure on the pitcher and defense right from the first pitch, like the Astros do with Springer, and that's why he wants Betts there. I agree with the concept, and am ok with Mookie in that slot and Beni second, although I agree with SH and GNQ that you could probably flip those two.

the biggest question, and debate in this thread, has been on the third slot. I think there's universal agreement that JDM is either #3 or #4, but if you prefer him at #4 you have to find someone suitable at #3.

- Devers is too young, and is less than ideal if Beni is #2;
- 2016 Hanley would be suitable, but we won't know for a while if that's who we get.

Bogey and JBJ are the interesting choices. JBJ has the bigger power upside, but again this probably only works if you flip Mookie and Beni.
I know that folks are down on Bogey for his lack of power. But two things: (1) I think a lot of the reason for the drop-off last year was due to the hand injury; and (2) Approach: he has acknowledged that he became too passive last year, and that he took way too many good pitches to hit, and that he's looking to be more aggressive earlier in the count. I move to the #3 or #4 hole just might help force him (or aid him) in being more aggressive. It might be a nice psychological boost (or kick in the rear).

On days when Vasquez or Leon are catching (i.e., most days), I would bat the C at #8, and leave #9 for someone faster - Nunez to start the season, or JBJ. Or Swihart would be good on days he catches. I like #9 as the "2nd leadoff" hitter (it takes me back to my role in HS).
 

Pozo the Clown

New Member
Sep 13, 2006
745
Bogey and JBJ are the interesting choices. JBJ has the bigger power upside...
I'm not sure that I necessarily agree. I have a gut instinct that X is going to really blossom power-wise this year (sorry, no supporting statistical analysis from my GI tract). In fact, I'm willing to make a wager (loser donates to the Jimmy Fund) that if both JBJ and X rack up at least 450 ABs and that they are within 50 ABs of each other at regular season's end that X will out-homer JBJ. What says you Dewey's Cannon?
 

Savin Hillbilly

loves the secret sauce
SoSH Member
Jul 10, 2007
18,783
The wrong side of the bridge....
So far in their careers JBJ has had a HR/FB more than 50% higher than Xander's, and a Hard% of 34.2% to Xander's 30.8%. Looking at the Statcast numbers, JBJ has consistently had a much higher Barrels/PA number than Xander's--in Xander's best year, 2016, JBJ led him 5.8 to 3.9.

By every measure I can find, JBJ has hit the ball harder from day one, and it's not close. For Xander to out-homer him, he'd have to break out pretty dramatically and/or change his approach pretty drastically.
 

MikeM

Member
SoSH Member
May 27, 2010
3,114
Florida
Even with a decent overall rebound JBJ is way too streaky to hit 3rd in this lineup imo. Those out machine stretches he has while hitting in front of JDM would be absolutely brutal.
 

SouthernBoSox

Member
SoSH Member
Jul 23, 2005
12,105
Pretty sure your opening day lineup is out there today....


I gotta say, I feel really good about it.
 

Red(s)HawksFan

Member
SoSH Member
Jan 23, 2009
20,881
Maine
I like it, too, although I expect opening day catcher won't be Leon. Didn't Vazquez catch his last outing?
He did. Cora has suggested that all pitchers were going to be working with all catchers. Unless we're a month or so into the season and Sale has pitched exclusively to Leon, I'm not sure we can say for certain that last year's "caddy" system is still in place.
 

Sampo Gida

Member
SoSH Member
Aug 7, 2010
5,044
So based on Coras comments I think we will be seeing this lineup quite a bit

Betts-RF
Benintendi-CF
Hanley-DH
JDM-LF
Boegarts-SS
Devers-3B
Nunez-2B
Moreland-1B
Vazquez-C

Obviously, JBJ can force the issue and grab playing time for Bennie in CF with a hot start and Bennie can lose time with a cold start. If both start off hot we could see JDM get more DH time, and less Hanley at DH (more at 1B depending on who is hitting)

Cora seems like he will adjust on the fly so I wouldn't bet on anything. I am satisfied for now that XB will be dropped down. Hopefully he can move himself up by being more aggressive. Stay tuned
 

Pozo the Clown

New Member
Sep 13, 2006
745
So based on Coras comments I think we will be seeing this lineup quite a bit

Betts-RF
Benintendi-CF
Hanley-DH
JDM-LF
Boegarts-SS
Devers-3B
Nunez-2B
Moreland-1B
Vazquez-C

Obviously, JBJ can force the issue and grab playing time for Bennie in CF with a hot start and Bennie can lose time with a cold start. If both start off hot we could see JDM get more DH time, and less Hanley at DH (more at 1B depending on who is hitting)

Cora seems like he will adjust on the fly so I wouldn't bet on anything. I am satisfied for now that XB will be dropped down. Hopefully he can move himself up by being more aggressive. Stay tuned
Do you have a link to said comments?
 

Red(s)HawksFan

Member
SoSH Member
Jan 23, 2009
20,881
Maine
Yeah, if everyone is healthy, that lineup *might* be seen 5-10 times this year. Depends on your point of view whether that constitutes "quite a bit". I'd be curious to see these alleged comments too.
 

RedOctober3829

Member
SoSH Member
Jul 19, 2005
55,452
deep inside Guido territory
So based on Coras comments I think we will be seeing this lineup quite a bit

Betts-RF
Benintendi-CF
Hanley-DH
JDM-LF
Boegarts-SS
Devers-3B
Nunez-2B
Moreland-1B
Vazquez-C

Obviously, JBJ can force the issue and grab playing time for Bennie in CF with a hot start and Bennie can lose time with a cold start. If both start off hot we could see JDM get more DH time, and less Hanley at DH (more at 1B depending on who is hitting)

Cora seems like he will adjust on the fly so I wouldn't bet on anything. I am satisfied for now that XB will be dropped down. Hopefully he can move himself up by being more aggressive. Stay tuned
Tell me again why JBJ has to "force" any issue in CF?
 

Sox Puppet

Member
SoSH Member
Dec 7, 2016
726
I know this isn't one of the options being discussed, but Nunez was very solid at the top of the lineup last year, including a .916 OPS in 174 PA as the #2 hitter. I wouldn't mind seeing a top 4 of Benintendi - Nunez - Betts - JD. Of course, batting Nunez #9 essentially puts him at the "top" of the order after the first round of ABs.

Nunez stats by lineup spot last year:
https://www.baseball-reference.com/players/split.fcgi?id=nunezed02&year=2017&t=b#lineu::none
 

Savin Hillbilly

loves the secret sauce
SoSH Member
Jul 10, 2007
18,783
The wrong side of the bridge....
I know this isn't one of the options being discussed, but Nunez was very solid at the top of the lineup last year, including a .916 OPS in 174 PA as the #2 hitter. I wouldn't mind seeing a top 4 of Benintendi - Nunez - Betts - JD. Of course, batting Nunez #9 essentially puts him at the "top" of the order after the first round of ABs.

Nunez stats by lineup spot last year:
https://www.baseball-reference.com/players/split.fcgi?id=nunezed02&year=2017&t=b#lineu::none
174 PA is a really small sample. For his career, Nunez' lines are almost indistinguishable for the four batting order positions in which he's had the most PA (1st, 2nd, 7th, and 8th).

He's a pretty good hitter, but still probably the 8th best hitter in the likely Opening Day lineup, which just tells you what a fun year we have to look forward to.
 

Pozo the Clown

New Member
Sep 13, 2006
745
I have a gut instinct that X is going to really blossom power-wise this year (sorry, no supporting statistical analysis from my GI tract)...
Keith Law feels the same way: "Bogaerts always projected to hit for 20-plus homers, showing easy power as a teenager in A-ball, and I fully expect him to bounce back to his 2016 power levels and probably even exceed them."

https://nesn.com/2018/03/why-espns-keith-law-believes-xander-bogaerts-could-break-out-in-2018/