Obligatory batting order post
March 29, 2011 at 12:42 am by Capitol Avenue Club under Atlanta Braves
Tool shed: Optimizing Your Lineup By The Book | Beyond the Box Score, 2011 ZiPS Proections – Atlanta Braves | Baseball Think Factory (also available on Fangraphs player pages).
Data: Below are three images. I’ve listed the Braves eight starting position players (Brian McCann, Freddie Freeman, Dan Uggla, Chipper Jones, Alex Gonzalez, Martin Prado, Nate McLouth, and Jason Heyward) along with their ZiPS projected weighted on base average, on base average, and slugging average. Each image contains the same information, but they’re sorted differently. The first image is sorted by wOBA, the second by OBA, the third by SLG. The column that the data is sorted by is indicated with a ‘«’. For the record, *=left-handed hitter and #=switch-hitter. No marking indicates a right-handed hitter.
Going through slot-by-slot. These quotes are pulled from the line-up optimization article at BtB.
Lead-off:
The Book says OBP is king. The lead-off hitter comes to bat only 36% of the time with a runner on base, versus 44% of the time for the next lowest spot in the lineup, so why waste homeruns? The lead-off hitter also comes to the plate the most times per game, so why give away outs? As for speed, stealing bases is most valuable in front of singles hitters, and since the top of the order is going to be full of power hitters, they’re not as important.
I’m putting Chipper Jones here. He’s the team’s second-best on-base threat according to ZiPS projection, and the best one you’re going to want to save for a role that involves both table-setting and run producing rather than leadoff. I want to re-emphasize something: “As for speed, stealing bases is most valuable in front of singles hitters, and since the top of the order is going to be full of power hitters, they’re not as important.” There may be a better lead-off hitter on the team than Chipper, I’m willing to listen, but if the justification for Chipper being unqualified for the role involves the fact that he’s not a great stolen base threat the debate is framed with a faulty premise and I will not talk to you about this topic.
The #2 spot:
The Books says the #2 hitter comes to bat in situations about as important as the #3 hitter, but more often. That means the #2 hitter should be better than the #3 guy, and one of the best three hitters overall. And since he bats with the bases empty more often than the hitters behind him, he should be a high-OBP player. Doesn’t sound like someone who should be sacrificing, does it?
This is where Jason Heyward hits, and it’s the most obvious decision of the eight if you’re familiar with The Book’s take on optimizing a batting order. You give him a ton of plate appearances and he comes to the plate in tons of crucial, run-producing situations. This is where your best hitter goes (provided he’s well-rounded), Atlanta’s best hitter is Jason Heyward (who is very well-rounded).
The #3 spot:
The Book says the #3 hitter comes to the plate with, on average, fewer runners on base than the #4 or #5 hitters. So why focus on putting a guy who can knock in runs in the #3 spot, when the two spots after him can benefit from it more? Surprisingly, because he comes to bat so often with two outs and no runners on base, the #3 hitter isn’t nearly as important as we think. This is a spot to fill after more important spots are taken care of.
—snip—
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. After positions #1, #2, and #4 are filled, put your next best hitter here, unless he lives and dies with the long ball.
I pulled Sky’s quote about the #5 spot because I wanted to point out that homers can be leveraged in the #3 spot. To this end I’ve chosen Dan Uggla to hit 3rd. He’s not an ideal fit for the batting order position, ideally you’d have your best power hitter fourth, but because the Braves don’t have a guy that truly lives-and-dies by the homer and to maintain a lefty/righty balance, I think he’s the best option to hit 3rd that Atlanta has.
The cleanup hitter:
The Book says the #4 hitter comes to bat in the most important situations out of all nine spots, but is equal in importance to the #2 hole once you consider the #2 guy receives more plate appearances. The cleanup hitter is the best hitter on the team with power.
I picked Brian McCann for this one–even though Uggla is probably a better fit–to maintain the lefty/righty balance. ZiPS projects the pair to finish with identical slugging averages, with Brian McCann’s superior double rate compensating for Uggla’s advantage in homer rate. The offense worked last year with McCann hitting cleanup, it should continue to work with him hitting cleanup in an optimized lineup.
The #5 spot:
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. After positions #1, #2, and #4 are filled, put your next best hitter here, unless he lives and dies with the long ball.
This quote is redundant, I posted it when discussing the #3 spot. Anyway, Martin Prado is my pick for this spot. Again, continuing with the lefty/righty balance theme and because Martin’s game revolves around hitting a lot of singles and doubles.
The #6 spot:
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.
This quote deals with the 6-9 hitters. Given that Nate McLouth is the best base-stealing threat on the team I’ve placed him 6th to optimize the value of those steals. Also, lefty/righty balance.
The #7 spot:
I’ve got Alex Gonzalez in the #7 spot for a few reasons. Reason number one: you guessed it, lefty/righty balance. Reason number two is the #8 hitter hits in front of the pitcher, which sometimes leads to a player getting pitched around. Therefore, it’s bad to put a hitter there who has no plate discipline at all.
The #8 spot:
Freddie Freeman hits 8th. I like him here for one reason, other than lefty/righty balance of course. The reason is the part of Freeman’s game that needs the most work right now is his approach, and I think that putting him eighth where he may be pitched around more than he would elsewhere gives him more opportunities to work on his patience. He’s already got the contact/power part of his offensive game down, with more patience he could really turn into a legit middle-of-the-order threat. Also, Fredi Gonzalez’s justification for doing so isn’t wrong, putting Freeman 8th where the pressure’s off and he can get his feet wet seems like a good idea to me.
And there’s your ZiPS projection aided, The Book optimized batting order:
Loading ...
So I’ve often wondered, if lineups all followed the recommendations by The Book, wouldn’t that increase the role of the Number 3 hitter? Reason being, as there is now higher OBP players in front of the hitter, wouldn’t he have more opportunities to hit with runners on base? This being because managers no longer select for speed as an issue, hence more opportunities, therefore the better hitter is needed there.
Just saying I think Chipper’s role last year was more important when Heyward was hitting in the 2-spot than when someone else was.
Just my 2 cents.
great images!
Someone please forward this post to Fredi Gonzalez before Thursday.
Jason Heyward is projected to lead the team in OBP, SLG, and wOBP. He’s going to be batting 6th.
ugh.
Awesome, awesome, awesome post. Thanks for the great analysis.
Well, my only argument against Chipper in the leadoff spot is I don’t think he’d like it at all and may silently protest the move, especially coming from Fredi as opposed to Bobby. I also don’t think he’d like running the bases that much. As little as it would matter, I’d be fine keeping Chipper in the #3 spot, just for the sake of keeping people happy.
Another thing is that I think for certain hitters, lineup position can change their approach, for better or worse. I think Martin Prado may be more inclined to be patient in the leadoff spot, which is his biggest fault as a hitter, so I’d hit him there.
I’d bat Uggla 4 and McCann 5
I’d bat Nate #6
If I can’t trade Gonzo, I’d bat him 7th, I guess. Though I’d be tempted to bat him 9th, but maybe I’m just spiteful.
Freddie 8th for the reasons you mention. Though I worry that the Atlanta coaches won’t stress the importance of patience and “clearing the pitcher”. The move could backfire if he gets frustrated with walking and starts hacking even more.
I was thinking the same thing as 1/Yaz…
If all this line-up optimatization is based on actual line-ups (which are not optimized), then how do we really know what is optimal?
One other thing I meant to mention is that I’ve always felt that The Book doesn’t adequately take into account the amount that NL leadoff hitters are going to come to bat with 1 or 2 outs and a man on 2nd, because of how much NL managers have the pitcher sacrifice, which means that I think the #1 spot should value singles and doubles a lot more than it does out of the leadoff spot and perhaps walks very slightly less.
Heyward should bat somewhere high in the order for the simple fact that by batting him sixth instead of second, the Braves are costing him at least a few games worth of plate appearances over the course of the season.
Even if one doesn’t quite grasp the concepts of building a lineup by “The Book,” he or she should understand that we don’t want to take Heywards bat out of the equivalent of a few games.
Offense is about gaining as many bases as possible and avoiding outs as often as possible. Batting McLouth second versus Heyward does neither. Heyward is better at gaining bases and at avoiding outs.
I like that better than the line up it appears Freddi will use. I’d probably keep Chipper batting 3rd and Prado 1st under the principle of if it’s not broken, don’t fix it. Uggla would then hit 5th.
Fantastic write-up. Too bad we will never see this line-up used.
A few comments:
1. Yaz/1 makes a good point: if the number of important situations projected to be faced by each spot in the lineup is based on analysis of traditional lineups, wouldn’t you expect slightly different results for a Book lineup? That said, the Book lineup isn’t that different from traditional lineups, and MGL/Tango surely anticipated this issue when writing the Book.
2. fjrabon/5 makes another good point: don’t upset the apple cart; if Chipper wants to bat third, he bats third. He bats first for me in video games, but not real life. The difference between the optimal lineup and a decent one where the good hitters generally bat before the bad ones is <10 runs a year.
3. But who bats first if it isn't Chipper? ZiPs seems down on Prado, who's projected to have just a bit better than average OBP. But he's the guy.
4. 8th really is a great spot for Freeman. It's low pressure, and Chipper and the guys will be telling him that he doesn't need to do everything and that turning the lineup over is really important. Don't be afraid to take a walk with 2 outs and a man on.
Scott Proctor has been released! Let all rejoice!
It still blows my mind that Heyward will be hitting 6th
I still have nightmares that if Nate doesn’t work out in the #2 spot that Fredi will bat Sea Bass there, as he did for most of spring.
Yaz & ThePetis & Hizouse,
Ask MGL or Tango, I don’t have the analytical chops to answer that.
5, 9
I’m sure the “Chipper might get mad” scenario wasn’t really factored into putting together the lineup. In fact, I don’t really think he would get mad at all being put in leadoff as long as he understood why. Not like you’re moving him down to the 8 spot.
Like everyone else, Heyward hitting 6th pisses me off. Do you think it’s purely an old-school thought process, thinking he will have more RBI opportunities? Or maybe they might be trying to change his approach to being more aggressive to increase his power instead of being so patient all the time? I don’t agree with that, just trying to understand the reasoning.
My lineup would be the typical boring: (LHP) Prado / Heyward / Chipper / Uggla / McCann / McLouth / Gonzalez / Freeman. And (RHP) Prado / Heyward / Chipper / McCann / Uggla / McLouth / Gonzalez / Freeman.
Though I would consider flipping McLouth and Freeman in the order.
fjrabon / 15
So Alex Gonzalez is Sea Bass right? And that’s referencing Dumb and Dumber correct? But I don’t understand the connection. Please enlighten me.
18: Yes, it’s Gonzalez. I’ve never understood it to be a movie reference; I always thought it was because he looked like a sea bass. (It’s the mouth.)
CAC, what team, if any, do you know of that does the best job of structuring their lineup with the method you used as opposed to the traditional viewpoints?
I’m sure the “Chipper might get mad” scenario wasn’t really factored into putting together the lineup
Oh, no. I just want to add that this isn’t necessarily what I would do or what I’m suggesting Fredi should do, just what The Book says.
Graham/20,
The Braves last year when they had Heyward 2nd. Seriously. The #2 spot in the line-up is the most misused line-up spot out there and managers could do more good by moving their best hitter there than any other adjustment.
Anyone else find it odd Prado and McLouth and both projected at .340?
Don’t really want to argue the numbers, it just seems weird since Prado hasn’t been under 350obp/450slg in the past 3 years but zips has him for 340/440? Seem to have him dropping 20 points under his babip through his first 1500 PAs.
Just seems to be slightly over projecting McLouth and putting Prado under.
As far as the Sea Bass nickname goes, from what I’ve read, Kevin Millar gave him the nickname a long time ago when they both played for the Marlins. Not sure why though.
How would you feel about the half of the order being:
Gonzalez
Freeman*
Pitcher
McLouth*
I can see McLouth’s speed being a real asset there, going from first to third on singles by Prado/Chipper (whoever you have at leadoff) or coming home on a single after a stolen base.
(Of course I do understand that some players might interpret batting 9th as being the ultimate insult. Dunno if that would affect their play or not.)
And of course I meant the bottom half of the order.
I have a feeling Heyward will be hitting 2nd and Freeman will be hitting 6th by the middle of the season. Nate, Gonzo, or a replacement CFer will be hitting 8th. Expect Heyward to be in the 3rd spot every day Chipper is out of the lineup.
I do like the idea of batting Nate 9th, followed by Prado, Heyward and Chipper. Prado is the exact kind of hitter you want at the plate with a base stealer on 1st. Having Nate hit 9th also essentially makes Heyward bat 3rd and Chipper bat 4th after the first inning.
Jeff,
Covered that somewhat in the 2011 Braves hitting post. McLouth’s preferred method of production involves walking and hitting the ball in the air for extra-base hits. Prado’s involves hitting singles. Single rate carries more variance, so you have to regress Prado’s batting average to the mean more than you regress McLouth’s secondary average.
You just can’t take Prado’s production over the past few years at face value.
Prado is the exact kind of hitter you want at the plate with a base stealer on 1st.
Why?
I’m not sure enough work is done on bottoms of lineups. I think it makes some sense to bat the pitcher as high as 7 in a batting order, especially if you figure the pitcher might only get 2 AB’s in a game (the others are PH’s).
Why not clear the pitcher earlier so you have some productive hitters coming back around before you get back to the top where your best hitters are?
Maybe there isn’t enough of a statistical difference to matter.
May the braves win it all in 2011 and beyond, let’s bring down those Giants or the Phillies in the postseason and that Nation known as the Sox in the Fall Classic.GO BRAVESNATION!!!!!!!!!!!!!!!!!
Prado also has a better percentage of hard hit balls and a greater contact rate. Prado also hits a lot of doubles and has one of the higher line drive percentages on the team. His HR + Double rates have steadily increased and then been stable the last two years. He’s also beat Zips estimates every year. At some point, you have to say an already questionable at times projection system might be off for him. This is 2 full years of a TON of AB’s.
interesting fact is that Prado is more of a line drive hitter than Heyward, who has one of the higher GB rates on the team, which will likely make him a guy we should expect to have wildly varying batting averages.
Heyward”s 55% ground ball rate seems like it could really be sapping his potential. It’s one of the only “bad stats” Heyward has.
Interesting quote from Keith Law’s blog today.
“Atlanta has the talent to surpass the Phillies, but they keep trying to sabotage themselves, like talking about batting Jason Heyward sixth, moving Martin Prado to left to install Dan Uggla at second, or choosing Brandon Beachy over Mike Minor based on the tiny sample of spring training.”
Unfortunately, these eight players will probably only start 35-40% of games even if there are no DL trips all season. Mac and Chipper will likely each play in the 120-130 range, and I’m not too impressed with a lineup that includes neither of them. I’d argue for Fredi to stagger their days off, but I’d understand the argument to maximize games with all hands on deck.
When Ross plays, I’d expect Uggla and Heyward to move up a spot in the lineup (assuming J-Hey is hitting 6th) and Ross be plugged in at #6.
Nick S is probably right that Heyward takes the 3 spot when Chipper is out. If Hinske/Conrad replace Chipper on a particular day, they probably hit 6th, while Young or Hicks would likely hit eighth, bumping (kick his ass!) Sea Bass and Freeman up a spot.
My apologies regarding the Dumb and Dumber reference for those who are unfamiliar.
29,
According to the book:
Stolen bases are most valuable ahead of high-contact singles hitters, who are more likely to hit at the bottom of the lineup.
Also, Prado takes a lot of pitches and works the count, giving the base stealer time to read the pitcher. He can also hit to the right side to take advantage of the hole created by the first baseman holding the runner.
Gonzo is pretty much the worst player on the team to hit behind Nate if you want to maximize the value of his base stealing. He is most likely to have a quick AB ending in an out, which would either bring up the pitcher or end the inning. Both of those cases completely negates the value of a stolen base threat.
A 2 out walk by Nate is next to useless with Gonzo and the pitcher batting behind him. If he is hitting 9th he at least has a chance to steal 2nd and be driven in by Prado or Heyward.
This is only in reference to maximizing Nate’s base strealing value…
At some point, you have to say an already questionable at times projection system might be off for him. This is 2 full years of a TON of AB’s.
2 years worth of AB’s doesn’t get you to a 95% confident estimate of true talent level. Closer to 10 years would.
Of course a projection system might be wrong. Projections are based on algorithms and can’t account for everything. ZiPS is the best one we’ve got, though, and if it’s incorrect in it’s estimation of true talent level it’s because a) we don’t have enough data or b) something about the player has changed. In this case we don’t have enough data to confidently say Prado’s MLB results thus far = his true talent level.
I am not saying ZiPS is right and Prado’s batting average on balls in play is going to drop, I’m saying we can’t confidently say what’s happened is what’s going to happen based exclusively on numbers yet.
Did you link to the rational discussion flow chart because people who disagree with you presumably will not “stop using [faulty arguments]” once shown to be faulty, or because you can’t “envision anything that will change your mind on this topic?”
Sabermetric propositions are true empirically, not logically. I could imagine a baseball world in which this advice might not be so wise (the dead ball era being one). Or a team like the A’s might attempt to construct an entirely OBP and speed team, eschewing power, for budgetary reasons. If all of your hitters are singles hitters, you may as well put the basestealers near the top of the order. Finally, managers might also prefer, ceterus paribus, a speedier high-OBP player to a slower one in order to make him more concerned about holding runners while more often while facing the team’s best hitters in the #2–4 holes. I’m sure that study has been done, and the effect is subsumed for many teams. But what is true on average is not true for all lineups.
I’m not saying I disagree with the post, but just that cloaking opposing arguments as inherently irrational is probably a bit much.
prado
heyward
chipper
mccann
uggla
freeman
gonzalez
(pitcher)
mclouth(his speed is like having another leadoff batter)
@ David / 38
I think what the issue is, people tend to stick with their opinions over the observed statistical findings. For example, “The Book” states that over its’ observed historical reference period, the #2 spot if an important spot in the line-up due to the propensity for that spot having people on-base, number of at-bats seen, etc; therefore should have “one of the best three hitters overall” filling that spot.
If a person were to argue with that point, they’d need an argument that counter-acts said statistically evidence with something beyond “thats where you need speed!”.
Basically, if you are going to argue, come with facts and/or substance to back up your belief.
I’d have no complaints with that lineup whatsoever. As another poster pointed out, I’d be curious to see it shuffled when Chipper takes a day off or when we play the Sunday lineup…just saw on Twitter that Jerry Crasnick says the Braves are targeting a RH bat for the bench. Be curious who they are looking towards and how soon.
I mainly just wanted to throw a link to that chart up so people saw it.
In this case Chipper’s lack of speed doesn’t disqualify him from being an effective leadoff hitter, though, and I think it’s absolutely appropriate to quash the argument before it starts.
I understand how it is calculated and Prado maybe isn’t actually a .355 woba hitter. I think I am more wondering why they have McLouth for 17 HR’s in about 520 PAs, while also putting Prado at three year lows in basically everything.
Anything in The Book about players who see large amounts of pitches?
That’s always been one of the reasons I love having Chipper as a top-3 hitter: not just because he gets on base but because he often sees 20+ pitches from the other team. He just tires out the starters. And it’s why I hate it when Chase Utley’s in the opposing line-up.
I am aware of the literature available for lineup optimization, but they all leave out one key element (and perhaps the most important one): THE HUMAN ELEMENT. The fact is, baseball is played by humans, not robots. The mind can play crazy tricks on people. So, with this in mind, if batting McClouth #2 in the order helps him to build his confidence to the point where he hits like Pittsburg Nate, then that is fine by me. Pittsburg Nate at #2 and Heyward at #6 is far greater than Heyward at #2 and Atlanta Nate at #6. I’m not saying this alone will be a factor, but it definately won’t hurt to try for a few weeks to start the season.
I think that’s a bunch of bullshit. If a guy needs to be handed the #2 spot rather than the #6 spot in the order to be confident in himself enough to produce he isn’t a major-leaguer. These guys are professional players, enough with the babying.
Edward,
ZiPS uses minor-league translations as part of the equation and Prado wasn’t nearly as good in the minors as he was in the majors. McLouth hit 26 homers in 2008 and 20 in 2009.
The only problem with the book on lineup construction that I have seen is this. If the vast majority of baseball managers have constructed their lineup in the traditional way of thinking, then wouldn’t the statistics used in the book be changed if a manager changed his lineup to the book? I guess what I’m trying to say is, that if a manager changed his lineup to exactly what the book suggests, then wouldn’t the most common scenarios that each hitter in the order faces change because of the new lineup? Hopefully someone more knowledgeable than me can understand what I am trying to say and clue me in.
@46
Hey! I didn’t ask that!
Fredi’s latest argument for hitting McLouth 2nd sounds a lot like Joshua’s -
“I know all my people from SABR,” Gonzalez said, of the statistics gurus at the Society for America Baseball Research. “But hey, we’re dealing with human beings. Sometimes guys get comfortable at a certain spot and they’re producing and they’ve got confidence and go get him.”
Also, this should add to the Mather to Francoeur comparisons -
“They want me to shorten my swing a little bit, which is always kind of what people say,” Mather said. “It all comes down to numbers. There are guys who have good numbers with long swings, and guys have bad numbers with short swings. So put up numbers, you’ll be back.”