BBTF's Newsblog Discussion :: Goold: Why doesn’t Albert Pujols bat 4th? Ask The Man

Baseball Primer Newsblog
— The Best News Links from the Baseball Newsstand

Wednesday, February 25, 2009

Goold: Why doesn’t Albert Pujols bat 4th? Ask The Man

Goold…La Russa…Musial…Baseball-Reference.com…Dewan…Schoendienst…Tracer Alert!

Into that volley of lineup questions and La Russa replies, I floated this trial balloon: Would you ever, during spring training, take a look at hitting Albert Pujols cleanup? Just to see …

La Russa looked like he had just swallowed a dime.

“Where did Stan hit?” he said. “I’ll leave it that.”

...So Musial hit most in the No. 3 spot, but appears to have hit best in the No. 4 spot — with far fewer at-bats and hardly any real value as a oranges-to-oranges comparison. The reason for all of this is clear: Baseball tends to bat its best hitter No. 3. That’s part of La Russa’s theory. He wants Pujols to bat in the first inning — even if there is ample evidence that average outcomes of innings show that the cleanup hitter often hits with a man on base in the first inning and gets more cracks with men on base.

...La Russa used history as his advocate.

“I’ll ask Red,” La Russa said, referencing Hall of Famer Red Schoendienst. “I should ask him that. Why didn’t you hit Stan cleanup?”

Repoz Posted: February 25, 2009 at 05:04 PM | 33 comment(s) Login to Bookmark
Tags: cardinals, history, sabermetrics

Reader Comments and Retorts

Go to end of page

Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.

1. phredbird

Posted: February 25, 2009 at 05:16 PM (#3085955)

he's gotta be kidding. even babe ruth batted 3rd. if its true that statistics show that cleanup hitters hit with a man on base enough to make that arrangement less than optimal, i can only say managers are probably thinking its beside the point. its maybe more about the pressure it puts on opposing pitchers right away.

2. Hello Rusty Kuntz, Goodbye Rusty Cars Posted: February 25, 2009 at 05:22 PM (#3085974)

"Because I didn't start managing until Stan was retired," Red would deadpan.

3. The importance of being Ernest Riles Posted: February 25, 2009 at 05:22 PM (#3085975)

1: in that case, Albert should bat #2.

4. OCF Posted: February 25, 2009 at 05:31 PM (#3085995)

1: in that case, Albert should bat #2.

I was about to say that myself. There's very little tradition behind it, but batting the best hitter on the team - a high-OBP slugger, the kind of talent that is far from freely available - second? That really does make a lot of sense.

For one thing, how often does such a great slugger hit his HR with the bases empty even batting cleanup? More than you might think. Best to move him up to get him more PA - and with one guy tasked with trying to get on base ahead of him.

5. plim Posted: February 25, 2009 at 05:46 PM (#3086025)

how often does such a great slugger hit his HR with the bases empty even batting cleanup

you mean like hitting a homer to lead off the 2nd inning?

6. Non-Youkilidian Geometry Posted: February 25, 2009 at 05:48 PM (#3086027)

For one thing, how often does such a great slugger hit his HR with the bases empty even batting cleanup? More than you might think. Best to move him up to get him more PA - and with one guy tasked with trying to get on base ahead of him.

But how do you know it is really "best"? There are advantages and countervailing disadvantages to all of the options, and I don't think you or I -- or anyone else, for that matter, including Tony LaRussa -- can accurately determine how they stack up against each other based on intuition, without the benefit of statistical analysis.

My recollection is that the statistical work that has been done on this question ultimately concludes that batting order actually makes very little difference.

7. Home Run Teal & Black Black Black Gone! Posted: February 25, 2009 at 05:55 PM (#3086038)

I believe the difference between a good and bad batting order is about two wins, which is either meaningless to a 65-win team or hugely important to a 88-win team.

And I think most studies conclude that you should hit your best batter third; unless, IIRC, you really don't have two good OBP guys to lead off with, in which case you should just give him the ABs. I seem to remember various recommendations that Barry Bonds, back in the days when he solely powered the Giants' offense, should hit second.

8. Randy Jones

Posted: February 25, 2009 at 05:59 PM (#3086043)

Pretty sure the studies on batting order said that the best hitter should be 4th. You get fewer PAs than the 3rd hitter, but more PAs with runners on, enough to make up for the difference.

9. The District Attorney Posted: February 25, 2009 at 06:05 PM (#3086054)

Yeah, if anything I would think hitting a great overall hitter 3rd is slightly counterproductive. You want SLG there to be sure, but OBP is a little wasted. At least if the guy is hitting 4th, he will lead off a lot of innings and the OBP will be useful then. Unless you just have a super lineup and have five or more terrific hitters to hit 1-5 (in which case, who cares what your batting order is? ;-), my ideal #3 hitter would not be the highest OBP guy, but would have high SLG and few GIDP.

This sort of logic is of course incorporated into the Baseball Musings lineup analyzer and taken to a ludicrous extreme, as you'll see if you put in the Mets' PECOTA projections and it tells you to hit Brian Schneider 3rd. (That's what it did last year, anyway.)

Hitting the best overall hitter 3rd, well, "it's always been done that way" (probably not literally always, but for quite a while anyway), that's true... but it doesn't necessarily mean it's optimal.

I use the word "optimal" advisedly, as it pretty clearly doesn't matter much either way.

10. Home Run Teal & Black Black Black Gone! Posted: February 25, 2009 at 06:08 PM (#3086057)

http://www.hcs.harvard.edu/~mbodell/battingOrder2001.html says the best orders are in descending OPS. In other words, best player at the top. Second best next. Etc.

http://fantasyscope.wordpress.com/2008/01/22/how-to-optimize-a-lineup-sabermetric-style/ uses the Baseball Musings tool and provides two examples. For the Yankees, with some good OBP guys in the team, it has Rodriguez hitting fourth. For the 2007 White Sox, with no good OBP guys on the team, it has their best guy (Thome) first and their second best (Konerko) second.

http://www.baseballmusings.com/cgi-bin/LineupAnalysis.py

I come away from this brief search with my opinion reinforced: Unless you can present above league average table-setting opportunities for your sluggers, make your first concern getting them more ABs.

11. More Dewey is Always Good Posted: February 25, 2009 at 06:10 PM (#3086065)

I believe the difference between a good and bad batting order is about two wins, which is either meaningless to a 65-win team or hugely important to a 88-win team.

By "bad lineup", do you mean a lineup that nobody would use (like leading off with the pitcher and batting Pujols 7th), or a lineup that conceivably would exist in the real world, but doesn't make a whole lot of sense?

Because I would think that the difference between batting Pujols 2nd or 3rd or 4th would be so marginal as to be non-existent.

12. Famous Original Joe C Posted: February 25, 2009 at 06:19 PM (#3086075)

Pretty sure the studies on batting order said that the best hitter should be 4th. You get fewer PAs than the 3rd hitter, but more PAs with runners on, enough to make up for the difference.

FWIW, "The Book" (MGL/Tango/Dolphin) say:

Your best three hitters 1, 2, and 4 (and the very best at 2).
Fourth and fifth best hitters in 3 and 5.
Slots 6 through 9 in descending order of quality.
Also, 1 and 2 should be players who walk more often than the 4 and 5.

13. Home Run Teal & Black Black Black Gone! Posted: February 25, 2009 at 06:20 PM (#3086080)

By "bad lineup", do you mean a lineup that nobody would use (like leading off with the pitcher and batting Pujols 7th), or a lineup that conceivably would exist in the real world, but doesn't make a whole lot of sense?

Yeah, I'm a little fuzzy on that. The first link I posted showed a 4-win swing between the absolute worst imaginable and the descending-by-OPS lineup. I can't locate any Prospectus/Hardball Times studies on this (which I'm sure I've read before), but, with the added proviso that I'm going on memory here, I think the difference between a crappy but plausible lineup and an optimal lineup can be as high as two wins.

Yeah, it probably doesn't matter a ton whether you bet Pujols 1-5; how many times, though, have we seen teams trotting out sub-.330 OBP guys in the 1 and 2 hole because (a) They're speedy! (b) They're slap hitters! and (c) Why, you've got to hit some combination of SS/2B/CF in the first two spots!

Juan Pierre batting first for the Dodgers all year has got to cost them a win from sheer lineup construction-related stupidity.

14. Home Run Teal & Black Black Black Gone! Posted: February 25, 2009 at 06:21 PM (#3086083)

bat*

15. Tango Posted: February 25, 2009 at 06:22 PM (#3086088)

I spent a month of my life answering this question. Don't make me feel ashamed that I should have spent it playing with my kid.

Go to Amazon, use their Look Inside feature and do a search for "CHAPTER 5" then read the research.

I will highlight that moving your pitcher from best spot (8th) to worst spot (cleanup) costs about 15 runs per season.

So, you can imagine how much impact moving Pujols from 3rd to 4th will have.

16. Randy Jones

Posted: February 25, 2009 at 06:24 PM (#3086093)

FWIW, "The Book" (MGL/Tango/Dolphin) say:

Your best three hitters 1, 2, and 4 (and the very best at 2).
Fourth and fifth best hitters in 3 and 5.
Slots 6 through 9 in descending order of quality.
Also, 1 and 2 should be players who walk more often than the 4 and 5.

Yeah, that's it. I haven't read The Book in a long time.

17. TomH Posted: February 25, 2009 at 06:37 PM (#3086114)

FWIW, "The Book" (MGL/Tango/Dolphin) say:

Your best three hitters 1, 2, and 4 (and the very best at 2).
Fourth and fifth best hitters in 3 and 5.
---

I understand the reasoning and math behind this, but I'm still uneasy with the conclusion that the #1 hitter is more important than the #3, considering that the #1 hitter comes up to the plate 5% more, but hits with 25-30% fewer men on base.

I'm about to have published an article in SABR's quarterly By The Numbers, dealing with how batters who hit leadoff are overrated by most of our modern stats (linear weights, runs created, base runs); because they GET credit for batting more often, but they do NOT GET credit taken away for batting in less-leveraged spots than vevery other lineup spot (including #9).

18. You can keep your massive haul Posted: February 25, 2009 at 06:38 PM (#3086117)

Because Wieters isn't on the team to bump him out of the #3 spot?

19. Home Run Teal & Black Black Black Gone! Posted: February 25, 2009 at 07:01 PM (#3086162)

TomH,

That's a damn fine point. I guess it's like looking at leadoff hitters the same way we do closers, huh?

20. phredbird

Posted: February 25, 2009 at 07:12 PM (#3086184)

i can see the arguments for making your best hitter bat #2, but if, as everyone seems to agree, the advantages are in tiny increments, i'd still feel more comfortable putting him #3. he's guaranteed an atbat in the first inning, and he's also given a better shot at coming up with a man on. and the best possible outcome is still in order.
think like a manager. you want as many sure things as possible.

21. villainx Posted: February 25, 2009 at 07:15 PM (#3086190)

I just hope La Russa manages his drinking better this spring training.

22. Home Run Teal & Black Black Black Gone! Posted: February 25, 2009 at 07:16 PM (#3086192)

But since your best hitter usually has the best chance of getting on base (including second base) wouldn't you want the decent hitters behind him to have 0/1 outs to knock him in versus 1/2?

23. Home Run Teal & Black Black Black Gone! Posted: February 25, 2009 at 07:17 PM (#3086195)

wouldn't you want the decent hitters behind him to have 0/1 outs to knock him in versus 1/2?

This should say: Have 0/1 outs on the board when trying to knock him in versus 1/2.

24. Swedish Chef Posted: February 25, 2009 at 07:19 PM (#3086199)

Is there a difference between the AL and NL? Seems like you really wouldn't want your best guy directly after the pitcher.

25. JMPH Posted: February 25, 2009 at 07:27 PM (#3086211)

Is there a difference between the AL and NL? Seems like you really wouldn't want your best guy directly after the pitcher.

One way to avoid this is to go with the second leadoff man theory--batting your pitcher eighth so that someone sets the table for the leadoff man. La Russa is no stranger to this.

26. OCF Posted: February 25, 2009 at 10:05 PM (#3086458)

What should you do if your lineup has Ruth and Gehrig? The actual Ruth/Gehrig lineup had them batting 3/4; is that optimal? Post 16 suggest the possibility of having them either 1/2 or 2/4 with a lesser hitter between them.

27. Obama Bomaye Posted: February 25, 2009 at 10:25 PM (#3086491)

Plus in the current game you might want to split the lefties. (Though I'm guessing neither had too severe a platoon split.) In the '20s, not a big deal.

Per The Book, you'd probably want the '27 Yankees to go something like:

Combs
Ruth/Gehrig
Meusel/Lazzeri
Gehrig/Ruth
Lazzeri/Meusel
Collins
Koenig
Pitcher
Dugan

Of course, that's based on what we know happened. At the beginning of the season you would have somewhat different expectations for some guys. You don't know how much Lazzeri will improve or that Gehrig will have so much HR power.

28. Famous Original Joe C Posted: February 25, 2009 at 10:26 PM (#3086492)

What should you do if your lineup has Ruth and Gehrig?

Bat them in the top four and score a crapload of runs?

29. Hello Rusty Kuntz, Goodbye Rusty Cars Posted: February 25, 2009 at 10:43 PM (#3086510)

What should you do if your lineup has Ruth and Gehrig? The actual Ruth/Gehrig lineup had them batting 3/4; is that optimal? Post 16 suggest the possibility of having them either 1/2 or 2/4 with a lesser hitter between them.

I remember discussion of whether the White Sox should bat Frank Thomas and Albert Belle 1-2 on r.s.bb or whatever the BTF of the time was.

30. OCF Posted: February 25, 2009 at 10:48 PM (#3086523)

how many times, though, have we seen teams trotting out sub-.330 OBP guys in the 1 and 2 hole because (a) They're speedy! (b) They're slap hitters! and (c) Why, you've got to hit some combination of SS/2B/CF in the first two spots!

On the other hand, what if the people you have that fit that general description all have OBP in the .400-.420 range? The bb-ref page I have open right now is that of the 1999 Indians. What they did was bat Lofton/Vizquel/Alomar (lots of OBP, lots of mobility on the basepaths, only Alomar any kind of HR threat) 1/2/3 and then put Manny's .660 SLG in the cleanup spot. Whether that was or wasn't best for the team, it certainly was best for Manny piling up RBI. Theoretically, I suppose Alomar (probably the 2nd best offensive player on the team when you include his baserunning) would have been better placed at #2 than #3, and there a lot of other things you could do with that lineup. What you couldn't do no matter what you tried would be to prevent that team from scoring lots of runs.

31. Ron Johnson Posted: February 25, 2009 at 11:06 PM (#3086545)

James was able to get the difference between the best and worst lineups up to 50 runs. But that was using the 1930 Cubs who:

A) scored a ton of runs
B) had some pretty specialized players (as well as a few stiffs) English in particular and Cuyler to a lesser extent were unusually strong on the run scoring side.

A huge chunk of that 50 runs had to come from separating Woody English and Hack Wilson.

32. Srul Itza Posted: February 25, 2009 at 11:10 PM (#3086550)

Don't forget that Jim Thome at .426/.540 and David Justice at .413/.476 were also on that 1999 Indians team. Damn, that team was stacked.

33. OCF Posted: February 25, 2009 at 11:15 PM (#3086558)

A huge chunk of that 50 runs had to come from separating Woody English and Hack Wilson.

My attempt to sabotage the 99 Indians:

1. Diaz
2. Thome
3. Vizquel
4. Fryman
5. Justice
6. Alomar
7. Lofton
8. Sexson/Cordero
9. Ramirez

Ah - they'd still score a ton of runs. But Manny would probably only have something like 110 RBI, not 165.

You must be Registered and Logged In to post comments.

<< Back to main

Wednesday, February 25, 2009

Goold: Why doesn’t Albert Pujols bat 4th? Ask The Man

Reader Comments and Retorts

BBTF Partner

Support BBTF

Hot Topics