It’s an affront to Managing 101: Whether you look at the batting order as a way to set up the offense or simply as a vehicle for distributing the most plate appearances to your best hitters, it has long been accepted baseball doctrine that your leadoff hitter should have a high on-base percentage. Yet, every day you can look at the box scores and see Orlando Cabrera and his .276 on-base percentage taking the first hacks of the day for Dusty Baker’s Cincinnati Reds. Drew Stubbs opened the season in the leadoff spot, occasionally yielding to Chris Dickerson, but when both started slowly, Cabrera got the nod and has been rooted at No. 1 ever since.

At .238/.276/.327 (.231 True Average), the 35-year-old shortstop is hitting well below his career rates of .273/.320/.395 (.250 TAv), yet as those rates suggest, even peak-form Cabrera isn’t exactly Max Bishop or Rickey Henderson. His highest single-season on-base percentage was .347, recorded in 2003. He also reached his career high in walks that year with 52. Cabrera does have one small positive as a leadoff hitter. Since 2001, he has been an excellent percentage basestealer, swiping 193 bags in 232 attempts, an 83 percent success rate. This season (through Wednesday) he’s 9–for-9. Since, to invoke a cliché, you can’t steal first base, and Cabrera has never been a high-volume basestealer in any case (averaging 21 per 162 games since 2001), his baserunning does not add to his credentials as a leadoff hitter.

Baker’s inability to exploit the leadoff spot is nothing new. Now in his 17th season as a manager, the aggregate rates for a Baker leadoff hitter are .264/.327/.382. Baker leadoff hitters have drawn more than 87 walks in a season only once, in 1998. More often, they have drawn fewer than 60. Overall, his leadoff hitters have averaged 749 plate appearances and 61 walks per season. While we would expect leadoff hitters to have a higher walk rate than the average hitter, Baker’s leadoff men have actually had a slightly lower walk rate than the average National League non-pitcher.

Despite Baker’s oft-expressed disdain for slow runners who clog the bases no matter their other positives, it’s not that he has been blinded by speed, choosing fast players to lead off, or if he has, it hasn’t shown up in the results. Baker leadoff men have stolen as many as 58 bases and as few as 16 (with 12 caught stealing); the average has been 29. With the exception of his 2006 leadoff man, Juan Pierre, and the two half-seasons in which his teams traded for Kenny Lofton (with the 2002 Giants and 2003 Cubs), he hasn’t had a classic burner at the top of the order, and by the time he got Lofton, the outfielder was 35 and his basestealing had been greatly curtailed.

Baker is not unique among managers in having trouble in finding an adequate leadoff hitter when the choice is not obvious. Think of Ralph Houk, who led off .261/.295/.316-hitting Bobby Richardson for the 1961 Yankees and got 90 runs out of the leadoff spot (80 of them belonging to Richardson, who had 706 PAs) despite having more home runs than any team in history coming up next. This was the kind of choice that Houk indulged in again and again during a 20-year managing career, but when he got to Detroit and had Ron LeFlore, he had an epiphany and stopped leading off the crappy second basemen (he batted them second instead). Casey Stengel once said that he didn’t need his hair parted with an axe to get an idea from the outside. Houk needed the axe. Similarly, when Baker had Lofton he knew what to do with him, but the rest of the time he couldn’t sort out his options.

In Baker’s defense, identifying the leadoff man on his roster hasn’t always been easy. The big on-base threat on his first team was Barry Bonds, who seemed a more obvious three-four hitter (Baker batted him fifth, which suggests a problem greater than just his taste in leadoff men). A good, though unorthodox choice, would have been Will Clark, who had lost much of his home-run power at 28. Robby Thompson was having his best season (his last good one) but he hadn’t been a great on-base threat at prior points in his career. Baker went with the veteran, Willie McGee, then the punchless outfield gloveman Darren Lewis. Result: abject failure. The first spot in the Giants’ order hit .252/.301/.313 that year.

Baker stuck with Lewis in 1994 and was still using him in July 1995, at which point the Giants included him in a package that brought back Deion Sanders and four other players from the Reds. Though Sanders had hit only .240/.296/.326 to that point in the season, Baker made the football star his leadoff hitter. Sanders hit well enough in 52 games (.285/.346/.444; .273 TAv) that he qualifies as one of Baker’s best choices, but he left as a free agent at the end of the year and was incapable of maintaining that level of production in any case.

For the next several years, Baker would devote himself to making Marvin Benard his leadoff man, but Benard’s middling hitting and resultant inconsistency meant that the manager would have frequent recourse to Stan Javier, Darryl Hamilton, F.P. Santangelo, and Calvin Murray. For two years, 1998 and 1999, the aggregate result was quite good. In the former season, Giants leadoff hitters average .297/.392/.386, in the latter, .288/.368/.458. After that season, Benard’s time as a useful hitter would be over, while Murray’s would never come, but it took Baker two years to be sure.

Since the Benard years, Baker’s leadoff men have been a fascinating congeries of ill-fitting spare parts, including (but not limited to) David Bell, Tom Goodwin, Tsuyoshi Shinjo, Mark Grudzielanek, Lenny Harris, Jose Macias, Todd Walker, Corey Patterson (in three different seasons), Jerry Hairston, Neifi Perez, Ryan Freel, Jay Bruce, Willy Taveras, and the aforementioned Lofton, Pierre, and Dickerson. Baker’s career is bookended by the disaster of his first season and 2010 leadoff men, but those aside, the nadir of his mid-career choices came in 2005, when Hairston, Patterson, and Perez, the last two of which with career OBPs below .300, combined to hit .245/.299/.358 in the leadoff spot. Hitters at the top of the Cubs’ order that year scored a grand total of 83 runs.

Sometimes, Baker’s teams have even outsmarted themselves. In 2005, primary leadoff hitter Walker was hitting a solid .298/.346/.472 when Baker kicked him out of the top spot for good, first for Grudzielanek, then for the Moby Dick of Baker’s career, his irresistible force, Patterson.  The current Oriole led off for 55 of the final 58 games of the season, hitting .211/.263/.359 in the spot. Give Baker this much credit: Patterson’s leadoff rates were better than his overall .215/.254/.348.

It is difficult to understand why a manager would choose to handicap his team the way Baker has in most seasons. He is clearly not a stupid man, and yet he continually makes batting order choices that are difficult to describe in any other way. It’s as if some psychological trauma prevents him from seeing the errors in his thinking. It might be that his career provided poor role models, leadoff exemplars that he’s still trying to find facsimiles of for his own batting orders, someone like Ralph Garr.

Baker played with Garr in the Braves’ system, and then again in the majors, where Garr was the leadoff hitter. Garr was a terrific hitter at his best, but his approach didn’t leave much room for error. He was an Ichiro Suzuki-style hitter, averaging .306 for his career, peaking in 1974 with a league-leading .353. He had speed and a contact-oriented, line-drive approach which led to as many as 17 triples in a season. The downside was that he rarely walked, so when he didn’t hit over .320, which was most of the time, he didn’t contribute that much on offense. His career OBP was only .339.

At his peak, Garr would have made a fine third- or fifth-place hitter, but he spent his entire career hitting first or second. This was how baseball liked its leadoff men in the 1970s and early ‘80s; walks weren’t important, but if you hit .290- or .300-worth of singles and threw in some steals, you were getting the job done. Garr wasn’t very different from Mickey Rivers, Dave Collins, Willie Wilson, or Mookie Wilson, and McGee wasn’t too far off either, though he spent more time hitting lower in the order (McGee’s MVP-winning 1985 is a dead ringer for Garr’s 1974). They almost exclusively hit their way on base rather than got there through a combination of hitting and selectivity.

Baker saw Garr at the height of his career. They were together at Triple-A Richmond when the latter hit .386/.426/.522, in the majors for part of the season in 1971, when Garr hit .343/.372/.441 in 154 games, shared the outfield with him as a regular in 1972 when Garr hit .325, and in 1974 when he won the batting title. In none of those seasons did Garr take more than 29 unintentional walks. Perhaps Garr remains Baker’s leadoff role model. Yet, there are only so many .300-hitting line-drive hitters with speed, and Baker hasn’t had them. Moreover, Baker’s teams, in fact, any teams, are almost certainly not looking to draft them. If Ichiro had been born in the United States, he’d be a cornerback. Pure singles hitters who put the occasional liner in the gaps, no matter how fast, are not beloved of scouts and scouting directors.

Baker’s subsequent career exposed him to a more rounded leadoff hitter in Davey Lopes, a terrific percentage basestealer who also had power and patience. He seems to have made less of an impression than his successor, Steve Sax, another player who sometimes hit his way on base, and Dan Gladden, who for a glorious half-season was Garr all over again. Baker went out with the A’s, playing with them in 1985 and 1986. The A’s employed three primary leadoff hitters in that period: Tony Phillips (career .374 OBP), Collins (.338, but fast), and Alfredo Griffin (.285, and perhaps the worst baserunner in modern history).  If you not only saw but accepted Griffin as a leadoff man, it probably required no cognitive acrobatics to place Perez atop a batting order, for they were particularly the same player.

As for the psychological insult that made Baker so wary of those “base-cloggers,” there is no smoking gun, but perhaps the blow came on June 3, 1973. The Braves were managed by future Hall of Fame third baseman Eddie Mathews, who might have been a bit over his head. Garr was already established in the majors, but had gotten off to a painfully slow start, hitting .234/.254/.292 through the end of May. At that point, Mathews decided to drop him out of the leadoff spot. Here is the lineup card that Baker saw on June 3, with the rates of the starters coming into the game:






Johnny Oates




Mike Lum




Darrell Evans




Hank Aaron




Ralph Garr




Dusty Baker




Davey Johnson




Marty Perez




Ron Reed



The late Johnny Oates, best remembered as the manager of the Orioles and Rangers, was a career platoon catcher who never hit much, batting .250/.309/.313 in a career that lasted 593 games and 11 seasons. He was best known not for getting on base, but for having an excellent arm behind the plate, something attested to by his career 38 percent rate of throwing out basestealers. He was 27 and coming off of a season in which he’d hit .261/.332/.364 for the Orioles, a performance that was better than it looks given that the American League hit only .239/.306/.343 that year. Still, it was an 85-game, 288-PA performance that was just decent. It didn’t mark Oates as the second coming of Ed Yost. In the game, Oates went 1-for-4 but didn’t score a run—the Braves were shut out by the Cubs’ Milt Pappas and Bob Locker.

June 3 wasn’t the first time Mathews had led off with Oates; he tried it in early May, with Aaron hitting second and Baker hitting third. This was, however, the first time that he stuck with it, leaving the catcher at the top of the order for 10 of 12 games beginning a few days later. Oates hit .250/.321/.313 and a Braves team that went 76-85 overall went 8-4. Baker himself got crazy hot, going 17-for-40 (.425) and driving in eight runs, so it’s not like he can harbor an irrational hatred of Oates for clogging up the bases so much that he himself somehow slumped.

 No, the humiliation of hitting behind a journeyman catcher does not seem to be the source of Dusty’s pain, but an antiquated sense of what makes a proper leadoff man might be the real source of the problem. A gar is a slippery fish that swims in brackish waters; for Baker, a Garr may be a slippery notion that swims a troubled mind. Any leadoff man might be a Garr if he is fast enough and can be made to hit for a high enough average. Baker is forever seeking his old friend and teammate, and the potential damage to his team’s standing is no obstacle to the continuation of that quest.   

Thank you for reading

This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.

Subscribe now
You need to be logged in to comment. Login or Subscribe
The old belief in leading off a relatively quick middle infielder still lives. Both Dave Trembley and Juan Samuel (no difference in managerial approach... including the losses) have led off with Patterson (.306 OBP) and, believe it or not, Julio Lugo (.290 OBP) leading off this year during the absence of Brian Roberts. Kind of a 'let's start with one hand tied behind our back' approach. Trembley also tried Adam Jones (.298 OBP), during his deep-in-a-slump days, and do not appear to have looked at the stats where Nick Markakis (.390 OBP, 3 HR, 23 doubles) appears to be the best choice. Do any of these managers read?
It's a minor point, but I think leadoff hitters actually have a slightly lower walk rate than other lineup slots other than P. From 2008-2010, here are the numbers:

#1: 4923 BB in 55974 PA (8.80%)
all non-P: 39689 BB in 443068 PA (8.96%)
all: 40249 BB in 457678 PA (8.79%)

Of course, one would have to assume that leadoff hitters are much more skilled at drawing walks than the norm, but pitchers are also highly motivated to avoid walking them.
Sort of like the blackjack player thinking that a couple of 21's means I'm hot and I need to bet more. Dusty thinks the hot streak is just around the corner with these guys and he doesn't want to miss out.
Thanks for the great article Steven. Had to go look up Alfredo Griffin again - I'm always amazed that he lasted as long as he did. And also amazed that one season he stole 18 bases - in 41 attempts! Didn't he ever get told to stop trying to steal?
And he's the Angels first base coach. How?
"[I]t has long been accepted baseball doctrine that your leadoff hitter should have a high on-base percentage."

What's your definition of "long"? I think that SABRmetricians have been telling each other this for the past 30 years, but it's only in the past fifteen that the idea has seen wider mention.
I debated how to phrase that particular point to account for the very issue you mentioned, but in the end I settled on the definition of "long" as "long enough that Dusty should know better" or "longer than his managerial career."
Well, Dusty's "walks clog the bases" line was four or five years ago and a lot has changed in my life since then :P
I think we overlook the powerful effect of symmetry on the human psyche. We love things that are symmetrical; it's in our hard wiring.

There's something that just feels right about a bell-shaped lineup. Add in the aesthetic beauty of the imagined sequence:
1) Lead-off slaps a single and steals 2B.
2) #2 hitter exhibits bat control in grounding out to 2B, advancing the runner to 3B.
3) #3 hitter either gets a hit or a sacrifice to drive in the runner.

I think that when many managers make out a lineup card, this scenario is on their mind. They choose players specifically based on their perceived ability to bring this sequence to life. The abstract value of OBP is trumped by the anticipated pain caused by watching a slower runner unable to put himself in scoring position (an antiquated concept which itself devalues getting to 1B) or fail to score. It feels lazy.

Dusty seems to base much of his coaching philosophy on what he learned and felt as a player. It was about being active and making things happen. Playing the game the right way involves scenarios such as the one laid out above and his lineups seem designed to realize it. Abstract concepts about designed to take greatest advantage of all possible scenarios just don't fit in to his mode of thinking. The idea of a slower player who may not have great bat control being at the top of the lineup just runs counter to everything he values. If he can get OBP from that fast guy, great. If not, oh well, at least he can make things happen.
I was about to write that Dusty must have an irrational hatred of sabermetrics. But when you write "It was about being active and making things happen," something clicked. Walks are the foundation of OBP, but they're also boring and involve little movement. They're the absence of swinging the bat. It may be that Dusty finds them abhorrent for the reason that they seem like the antithesis of "exciting" baseball.

If so, it's harder than you might think to disagree with that reasoning. Walks are definitely boring compared to hits. However, if his goal to win more games trumps his goal to have lots of swinging, he'd better get a walker in there.
Has Dusty done more damage to the Reds with his leadoff choices than Cito Gaston has done to the Blue Jays with his choice of the 2 and 3 hitters. 70 games into the season and he thinks it might be time for a change. You think?
Weaver on Strategy - paperback, 1982
Two things, from a Baker criic:

- He hasn't had a lot of leadoff options with the Reds. Yes, he's usually found the worst possible person to lead off, but it's not like Jonny Gomes or Alex Gonzales were good candidates.

- I would consider Taveras (47 SB per 162 G, career) and Freel (39/162) both "classic burners." Freel was hurt, and Taveras was horrific, but both were clearly speedsters, particularly in modern terms.

I agree with the general premise, though. A manager's playing career often shapes how he sees the game.
... but a manager has a say in what options make up their roster and their lineup, so I think lacking an option beyond Jonny Gomes or Alex Gonzalez is Baker's fault.
Really? Walt Jocketty and ownership have no say in roster construction? Again, I can't believe I'm defending Dusty Baker, but there's a difference between accountability and scapegoating. Alex Gonzales, in any event, was there before Baker arrived.

But who would you lead off on this year's roster? If it's not Brandon Phillips (Dusty's latest choice - so far, so good), then who? If it's someone not on the roster, how do you get them there? Who sits? Who's traded? How do you pay for it?

Last year, Taveras just imploded. I didn't expect success, but I did expect something just south the neighborhood of his (at the time) career .330 OBP.

I'm sure Baker would've played Patterson anyhow, but the legitimate candidates (Freel, Keppinger, Hopper, Dickerson) simply couldn't stay on the field.
Just because I say Baker has input in roster construction doesn't mean Jocketty and ownership somehow have no say. Nor do you hire and retain an expensive manager salary-wise like Baker unless you take in his input, nor do you give him players without asking what he thinks. I would also think managers have some say in who makes the roster out of spring training and who gets promoted/demoted during the season.

So to reiterate, rephrase, underline and highlight, Baker has input but not complete control of his roster construction.
How does the manager have a say in roster construction?
Another great article. I have often wondered why Dusty insists on batting awful lead-off hitters. Having to watch Neifi "First Pitch Express" Perez lead-off for the Cubs in '05 was particularly excrutiating (and he wasn't even particularly fast). The other question that leaps to mind is why teams are so eager to employ Dusty.