« Rodrigo the meanest bull in the pen | Main | What's on second? (Or is that who?) »

The case for bullpen generalists

Washington Post columnist Thomas Boswell's latest column discusses the Orioles' attempt to break with the modern orthodoxy of bullpen usage. Instead of going for lefty-lefty and righty-righty matchups that typically last one or two batters, manager Lee Mazzilli has often gone with "the next best pitcher" regardless of handedness.

No single person is credited with the strategy, but if the quotes from the story are any indication, team co-VP Mike Flanagan is a key proponent.

"We finally got sick of seeing our games being lost by our 11th or 12th best pitcher in some matchup situation," said co-general manager Mike Flanagan. "Too often, we never even got the game into the hands of our closer because we'd lost somewhere along the way with one of our worst pitchers. Now, we usually only warm up one reliever, then we bring him in -- our Next Best Pitcher -- regardless of who is hitting."

Flanagan also broached this philosophy in a Peter Gammons story on ESPN.com in February.

The O's bullpen should be deep, but Flanagan says, "we want to get away from the lefty-righty thing. We could have from three to five left-handers out there, but John Parrish gets right-handers out better than left-handers, and the rest are equally good against either side. That knee-jerk matchup thing is something we're trying to get away from."

Yesterday came suddenly

Boswell calls the Orioles' strategy "a radical new idea about how to construct a bullpen," but in concept it is actually more of a reversion to the bullpens of an earlier period, before the hyper-specialization of bullpen roles evolved in the 1980s.

Prior to the rise of the save statistic and the closer, bullpen roles were less narrowly defined. Teams used to carry four or five relievers instead of the six or seven common today. Although there was a loose hierarchy in place—the better relievers pitched more often than others did—nearly any reliever could be summoned to enter a game at nearly any time. Relievers did not pitch in as many games as they do today, but when they did pitch they often went multiple innings.

Take your baby by the hand

Then in the 1970s, the era of relief specialization began. Left-handed specialists, closers, setup men—all these roles have grown into their present-day usage over the last 25 years or so. Some trace the rise of the modern relief pitcher to Sparky Anderson's Reds teams of the '70s, but Tony La Russa, whom Boswell calls "a serial matchup recidivist," was the man most responsible for the trend of increasing bullpen specialization. The availability of context-based statistics, such as batter-versus-pitcher records and platoon splits, made it possible for La Russa and others to construct a bullpen of pitchers with specific skills that could optimally "match up" against hitters in crucial situations.

Such a bullpen might contain some pitchers who excel against left-handed batters, and others who are better against righties; some pitchers who are particularly good at getting ground balls, others who can bring the high heat, etc. The emergence of the save statistic led to "closers," who are used almost exclusively in ninth-inning save situations, and "setup men," who typically enter in the seventh or eighth inning to bridge to the closer. With so many pitchers confined to rigid roles, most relievers have come to be used for an inning or less per appearance. La Russa's mix-and-match, percentage-based bullpen usage has become standard practice around the major leagues.

Take it to the limit one more time

Yet a few baseball analysts, including some of the sabermetricians who initially supported the approach, have criticized the extent to which bullpens have evolved. The arguments against the modern bullpen usually claim that it has resulted in a misallocation of roster spots: too many seldom-used, mediocre pitchers in the bullpen, and too few position players on the bench to substitute in close games. That is, when a team uses its relievers to face a few batters at a time, it is obligated to carry extra pitchers to avoid using up its entire bullpen during a game. But having six or seven men in the bullpen leaves the bench with only four or five position players. As a result, managers have fewer offensive options to combat the very matchups for which those extra pitchers were designed. Flanagan and the Orioles seem to be aware of this strategic paradox. From the Boswell article:

If the Orioles ever establish a solid rotation, a moot point at the moment, they have such a strong bullpen behind them that they may trim down to just 10 pitchers. Yes, 10. Pure heresy. Is this 1956? Most teams now have 12. "If we can trust several relievers to work [multiple] innings, and our starters are taking us deep into most games, that would give us the flexibility to carry two extra hitters," said Flanagan.

Although that "if" is a big step, an extra hitter or two on the bench would provide a tangible benefit. The Orioles currently lack a power threat to insert in pivotal late-game situations. Another missing luxury is a second backup catcher for games in which Javy López is the DH.

Lived and learned from fools and from sages

The Orioles' neo-retro-bullpen approach, which Boswell has dubbed the "next-best-pitcher theory," may have its roots in the mind of Earl Weaver. Weaver preferred a small, quality pitching staff and a long, deep bench of position players. He kept a four-man rotation long after most teams had lengthened theirs to five, simply because "it's easier to find four starting pitchers than five" and he saw no medical reason why starters could not continue to handle the workload of pitching every fourth day. Moreover, he thought that "ten pitchers are too many" to have on a roster. His rationale in his classic 1984 managing text Weaver on Strategy was this:

With ten pitchers, one guy usually ends up rusting away. Rather than the tenth pitcher I'd rather have an extra player I could use to pinch-hit or maybe pinch-run. I believe that last regular player will help you win more games than a tenth pitcher. That extra player will be in the close games, while that extra pitcher will be on the mound in the blowouts.

—from page 79 of the 2002 edition of Weaver on Strategy

Given a four-man rotation, the logical implication is that Weaver wanted no more than five relievers. Nowadays, paring down to a nine- or ten-man pitching staff would be a bold move, to say the least. Not only has the role of the relief pitcher changed, but starters are also pitching less innings than ever before. Offense has taken the upper hand over pitching and defense in today's game, resulting in more pitches being seen per plate appearance and more pitches being thrown per inning than was the case 30 years ago. Add to that recent medical research linking high pitch counts and injury risk, and the effect is that starters typically last about six innings and rarely go more than seven before being pulled.

Always something there to remind me

Is this front-office regime conducting a reWeaver-ization of the Orioles? Even though it's been nearly twenty years since Weaver managed the team, it's apparent that some of his core philosophies about building and running a team hold significant influence over the current management. Last year, the new front office declared an organization-wide emphasis on on-base percentage, pitching, and defense—all Weaver hallmarks.

The material results of this ideology were scarcely evident at the major-league level in 2003, but changes have slowly been taking hold. Signing three power bats last offseason, for instance, signalled the return of more three-run homers to the Birds' lineup. Weaver himself dropped in on the Orioles' spring training camp this year to watch team workouts and chat with Mazzilli, and two weeks later the Post's Dave Sheinin spotted Maz studying a copy of Weaver on Strategy. Now the club has unrolled the "next-best-pitcher" bullpen model.

All of this is hardly surprising, given that many of the team's central decision-makers came of age in the 1970s, when Weaver was building his legend as skipper of the team of the decade. Out of the Orioles' top brass, Flanagan is the most prominent link to the Earl of Baltimore: Flanagan broke into the majors under Weaver and spent all or parts of ten seasons under Weaver's leadership, so he naturally remembers what worked so well back in the Birds' heyday. And coaches Rick Dempsey, Terry Crowley, and Elrod Hendricks are also steeped in the Oriole Way as it was defined by Weaver.

Turn the page

But the Orioles' strategy room is not an echo chamber of old, Birdbrained ideas. Rather, there is a combination of the old school and the new, of minds trained inside and outside the organization. Until they were hired, Beattie, Mazzilli, farm director Doc Rodgers, and scouting director Tony DeMacio had never set foot in the Orioles' organization. They brought fresh perspective from their experiences with other teams—Beattie, the Mariners and Expos; Mazzilli, the Yankees; Rodgers, the Reds; and DeMacio, the Braves, Indians, and Cubs. The third in command of baseball operations, Ed Kenney Jr., was once scouting director of the Red Sox. So there is a diversity of backgrounds and ideas at the top.

The resulting mix means that Weaver's ideas still carry some weight in the organization, but they are hardly the last word: other philosophies and scientific analyses are also part of the decision-making process. The Orioles have not switched to a four-man rotation, for example, but considering that Flanagan pitched in one for several seasons, such a move is not implausible. And the team has occasionally used the littleball tactics that Weaver disdained, like the sacrifice bunt and the hit and run.

Flanagan, whose baseball education includes stints as a starter, reliever, pitching coach, and broadcaster over the years, has been described as a big-picture thinker; Beattie arrived with an MBA and a detailed knowledge of baseball's business issues. Rodgers quickly made his mark as a disciplinarian by instituting a regimented minor-league system. Its uniform code of instruction across levels is reminiscent of the Oriole Way, but with some new rules emphasizing professionalism.

One of the lesser-known hires by the current front office is Dave Ritterpusch, director of baseball information systems, who was scouting director of the Orioles from 1973 to 1975, the heart of the Weaver era. Ritterpusch, who drafted Flanagan in 1973, has used psychological test results to grade the mental makeup of prospects, as John Eisenberg described in an article in the Sun this spring. The Orioles appear to be ahead of the curve in this area of player evaluation.

There may be times when Mazzilli or Beattie or Flanagan may ask, "What would Weaver do?," but for the most part the Orioles' decision-makers are their own men with opinions peculiar to their own personalities and experience. Apart from avoiding the stagnation of corporate monoculture, this motley group of minds has the potential to create a new legacy in Orioles history, one that reaches into the riches of the franchise's past, but also draws from the best policies from around the league and cooks up innovations of its own to propel the team toward a new run of success. And if that happens, Weaver, a loyal Oriole even in retirement, would be among the first to applaud it.

Other blog entries about the Orioles' bullpen:

Update (May 13): In my first draft of this entry, I was incorrect (or underinformed) about the Orioles going to a four-man rotation. I initially wrote that they had made "no suggestion" of implementing the idea. But after the Colorado Rockies decided to try the four-man this season, Jack Etkin of the Rocky Mountain News wrote an article on Monday about the issue that cites Flanagan willing to try four starters instead of five. His thoughts (emphasis added):

"I think that's something we will probably be able to do when we breed our own pitchers," Flanagan said, referring to a four-man rotation. "We certainly have talked about it extensively. If there's less supply and increased demand, why wouldn't you go to a four-man rotation?" ....

"I firmly believe it accelerates your progress as a pitcher," Flanagan said. "Because you pitch more, you don't have to be 100 percent to be successful. I think that's the thing that's harder in a five-man rotation, because you usually do feel better physically.

"When you pitch in a four-man, it takes away the overthrowing edge. And what gets through to you is you don't have to be overpowering to be successful. So you learn your craft faster. You learn how to get out of jams, even in ballgames when you're a little fatigued. One thing we would never do is say we were tired. That equated to performance. And I think today, what you see is players equate being tired with not being able to perform."

Flanagan is well qualified to speak on this topic, as he pitched in both four-man and five-man rotations. (As best I can tell, Beattie pitched only in five-man rotations.) A pitcher with an intimate knowledge of the rotation's effects is well equipped to counter the ingrained ways of baseball and reconsider this old idea. Two years ago, Rany Jazayerli of the Baseball Prospectus made a strong case for reviving the four-starter loop. (The argument spanned three articles: Part 1, Part 2, and Part 3.) What worked before does not always work today, but there's no question that the four-man rotation makes sense on a lot of levels. This is definitely something to monitor in the future.

Comments (1)


Whatever happened to our ex-pitcher, Don Boswell?


This page contains a single entry from the blog posted on May 10, 2004 4:18 PM.

The previous post in this blog was Rodrigo the meanest bull in the pen.

The next post in this blog is What's on second? (Or is that who?).

Many more can be found on the main index page or by looking through the archives.

Powered by
Movable Type 3.33