It's a scene that won't seem to leave our brains:
Oct. 16, 2003. Yankee Stadium. Eighth inning. ALCS, Game 7. Grady Little gives Pedro Martinez a go-get-'em pat on the arm, turns and heads back for the dugout.
It's more than six months later now. But in our minds, Grady Little is still out there. And in our minds, we still ask ourselves: What would we have done?
For managers and the pitchers they manage, this is the question: Take him out, or leave him in?
It is the question that hangs over every manager every game of every season. It is the question that leaves canyons under their eyes and fault lines in their forehead.
And for poor Grady Little, it was a question that helped add him to our national unemployment rolls.
Now obviously, it wasn't one decision in one game -- not even in a game that cataclysmic -- that got Grady Little fired. But it sure helped push him off a plank he'd already creeped way too far out on.
And you should never underestimate how much the ability to make just that sort of decision plays into a general manager's evaluation of how competent a manager is at his job.
"That's the biggest challenge a manager has," says Blue Jays GM J.P. Ricciardi. "No matter what your record is or what your payroll is, you'd better know how to handle a pitching staff, and you'd better know how to handle a bullpen. You've got to know when to take your pitcher out, when to get your bullpen up, when to put your bullpen in, because it's so magnified."
How magnified? All it took was one second-guess by Ricciardi of his manager, Carlos Tosca, on a talk show, and Tosca shot up the managers-in-trouble charts. All it took was one monumental decision that turned out wrong to get Grady Little booted out of his office. See a trend there anyplace?
You would think that trend would be a nightmare that would scare anyone who ever thought managing was a cool way to earn a paycheck.
But if you did, you'd be wrong.
"If it scared me, I'd do something else for a living," says Cardinals manager Tony La Russa. "That's just the reality of major league managing, and NFL coaching, and coaching in the NBA.
"I really believe that if Grady had gone to the bullpen and the bullpen had given it up, he probably would have taken even more crap. That's the reality of sport -- any sport. But especially our sport ... The great majority of the time, if the decision works, it was a great decision. If it doesn't work, you should have done something else. What he decided to do was bank on one of the great pitchers of our era."
The trouble is, the baseball bank never guarantees any return on your investment. So managers approach their defining question, "Take him out or leave him in?" knowing it's always a trick question.
But that's the breaks. They can't pass. They can't abstain. They can't punt. They can't run out the clock. They have to make that call, the best they can -- and then pray people aren't still talking about what a dopey decision they made 40 years later.
"I don't think anybody -- anybody -- has the right formula," says Marlins manager Jack McKeon. "It doesn't exist."
Ah, but here in the age of computers, videos, DVDs, assistant GMs in charge of massive calculations and numbers that stretch from here to Mars, people are actually working on that formula. We wish them luck. But all that information is adding yet one more layer to what was already the most controversial, most compelling, most second-guessable decision in sports.
So we've spent the last month asking managers, pitching coaches, GMs, assistant GMs, pitchers and even catchers to break down what goes into that decision. We found out -- and you'll find out -- it's even a more fascinating issue than we'd ever realized.
The information bank
Once upon a time, there was just the manager and his ever-reliable gut.
There were no printouts to show that Lou Gehrig was 22-for-46 lifetime against Firpo Marberry. You pitched who you pitched. You yanked him when you felt like it (probably about the 23rd inning). And that was that.
No one is too sure when that changed. But in 1968, a guy named Earl Weaver showed up in Baltimore with his trusty index cards, scribbled full of info on opposing teams. Nobody knew it then, but baseball's modern information age had just been born.
More than 30 years later, virtually every team is spitting out data that can tell managers more than most of them even want to know about their pitchers and the hitters they might be facing.
We surveyed three highly regarded information men for teams heavily into this research to find out what kind of brain food they were providing. Let's just say we've come a long way since Weaver's index cards.
There are basic pitcher-hitter matchup numbers and scouting reports. There are breakdowns on what opposing hitters hit against a pitcher as his pitch count progresses. There is a look at the recent effectiveness of each pitch in the pitcher's repertoire, and data on how each hitter has fared against different pitches in different locations.
There is info on how much rest the pitcher has had between recent starts and how much rest he'll get before his next start. There are breakdowns on how he pitches from the windup and the stretch, ahead in the count and behind.
And that's just the simple stuff.
If managers want to know even more, some clubs provide them video on demand that would allow them to dial up every duel the pitcher du jour has ever had with any hitter he might face that night.
"I've got all the numbers in front of me," says Boston manager Terry Francona. "But then I can also go to the video and watch every at-bat a guy has had against every pitcher. It's amazing."
Do managers even use this info
We've heard several front-office men grumble that they spend massive amounts of time, money and manpower cranking out all this sophisticated information -- yet their managers, they claim, often won't even use it.
"I don't think our coaching staff has been real receptive to this sort of data," says one exec. ... "Sometimes they go directly against the information," says another.
The big gripe is that all a lot of old-school managers really want is the stuff Earl Weaver used to scrawl on his cards: How has Joe Schmoe done against Bill Schmill?
Some managers might not even look at that. They might bypass all the specific stuff, says one executive, and "remove a right-hander who wasn't 'done' to use a left-handed reliever to face a left-handed batter." Whether the stats dictate it or not.
But that's just some teams. On a growing number of other clubs, the info plays a growing role in the judgment on when it's time to point the pitcher toward the shower knobs.
Phillies pitching coach Joe Kerrigan has spent more time looking at the numbers on his computer than half of Bill Gates' programmers. And he digests them all: "Third time through the order. What a pitcher allows the opposing hitters to hit as the game goes on. From 75-90 pitches, 90-105, 105-120. We have all those averages in our database. We have last year and this year, number of pitches thrown in his last three starts, number of days of rest over his last three starts."
And once those numbers are absorbed, they provide a context for what unfurls in front of everyone's eyeballs on the mound.
"The information gives you a little sound in the back of your mind going in," Kerrigan says. "If it's the third time through the order, if the average against him escalates when he gets from his 75th to 90th pitch, or 90-105, those are numbers that weren't put there by the pitching coach or the manager. Those numbers were put there by the pitcher."
But as information-oriented as Kerrigan is, he would no more use just numbers to tell him when a pitcher is done than he would keep his cruise control on to weave through a traffic jam. There are times to use the technology -- and times to turn it off.
"Even with all the background data going in," Kerrigan says, "you still only use that maybe 25 percent. And the other 75 is: How is the ball coming out of their hand? How's their delivery? How are their mechanics? How is the ball coming off the hitters' bats? If you didn't have any of that information, you still have to watch the game. It's the game that dictates the decisions."
And that's a sentiment echoed by even the most information-oriented managers -- even guys like La Russa, who was using matchup data 20 years ago and carries around a self-made chart telling him most recent pitch counts and days of rest for all his pitchers.
"For some people who believe that a lot more managerial decisions today are based on printouts and statistical analysis," La Russa laughs, "that ain't happening."
Are pitch counts overrated
On June 14, 1974, a guy named Nolan Ryan stepped to the mound in Anaheim to face the Boston Red Sox. Thirteen innings later, his manager, Bobby Winkles, told him he could call it a night -- after 235 pitches.
Now imagine what would happen if a pitcher threw 235 pitches these days. His agent would have a grievance filed before the relief pitcher was warmed up. "Outside the Lines" would do a 45-minute look at the evils of high pitch counts. Nine hundred talk-show hosts would send a thank-you note. And the manager would probably have to join the witness-protection program.
Heck, all that practically happens now if a pitcher even gets within 100 pitches of that. So it's fascinating what you hear when you toss that term, "pitch counts," out there.
"Jim Kaat pitched 250 innings for me as an 18-year-old in Missoula, Mont., and no one ever thought of pitch counts," McKeon says. "He later told me that was the best thing that ever happened to him. He could pitch a lot of innings and work on all his pitches, and he was in the big leagues the next year."
So it's no surprise that McKeon grumbles more than most managers about the babying of pitchers today. How, he wonders, is a pitcher ever supposed to learn how to go deep into a game if no one ever lets him find out he has it in him?
Good question. But while pitching coaches have been counting pitches with their dugout clickers for a quarter-century, there is more science being applied to the use of pitch counts now than ever before.
"With the youth of our pitching staff," says Indians assistant GM Neal Huntington, "we use pitch counts more as a protection device than an evaluative device or for proactive decision-making."
But because the Indians are one of the most information-oriented teams in the game, they have tried to explore every possible use of pitch counts. They have tried to pinpoint a pitch range where each pitcher begins to lose his effectiveness. And their coaching staff then applies that research in different ways.
There are times, for the sake of building confidence, they want their young pitchers to leave a game before trouble hits. So if they have a lead, they might be taken out before they reach the back end of their pitch-count range.
There are other times, Huntington says, when they'll be left in a game slightly beyond their normal range "to enhance their development" -- a concept that ought to make Jack McKeon happy. They won't be left in to reach what Huntington calls "abuse levels." But testing limits, in various ways, is a part of all athletes' development.
In some managers' offices, pitch counts are a subject more for paranoia than statistical analysis. But an executive of one team that studies this subject extensively says: "I think we expend more energy on preserving pitchers' health than we do second-guessing our manager."
Is 100 pitches really the magic number?
Do flashing lights go off in every dugout any time a starting pitcher reaches 100 pitches? You'd think so, given all the attention we seem to pay to it. But this just in:
All pitch counts are not created equal.
"If a guy is getting out of two or three jams in a game, if he's pitching out of the stretch all the time, there's an intensity to pitches you make when you've got a lot of baserunners on base," Kerrigan says. "How many more quality pitches do you have to make? Felipe Alou used to have a saying. He said every pitch a pitcher makes in those situations is almost like making two pitches with nobody on, because of the extra focus you've got to have, the wear and tear on your body and the intensity of those pitches."
So 100 pitches in a three-hit shutout are a lot different than 100 pitches in a five-inning nine-hitter. And 100 pitches in a 2-1 game can bear zero similarity to 100 pitches in a 9-2 game. In fact, 100 pitches for Kerry Wood may be the equivalent of 150 for Roger Clemens.
"If you have a veteran pitcher who may know what he's doing out there," says La Russa "he may throw 140 pitches -- but of the 140, he's only maxing out on 40. The other 100, he's taking a little off, putting a little on. But when the slop is flying, he'll reach back and make his best pitch.
"With younger guys, the reason there are so many injuries is that, with young guys, their answer when they start to struggle is to reach back. So of those 120 pitches, 40 are their best sliders and 80 are their best fastballs. So there's wear and tear on their arm. That's why you pay attention to pitch counts."
We know of several teams that are engaged in long, detailed studies to determine if there really is such a thing as a magic number. They're rapidly reaching the conclusion there isn't.
"Each pitcher," says an official of one club, "has to be taken individually -- 110 pitches is a lot for Smith but not a lot for Jones. So every analysis would have to only include the starts by the specific pitcher you're considering.
"That gives you maybe 33 starts a year to look at. But of course, not all those starts are created equal. (You have to factor in) weather, rest before the start, quality of competition -- and unlike a lot of things I look at, these extraneous variables don't all just even out over time. You'd also have to consider that a pitcher's effectiveness at high pitch counts might change year-to-year, maybe as that tiny tear in his labrum gets a bit less tiny.
"And sometimes the game dictates pulling a pitcher before his count gets high, because you have a fresh bullpen and want a matchup, or, in the National League, because you chose to pinch-hit. That limits your sample even further. So you end up with small samples of polluted data for most of your starters."
But is there a way to clean up that pollution? Some teams are trying to figure that out. In the meantime, managers apply their own standards.
Only Dusty Baker and Frank Robinson allowed their pitchers to go beyond 120 pitches in more than 20 games last season. Just three other managers -- Joe Torre, La Russa and Mike Hargrove -- pushed their pitchers beyond 120 a dozen times or more.
On the other end of that equation, Clint Hurdle (Rockies) never let a pitcher throw 120 pitches. Jimy Williams (Astros), Tony Pena (Royals), Eric Wedge (Indians), Lou Piniella (Devil Rays) and Mike Scioscia (Angels) did it only once.
So more and more, we're finding, it is actually 120 that has become the unofficial threshold -- even for an old-school guy like McKeon (who passed that line seven times).
"I don't believe in going (beyond) 120," says McKeon's pitching coach, Wayne Rosenthal. "But I'm not saying I'm never going to do it. It depends on the game. It depends on the guy. It depends on who's pitching in a given situation.
"I believe in this day and age, you have relievers for a reason. The complete game is not a big thing right now. If you have 11 men on your pitching staff, and you have a closer, and your starter has thrown 115-120 pitches, is that one extra inning going to help that guy?"
So if a pitcher was throwing a no-hitter in a blowout game and he had reached 120 pitches, would he and McKeon leave him in?
"I guess," Rosenthal laughs, "I'll say I'll make that decision when I come to it. ... But I'd rather him throw eight innings of no-hit ball and make his next start than go nine and miss his next three."
Should you ask the pitcher?
Game 4, 2001 World Series. Curt Schilling returns to the dugout after the seventh inning. He holds a 3-1 lead. He has thrown just 88 pitches. He has given up only three hits. His last pitch has just been clocked at 98 miles per hour. But he is also working on only three days' rest.
His manager approaches, to ask how he's feeling. Schilling tells Bob Brenly he's "gassed" -- but he "could pitch another inning." After weighing those mixed signals, Brenly tells Schilling, "That's enough." In marches Byung-Hyun Kim to attempt to get the last six outs. Cue the videotape.
This was a scene that sums up a manager's classic dilemma: When in doubt about whether to take your pitcher out, should you ask the pitcher himself?
"I've stopped asking pitchers how they feel," Brenly says now, "because most of them are going to lie to you anyway. That's the lesson I've learned -- in some cases the hard way. I've learned you have to trust your eyes, watch a pitcher's body language, watch how the hitters are reacting, rather than believe what he tells you.
"They all have something they do that lets you know they're getting near the end. They take more time between pitches. They ask for a new ball. They dig at the dirt. You start to recognize the signs -- and they're much more reliable than what they say."
La Russa was even more blunt: "You never ask a guy if he can continue," he says. "The general rule is, go out with your mind made up." In fact, La Russa says, he has had only one pitcher in the last two decades he trusted enough to ask -- Tom Seaver.
McKeon says he does ask -- but he demands honesty, because "we don't need no dead heroes." Kerrigan and Larry Bowa also ask. But they listen for more than just words.
"It's not the question," Kerrigan says. "It's the response. What kind of response? If you get, 'Heck yeah, I'm all right,' if you get one of those responses, you say, 'OK, go get 'em.' If you hear, 'I'm OK,' or, 'I don't know. I'd like to try it,' that's not the response you're looking for."
We've surveyed pitchers on this topic. Amazingly, they mostly agree with La Russa and Brenly. They'd rather stay out of this.
Mets reliever David Weathers, a former starter, says managers should make this call. If you ask pitchers, they "want to stay in the game -- whether they need to or not."
Josh Beckett, who became a World Series hero last October because of a Game 6 shutout he never came close to leaving, agrees. Asked what happens when managers ask, Beckett tells of watching Game 7 of the 1991 World Series, when Tom Kelly came to the mound to meet with Jack Morris.
"And Morris said, 'Get your butt back to the dugout. You don't have anyone better to put in this game than me,' " Beckett says, with a chuckle that tells you he can relate to that.
Even Schilling said last fall, the day after Little asked Pedro if he could get another hitter: "Don't ask me. Don't ever ask the pitcher. If you want to ask anybody, ask the catcher."
"Quite often, the manager will ask me what I think," says one of baseball's most astute catchers, Houston's Brad Ausmus. "And I'm always very honest. You have to be honest -- and let the manager decide after he has all the info. That's his job."
But it's a tougher job with some pitchers than others. Which brings us back to the pitcher we started with -- Curt Schilling.
Science or art?
After 150,000 baseball games over 129 big-league seasons, you'd think someone would have figured this out by now.
Take him out, or leave him in? Sure seems like a basic enough question. But after all those games and all those years, here's all we've really figured out:
There is no right answer.
"There are too many things you can't control," says Brenly. "You crunch the numbers. You use the scouting reports. You try to do the best you can. But I've come to believe that whatever is going to happen is out of our control. We're just the vehicles.
"I really think," he laughs, "that there's some big guy up there in the sky putting quarters in this thing, and guys are running around, and we have absolutely no control. It feels like that sometimes."
Do managers who program all the numbers into their brains make better decisions than the guys who just do what comes flying into their gut? Theoretically, they should. But then how do you explain Jack McKeon?
He refers to the computer chip in his heart as often as he listens to the hard drive in his brain. But he also just won a World Series doing it that way. So maybe the secret is that simple: Trust your gut.
"I do," McKeon says, as if there were ever any doubt. "Yeah, it's been wrong a few times -- not as much later in my career as it was early in my career. But I trust it. We do things a little different sometimes, but it's because I trust it."
His pitching coach reads the printouts, counts the pitches, offers guidance when asked. But he has seen so many unorthodox calls by his manager that somehow work, he knows when to defer to McKeon's trusty gut.
"To me, if Jack has a gut feeling, I'm running with it," Rosenthal admits. "Once in a while, he might say, 'Get the right-hander up,' and it may not be the right-hander I'm thinking. But I go with his judgment."
Nevertheless, we live in an age in which the information is so good and so sophisticated, a manager would be insane not to use it. And they all do, in different ways.
Francona understands he was hired in Boston as much for his ability to handle people as for his ability to use a laptop. But he has come to believe that you can't make the right calls on handling the people if you don't see what's inside the computer -- because "your eyes can deceive you," he says.
"If you see a bunch of numbers, that's the truth," he says. "If something has happened 99 times out of 100, there's a pretty good chance it's going to happen again."
But it's also a long, long season. And those are human beings playing it -- and managing it. So without some feel for the impact of every move on the people affected by it, a manager probably won't be managing very long.
La Russa, for instance, is convinced that just the body language a manager uses in making a move can make that decision more likely to succeed.
"I always thought Jim Leyland had a pretty good style," he says. "He'd come barging out of the dugout like he was saying, 'I'm on top of this, and this is the time to do it.' I think your team sees that and their team sees that. And that's the way to do it -- instead of coming out and appearing uncertain because you're walking slow."
La Russa says he even tries to think ahead about the way his entire team will take the results of a given move, or even a given loss. So as renowned as he was for bringing in his closer in Oakland, Dennis Eckersley, only in the ninth inning, there were certain games that called for waving for Eckersley in the eighth.
"Sometimes, you have to think, 'How's this club going to take it if we lose?' " La Russa says. "If we did it with Eck, at least guys are saying, 'Hey, we lost with Eck.' "
But that's not the same thing as wondering: How is the media going to take it if this move doesn't work? Or: How will the talk shows handle it?
"If you can look in the mirror and say that was the right move, you can sleep at night," La Russa says. "That's really the test. And what you understand is, because it's the nature of the game, the crap is gonna fly. You've got to be strong enough to take the heat and don't get too excited, because you've got to do it again tomorrow.
"Paul Richards used to say, 'Trust your gut. Don't cover your butt.' And I truly believe that."
The guy upstairs pumping the quarters in the machine will always let you know how it turns out. In the end, then, the manager has to do whatever it takes -- not to win, to survive.
So how ironic is it that the guy driving the World Series float down the Information Highway last fall was the man least influenced by the information and the man who has survived the longest -- Jack McKeon?
"It's simple," chuckles McKeon's pitching coach. "You try to make the right decision and hope it works out. Or you can be like Jack. You make a gut decision, and it always seems to work out."