I was searching around for ways to make an undergraduate business course more interesting," says Robert W. Keidel, a senior fellow at the University of Pennsylvania's Wharton Applied Research Center. "This was a few years ago, at Temple University, and years ago, at Temple University, and since there were a number of varsity athletes in the class, and since I'm something of a sports junky myself, I started to use local sports teams -- the Eagles, Phillies, and 76ers -- as concrete examples of what I was talking about." The questions he began to throw at his class seemed simple at first: How does planning and executing a game plan against the Washington Redskins resemble setting up a long-linked production line? Why woud a brokerage firm reward a key executive the same way the Phillies do Mike Schmidt, their $2-million-a-year cleanup hitter? The more Keidel explored the parallels, however, the more excited he got about applying the model in his own organizational consulting work.

Keidel's personal history contributed something to the scheme. He was raised in a family that owned Robert Wooler Co., a commercial heat-treating plant in eastern Pennsylvania ("classic basketball organization," he says), and he took degrees froim Williams College and The Wharton School before spending three years in the Navy ("classic football"); later came a PhD dissertation on the Jamestown (N.Y.) area Labor-Management Committee, and the Temple faculty appointment ("classic baseball"). As a consultant and "team-builder," he had used other types of metaphors before, but, he says, "the problem [with analogies] is what do you do after the initial uh-huh? I mean, it's cute and it frees people up, but it really doesn't take you very far. With sports and business, you go back and forth so much it's almost impossible to tell where one ends and the other begins."

Perhaps the most unusual aspect of Keidel's work is that in using the games of football, baseball, and basketball as the metaphorical lenses through which he looks at business organizations, Keidel notices properties of the lenses themselves that most casual fans overlook. Basic elements such as team structure, decision making, power flow, and reward systems are cast in surprisingly sharp relief.And while the modern era of players' strikes, franchise shifts, and corporate Super Bowls has inspired much commentary on how businesslike the realm of sports has become, little has been said about what makes the different games themselves distinctive. Keidel says a lot.

In fact, he has packed his observations into a highly readable and insightful book, Game Plans: Sports Strategies for Business, which was published in September by E. P. Dutton. Last summer, INC. sent senior writer Joseph P. Kahn to interview Keidel in western Massachusetts. Kahn's report follows:

We met on the campus of Mount Holyoke College, where Keidel was conducting College, where Keidel was conducting a seminar in team building for a large manufacturing company. When I found him, he was playing with some charts he'd drawn up to illustrate his vision of the conceptual relationship among football, baseball, and basketball organizations. I happen to like sports a whole lot more than I like charts, so I asked him which game had most engaged him in his younger playing days. "Baseball, I guess," he answered with a smile, "although my baseball career peaked when I was 15. I played some sandlot football, and I got cut from my high school basketball team. Now I mostly play golf." Keidel also admitted that as a fan, his internal calendar gets most confused in October, when the three major sports seasons all run together. Given that viewpoint, and the fact that baseball's All-Star Game was on that night, it was not surprising that our conversation ran up-court or down-field as often as it roamed around the boardroom or factory floor. Like any good coach, moreover, Keidel began with the fundamentals.

INC.: Your basic premise is that baseball, football, and basketball represent three different basic models of organization, is that it?

KEIDEL: Right. The core difference among them is how they achieve co-ordination. In baseball, it's in the design of the sport. A baseball manager has very little to do other than to get the right players into the game at the right time, in the right batting order and location on the field. The players themselves act independently, although in a coordinated manner. Baseball is also very situational in terms of the teamwork required. Typically, on any given play, you get two or three players interacting for a brief period of time. The guts of the game is all individualistic: throwing the ball 95 miles per hour, catching it, hitting it.

Football is very different. Football relies on coordination from above: It's an organization meant for top-down direction by the head coach, with each player having a very narrow piece of the game plan. Basketball is still another game. Coordination is achieved through spontaneous interaction and mutual adjustment by the players themselves, with the coach acting as the catalyst.

My argument at its simplest is that football represents control and baseball represents autonomy, while basketball represents voluntary cooperation.

INC.: Those are loaded words: "control," "autonomy," "voluntary." When you use these particular analogies in your consulting work, do you run into trouble with people who have certain prejudices about the sports themselves?

KEIDEL: Oh sure. I think there's a knee-jerk resistance in terms of the depth of the models. "Oh yeah," people say, "Vince Lombardi and so on," referring to the legendary Green Bay Packers coach. They see sports as a facile area, with no substantive parallels worth exploring. Recently, I used basketball and football models with a group of hourly workers, to contrast where they and their managers had to go in terms of promoting flexibility of production. Personally, these guys identified with professional football and thought my analysis dumped on their favorite sport. Well, in a sense it did, but what I was really saying is that football management may be fine for football-type organizations, but it just isn't right for many businesses -- and certainly not if you're trying to promote flexibility. I like football. But I'm using sports as a way of looking at style, as the preface to team-building. The focus should be: What kind of game are we playing? What kind of structure do we need to play the game? How do we behave and how do we manage?

INC.: You don't make football sound like a great game to be played on any field, though. At one point you quote writer Pete Dexter as saying that football "condenses most of the things you hate to see coming in this life into a very short period of time: decision, pain, fear, embarrassment, confrontation, spit hanging from the bars of your face mask." That's not a hell of an attractive list.

KEIDEL: Again, we're talking about priorities. In business, football is a great model for high-volume manufacturing. It is also a wonderful model for consistency and coherence: McDonald's, Holiday Inns, whatever. And where the risk of failure is extremely high -- airlines, for instance -- it is absolutely critical. In fact, football works for any company that requires a common safety or quality policy. However, I'm suggesting that this isn't the leading edge of our economy. There are big negatives -- flexibility, adaptiveness, how the individual is treated, innovation. In the National Football League, of course, some teams are more innovative than others.

INC.: Like the Oakland Raiders, I suppose. They seem to thrive on misfits, other teams' castoffs, game plans drawn in the dirt.

KEIDEL: Right. Or [Coach] Bill Walsh and the San Francisco 49ers. They're very flexible and innovative. Still, it's the nature of football that requires the coach to be in control. [Ex-Minnesota Viking] Fran Tarkenton's book about his career as an NFL quarterback talks about player input in the offensive huddle. What he's really talking about is one of his receivers having the option of going left or going right on a given pass route. Now, that's not a whole lot of choice. Also, the fact that Tarkenton called his own plays would put him in the distinct minority of pro quarterbacks today. Most plays are sent in from the sidelines.

Only in football do you have a head coach, which may suggest that the rest of the team is just bodies. But in a world that's becoming less predictable and more turnbulent, game plans that are constructed and laid down from the top decay too quickly to be all that workable. That's the real problem with football approach. It's literally a World War II model.

INC.: I hope you're not crusading for the abolition of NFL football as America has come to know and love it.

KEIDEL: Absolutely not. I'm struck, though, by the extent to which football is idiosyncratic to North America. I've begun to think there's some tie-in with football's appeal and the prevalence of mass-marketing production in this country. No other country in the world has gone farther in that direction than we have. There may be a whole cluster of reasons to support why that's so, but at this point, all I can do is point to the parallels. Basketball has become the most international sport in the world, played in more countries than any other. Baseball is also popular in many cultures. Football isn't, and I don't think it ever will be. In fact, if anything, I think we will see football's appeal go the other way.

INC.: And yet NFL football remains our preeminent armchair sport. Are you suggesting there's something deep in the American psyche that craves the violence and voyeurism of televised football?

KEIDEL: I think so. Or at least deep in the American organizational experience, which is extremely controlled and asymmetrical from the top down. This country has taken mass production to new heights -- or depths, depending on your perspective. That has to bespeak a certain organizational mind-set.

INC.: Besides player coordination and cooperation, what's another fundamental difference in how these games are played?

KEIDEL: Well, scoring patterns, for one. In football, scoring is sequential. Teams may go for the Big Play, but effective offenses are basically built around sustained derives and ball control. Time of possession becomes an important statistic, because once you score, the other team gets the ball. The longer you keep it, the less time the other offense has to operate. You give up the ball in basketball, too, but it pro basketball, anyway, scoring is continuous, or nearly so.In the NBA, a shot is put up on the average of every 15 seconds. Teams can make up large deficits in a hurry. Only in baseball can a team continue to score without having to play defense. And only a baseball can a single player account for more than one "score" -- i.e., run -- at a time. What this means is that the old "three yards and a cloud of dust" football approach is out of sync with a game like baseball.[Statistician] Bill James has shown that teams with the highest regular-season batting average usually lose the World Series. How come? They need too many hits to score runs. They're not geared to the big inning, and big innings win baseball games.

INC.: Why these three sports? Why not use ice hockey, or soccer, or any other team sport familiar to most of us?

KEIDEL: Initially, I was goping to cover hockey, soccer, and possibly volleyball as well. I identified more with the other three, though, because I had played them as a kid. That, plus the fact that I think they really represent generic types. Soccer is rather basketball-like: Players coordinate themselves, they dribble the ball (with their feet, not their hands), the action is continuous, etc.Hockey, as others have pointed out, has elements of both basketball and football. The winningest teams of a few years ago emphasized dumping the puck in the corner and hitting people, in a way similar to control football. But today's best teams -- the Edmonton Oilers, for example -- play more in the European style, where the accent is on passing and speed. I like to think the same transition is underway in American business.

INC.: You write that baseball has "a fundamentally uncontrollable character" and that it is "partly random in nature." What are you implying by that?

KEIDEL: In baseball, much as in nature, the defense puts the ball in play. The good pitchers call their own pitches, with their catchers' help. The manager doesn't stand in the dugout telling [Houston Astros'] Nolan Ryan or [New York Mets'] Dwight Gooden when to throw a curveball. Each player is fundamentally on his own. All of the guidance and teamwork in the world won't help you hit a ball, so there's not a whole lot a manager can do. Earl Weaver of the Baltimore Orioles, whom I consider to be the consummate baseball manager, says it all when he says that baseball comes down to pitching, fundamentals, and three-run homers. For a manager, it's player evaluation and player-situation matching. If you intervene too much, it's like supervising an artist or inventor. You risk robbing the organization of the very skills it hopes to compete with.

INC.: On the other hand, you can also, like George Steinbrenner's New York Yankees or Gene Autry's California Angels, build yourself a pretty good team by hiring a whole bunch of high-priced free agents. You don't need to grow your own the way baseball organizations used to have to by developing strong farm systems.

KEIDEL: True. It's still a team game, where chemistry and leadership count enormously, but baseball teams can afford a lot more individualists, a lot more Reggie Jackson-types. It's the same in my professional realm, academia. If you want a top-notch faculty, you hire individual superstars. You don't make team players your first priority, because mediocre scholars will make you at best mediore. Chemistry among the players is important, but I like what [Chicago Cubs manager] Jim Frey says about that. Frey defines chemistry as "a double with the bases loaded."

INC.: Baseball is also a sport that lends itself to heavy statistical analysis. You even see computers in the managers' offices now, printing out data on every situational matchup imaginable. Individual performance is quantitatively measurable in ways that don't work as neatly for football or basketball.

KEIDEL: Database-ball, I call it. Yeah, that's true. In football, computers are mostly used to compile and evaluate scouting reports. I remember talking to Carl Peterson, who was director of personnel of the Eagles back when they went to the Super Bowl. Peterson said the good teams basically put it all up front. They design really good game plans, and they don't vary from them. All the situational decisions in a game have been planned in advance. With baseball, the same decisions become real-time situational. Pitchers versus hitters, one on one. When Weaver was asked about computers, he said, hell, all a computer does is what I do in my head -- the implication being that some damn computer will make a rival manager as smart as he is. Which may be true.

INC.: So now it's pitching, three-run homers, and software, eh?

KEIDEL: Good pitching still beats good software. But in terms of individual performance, think about the all-star concept in baseball and how different it is. Put together a team of Rickey Hendersons, Gary Carters, Dale Murphys, et al, play them against the rest of the league, and they'll win maybe 95% of the time. That won't happen in the other sports. The NFL all-stars would have a hard time beating the NFL champion. Ditto for the National Basketball Association all-stars. With rare exceptions, baseball teams are truly modular organizations. To a large degree, the same is true for universities, sales-oriented operations, and conglomerates.

INC.: Basketball, on the other hand, represents your working model for companies where organization has to be flat, flexible, and more responsive. Is that a fair assessment?

KEIDEL: The easiest way to see what basketball is, organizationally, is to contrast is to the classic football-type bureaucracy. In that kind of operation, research and developing designs a product, passes it along to manufacturing to make, which then passes it along to marketing to sell. None of the divisions care too much about the others. The analog is the platoon system in football. According to a lot of ex-football players I know, the primary identification on a football team occurs within the platoon unit. Players identify with offence, defense, or special teams more than they do with being a Steeler or Eagle. And that's a necessary thing. You let the offense and defense beat the crap out of each other in practice during the week so they'll be ready for Sunday's game.

With basketball, in contrast, you get much more interdisciplinary cooperation, more lateral versus hierarchical flow. Players switch from offense to defense instantaneously. They all have to be passers and scorers, they all have to play D. Teams practice a few set plays, but what happens on the court is entirely dictated by the flow of the game.

INC.: And that makes it, in your words, a game that "represents where we're going more than where we have been"?

KEIDEL: Basketball doesn't have the popularity that baseball and football have, which may be because we have not yet lived basketball the way we have lived other sports. Basketball is postindustrial. In meaning as well as chronology, it is newer than football, just as football is newer than baseball. Basketball is flow, process, spontaneity, chemistry. Although some of these words are current business buzzwords, their deeper meaning has only begun to penetrate the organizational psyche. Basketball signifies a different form of teamwork from the other sports.

INC.: Why isn't basketball more popular with the fans, then? College teams inspire wild fanaticism among their followers, yet pro ball, where the level of pure talent is so much greater, has been a very tough sell to a national audience.

KEIDEL: I like basketball because I appreciate the athletes, particularly on a team that plays as a real team. Great passing teams like the Boston Celtics are thrilling to watch, they really are. They hustle, they interact, they move without the ball, they think out there. Still, I believe that a number of structural changes will have to be made for basketball itself to become more popular. The players are so big now, so quick and talented, that they've outgrown the [original] dimensions of the sport. As a lot of writers have pointed out, baseball is still a wonderful design: 60 feet, six inches from the mound to the plate, 90 feet between bases. It works. With basketball, however, the guys playing today are way too good. They need to raise the basket, perhaps, or make some such basic dimensional change. Also, they need some changes that would make individual games more meaningful -- the guts of the game, I mean, not just those silly last two minutes, when there is a time-out every time the ball comes up the floor. Getting people to watch right now is a problem, because on a gross level, the whole [NBA] season is irrelevant.

INC.: You make that point with some interesting math. NFL football teams used to have 49 roster players per franchise; now they have 45. That's almost exactly twice the size of a professional baseball team (25), which is almost exactly twice the size of an NBA basketball team (12). Football teams play 16 regular season games, which is about one fifth of an NBA season (82), which is half a major-league baseball season (162). When you consider just the numbers, never mind the particular players involved or the nature of the game itself, they suggest radically different strategies for managing the individual team member.

KEIDEL: Sure. A single, regular-season NFL game has, at the very least, five times the significance of an NBA game. Each confrontation is a big-risk situation, a miniwar. In football, you get in the play-offs and it's sudden death: one loss, you're out. In baseball or basketball, you have multiple-game play-off series. Also in baseball and basketball, players who come out of the game -- in basketball, I'm talking about after a player's sixth personal foul -- can't go back in; in football, you can substitute freely after every play. What you wind up with is an interchangeable, spare-parts type of approach. And that's what you need, given the importance of a single game. The only way you can fight that kind of war is with an authoritarian structure and plenty of fresh troops.

INC.: Because the nature of each game is so different, I'd guess that you believe that a great coach in one sport might not be such a great coach in another?

KEIDEL: I'm not sure. There is a lot of management literature out suggesting that there is no such thing as a manager who is able to manage in any situation. Too many situational factors are critical. And I think that is probably true. I know sports people I've talked with feel that managers are very different -- that what might make one a particularly good coach in one game might actually work against him in another. To be good football coach, for instance, you have to be a super-organizer, a super-controller. That would frustrate you in baseball, maybe even more so in basketball. I have real concerns whether a highly technical coach like [New York Knicks] Hubie Brown can be truly successful in basketball. A coach like that might do extremely well taking a mediocre team to the play-offs -- but not taking a play-off-caliber team to the championship.

INC.: You're saying that you can't get too technical coaching a flow-and-process game like basketball?

KEIDEL: Jack Ramsay, the Portland TrailBlazers' coach, is a brilliant basketball thinker. He probably knows as much about court strategy as anyone; his team won an NBA title in 1977. What has happened since then? Leaving aside the injury to Bill Walton, their best player, they have gone downhill. They basically let go of all their stars, even passed up [1985 NBA Rookie of the Year] Michael Jordan. I wonder how far that team will ever go with Ramsay's approach. In football, you can go as far as your technical ability takes you, whether you're hot like Lombardi was or cool like [Dallas Cowboy head coach] Tom Landry. Incidentally, my impression of someone like [California Angels manager] Gene Mauch is that he's basically in the wrong game. He should be coaching football, not baseball.

INC.: I assume you mean because Mauch is famous for being such a strategist, for trying to outguess the opposing manager in every conceivable game situation.

KEIDEL: Exactly. Obsessing over silly decisions like, Do I have my guy bunt down the third baseline or first baseline? What Earl Weaver does is to arrange for someone to come off the bench and hit a three-run homer. And he's right. [University of Indiana's] Bobby Knight is another [misplaced coach]. He runs his [basketball] team just like a football coach would.

INC.: Some successful coaches are notorious for inspiring fear or hatred in their players. When the team wins, the locker room is full of happy talk about being "one big happy family." Does winning -- whether you're talking about winning a pennant race or a market share -- tend to obscure the structural weaknesses of an organization?

KEIDEL: I think so. The notion that success causes teamwork is at least as valid as its opposite. My consulting field is organizational change, and frankly, the only organizations that ever change are those that are hurting. Hurting implies not winning, and when you're winning you put up with a lot. That's one reason why winning sows the seeds of losing downstream.

INC.: One factor that has forever changed the way we look at professional sports is salary structure. How crucial to your overall analysis is the compensation issue?

KEIDEL: It's critical. If there is one variable you want to get right in terms of playing a particular game, it's the reward system. Baseball companies are individualistic to the extent that a salesman can earn more than his boss does: You want to encourage the individual to perform as well as he can, and reward him on that basis. Like baseball, one individual's game doesn't come at the expense of another's. A team can have several 30 home-run hitters, several 20-game winners on its pitching staff, and so on. Few baseball teams have too much hitting or too much pitching. So a reward system that is individual and performance based -- not just "show up and we pay you" -- makes a lot of sense.

INC.: What about football? Relatively speaking, the guys in the trenches don't make much.

KEIDEL: Again, football is very hierarchical.The players who make the bucks are the ones who handle the ball: quarterbacks, then running backs, then receivers. In baseball, high salaries are distributed over all of the positions. Football coaches also tend to make significantly more than their counterparts in the other sports, which makes sense, because they have certain intellectual demands made on them that the others don't. To the extent that in business we have got to get away from the old sort of pyramid design and encourage lateral interaction we have got to look at systems that reward cooperation between different functions, as well as performance within functions.

INC.: And what if you're running a basketball organization?

KEIDEL: If you want to play basketball, reward team behavior. In any matrix organization, a key issue is what percentage of a person's reward system -- pay increases, promotions, whatever -- is determined by team performance. If it is 90% individual and 10% team, you won't have a team. But it's hard to design reward systems that pay for team wins, not for individuals scoring 30 points a game.

INC.: In team situations like that, it often seems that an individual's contributions aren't accurately measured until he or she has left the organization and someone else takes his spot. The player may have more ability than the one he replaced, but the right mix isn't there.

KEIDEL: I'm sure there are lots of examples like that. Part of it is due to basketball's requirements for intimate teamwork and cooperation, which make interpersonal demands that don't exist in football or baseball. I mean, it's nice to have people cooperating and respecting each other and so on, but you can get away without those things in baseball or football organizations in ways you can't in basketball. Red Auerbach, the old Boston Celtics coach and general manager, is probably the best example of what I'm talking about. Auerbach could certainly resort to authority when he had to -- nobody ever accused him of being a push-over -- but that's not how he fundamentally ran his club. He was very sensitive to interpersonal dynamics, maybe the best ever. A tremendous motivator. The way he managed Bill Russell, particularly when Russell first came into the NBA, was absolutely brilliant.

INC.: Russell eventually became one of those rare commodities, the professional player-coach. Pete Rose has assumed the same role with the Cincinnati Reds. You never see that in football, though.

KEIDEL: No, nor would it make any sense to try it. Think about the locations of the managers/coaches in each game. In baseball, the manager is literally below ground, watching the field action from the dugout. Paul Owens, who used to manage the Philadelphia Phillies, once said that the toughest thing about managing a baseball team is standing up for all nine innings. Football is just the opposite. There, you have got the coach not only standing up but, in a sense, high above the team, since he's got spotters sitting in the stands to help him call formations. The ultimate symbol of that kind of hierarchical overview was Bear Bryant's tower at the University of Alabama. He stood up there, aloof and all-seeing, terrifying everybody. In basketball, though, the coach sits right there next to the action and communicates with his players -- he even touches them -- as they move up and down the court. There is a physical mirroring of the manager's role in all three sports.

INC.: Carrying that over to business, then, you'd expect to find management literally towering above the troops in a football-type organization.

KEIDEL: Often you will have an executive row that is sequentially and hierarchically arranged by virtue of level, which is an office sequence that mirrors the factory production sequence. This is classic football. In a baseball organization, the status system is apt to be more horizontal: The top broker, the top professor, the top writer gets the preferred office. It's hierarchy within class, not by level on the chart.

INC.: You're talking about a corner office with a window, not a suite of rooms six floors up in the tower.

KEIDEL: Right. And the basketball pattern is one that minimizes separation -- maybe offices are clustered around a meeting area, which encourages people to interact. I've read some interesting articles lately on office design. It's amazing to see how physical configurations reinforce value systems and, in my terms anyway, different sports models.

INC.: So if we see Apple Computer building a 40-story office tower in Cupertino, we should assume that they are changing their game plan?

KEIDEL: I would say that if that happens, they're into a whole different fruit.

INC.: Have you ever used videotapes of sports contests as instructional tools? You certainly seem to be arguing that we get a lot of our subliminal messages about authority patterns and decision making from sports.

KEIDEL: I haven't done that yet, but I might. I certainly encourage people I work with to look for these patterns whenever they turn on a game or go to the ballpark. Remember the 1984 NBA finals, for instance, when [Celtics coach] K. C. Jones allowed his two guards, Dennis Johnson and Gerald Henderson, to make their own decision about which one was guarding [Lakers'] Magic Johnson at any given moment? Now, contrast that with, say, Tom Landry deciding how many defensive backs he wants on the field and who they'll be covering. Or take a baseball game where the manager walks out to the mound and says to his pitcher, "Look, you wanna pitch to this guy or not? Wanna walk him? Want me to bring in a reliever?" The pitcher's really making the call. You don't see that in football.

INC.: And we seem to get a lot of subliminal messages from televised football.

KEIDEL: John Brodie, the ex-49er, made a commercial many years ago, and the tag line was, "I am the quarterback." I could see thousands of managers watching that and walking away, thinking to themselves, I am the quarterback. Therefore, I am the locus of authority." It was a message entirely consistent with the game, but it may not have done much for them as managers.

INC.: Speaking of applying lessons from sport, the Japanese borrowed baseball from us and turned it into their own national pastime. We, on the other hand, have been looking to them for inspiration in the fields of productivity and management technique. Just who's teaching whom here?

KEIDEL: The most interesting aspect of Japanese baseball is how they've tried to render it very grouplike. If you think about it, this makes sense, because Japanese organizations tend to be very basketball-like, yet they operate within a distinctly hierarchical -- or football-like -- culture. Their organizations are simply not as individualistic as ours are. Neither are their baseball teams.

INC.: It's also interesting that although plenty of ex-U.S. major league baseball players are encouraged to play over there, none is promoted as a true superstar. If an American player starts leading his team in home runs, for instance, he's either platooned or called upon to bunt more often.

KEIDEL: That's true. Reggie Smith, who starred with several American teams, wrote an article about his playing experiences over there, in which he referred to Japanese baseball as "small baseball." That was meant as a putdown pun to some degree, but he also meant it in terms of Mauch-type baseball, not Weaver baseball. In Japan, every man in the lineup is called upon to sacrifice for the one-run situation. The real concern I have with the whole Japanese thing is that their emphasis on group process has to be balanced against our emphasis on individualism. We're not going to change our fundamental nature, and I'm not sure we need to.

INC.: As a consultant, you obviously influence the direction of the teams you work with. What kind of game do you see yourself playing with them?

KEIDEL: My ground rule is that I work with groups that fall somewhere between a basketball team and a football team, 5 to 11 people. As a teacher, I've handled classes ranging in size from 7 students to 40. With 7, you can play basketball. With 40, you damn well better play football. In discussing companies [in the book], the difficulty I had with the basketball model is this whole size thing. I mean, you can only have this sort of reciprocal back-and-forth interaction up to a point. As your organization gets bigger, what can you do? Maybe you put in hierarchical controls, knowing that you risk losing a certain degree of spontaneity. Or maybe you develop a baseball-like design that encourages basketball within units and autonomy among them. Those are tough calls for any organization to make.

INC.: What's your time frame for working out these teamwork issues? Given that the sports models are so familiar to so many, does that speed up the discovery process?

KEIDEL: Well, the time issue gets integrated into a lot of other factors. But to go through my basic laundry list is about a six-day exercise, not including orientation and working downstream. There's no such thing as team-building in discrete doses. You can't subject a team to certain treatments and then just walk away. After all, you can call yourself the Boston Celtics, you can act just like the Celtics -- for a week or two, maybe -- but that doesn't make you the Celtics. In any kind of game, it's a long, long season.