[The players] play for men who "have come into the business for no other motive than to exploit it for every dollar in sight."
-"The Brotherhood Takes the Great Players," Sporting Life, November 13, 1889.

The NL "has no apology to make for its existence or for its untarnished record of fourteen years" because it "rescued the National Game from destruction threatened by the dishonesty and dissipation of the players." The owners argued that they gained little in serving as the ethical wardens of baseball, while, thanks to the reserve rule, the "salaries of players have more then trebled" since the National League's inception. The owners asked the public to reject the "overpaid players" who wanted to "again control [baseball] for their own aggrandizement, but to [the game's] ultimate dishonor and disintegration."
-Henry Chadwick, "The League Defines Its Position in the Controversy with the Brotherhood," New York Clipper, November 30, 1889

"[MLB Commish Bud Selig] should also have set up a management office to teach some of these teams how to run their baseball businesses, because while the playing field is unlevel and there are gross inequities inherent to the current system, the fact is that the Brewers, Tigers, Royals, Devil Rays and other teams are what they are for a reason -- bad management. The A's, Reds, Astros and Giants compete every year because of good management. The Red Sox are paying $110 million for a $70 million team because of previous bad management."
-Peter Gammons, ESPN.com, "Players, owners just don't get It", August 17, 2002.

Team Salaries

Here is a chart of team salaries, by sport, for the 2001 season (for MLB) or the 2001-02 season (for the NFL, NBA, and NHL). Linear trend-lines have been included. Each marked point represents a single team's salary within the league.

There are subtle differences between the capped and non-capped leagues, based on this chart. All four leagues have a steep drop-off after the top team or two and before the last team or two. Only the capped leagues have a virtually flat linear progression between the two extremes, however. The trend lines for the NFL and NBA are shallower than the trend lines for MLB and the NHL and, in fact, virtually parallel each other. Baseball and hockey have a "stepped" downward progression that is much steeper than the NFL/NBA. This is dramatically evident in the MLB line where every 3-6 teams there is a steep drop to the next cluster, a trend less dramatic but equally visible for the NHL.

Figure 1 illustrated that sports with some form of cap seem to have a smoother distribution of wealth, and (in numerical terms) a lesser disparity between the rich and the poor. Figure 2 shows this more clearly. It shows the single highest team payroll, the average of the top five teams, the overall league average, the average of the bottom five teams and the single lowest team payroll by league.

It is interesting to note that all four leagues are "top-heavy": the difference between the top five and the league average exceeds the difference between the bottom five and league average. The following chart shows the difference between the average of the top five team salaries in each league (the second set of bars in Figure 2) and the average of all teams (the third set of bars in Fig. 2) in both numerical and percentage terms:

The most highly paid teams in the capped leagues are much closer to the average team salary than they are in the un-capped leagues where the difference is greater than half the average team salary. Figure 4 shows the average of the 5 lowest team salaries subtracted from the average team salary in the league:

The bottom 5 in each league are actually closer to the average salary than the top 5 are, regardless of whether the league has a cap or not. This seems to indicate that a small number of "haves" in each league are standing out from a much larger number of "have-nots" or "have-lesses".

Competitive Balance

In the real world a cap, soft or hard, does serve to keep the teams closer together in terms of payroll expenditures. The intent of a cap, though, is not just to restrain labor costs. Increased competition is the other key element. People in favor of salary caps and other restraints on the cost of labor argue that they will increase the competitive balance around the league, allowing small market teams to compete on equal footing with large market teams. We'll look at this argument from two angles. First is how many different teams make the Championship game/series for each league over time. More teams making the finals means more competitive balance, fewer means less. The second angle is to examine the standard deviation of the winning percentages in a league over time. Less disparity in winning percentages means more competitive balance (a four team league whose teams finish 2-4, 3-3, 3-3, 4-2 is more competitive than one whose teams finish 0-6, 3-3, 3-3, 6-0). By looking at how these two indicators vary before and after the introduction of a cap in the NFL and NBA, and by comparing those two with the NHL and MLB, we can easily determine the effectiveness of caps at creating competitive balance in the real world.

We'll start with the hard-capped NFL and look at how many different teams won the Super Bowl. It's been 8 full seasons since the cap was instituted so the data is broken down into 8-year segments (Cap-era in green):

The NFL, pre-cap, seemed to be somewhat cyclical. Dynasties arose and dominated for several years before collapsing. Their reigns were followed by a stretch where anyone seemed to have a chance, but then certain teams took over the league again (noticeably not the same ones as dominated earlier, for the most part). Since the onset of the cap there has been a greater variety of Super Bowl finalists and a wider range have won. No single team has dominated the league for long, even the supposedly mighty St. Louis Rams of the last 3 years. Only three of the last 18 Super Bowls have been won (in 4 tries) by teams from Chicago, LA, Houston or New York City (the 4 largest markets in the US). Of course, Houston and LA haven't had teams for most of that stretch. Based on Super Bowl finalists the NFL is seeing much more variety since the institution of the cap.

Our second indicator is the standard deviation among winning percentages in the NFL by season. Larger numbers indicate that there is less balance because the winning percentages vary more greatly. Based on the Super Bowl numbers we can expect to see that the average of the standard deviation of the 8 seasons since the cap was instituted is less than the average of the standard deviations of the 8 years immediately preceding it.

As Figure 6 shows, the numbers contradict our expectations. While the cap era numbers are closer together than the pre-cap era numbers, the difference is by 0.0218%, the equivalent of 2 pennies relative to a $100 bill. More different teams are making the Super Bowl, but the overall competitiveness of the league has not changed in the years since the cap was implemented. The years with the second and fourth highest standard deviations were under the cap, while the years with the second and fourth smallest standard deviations were pre-cap. The NFL's cap is improving parity at the very top of the league, it seems, but has done little if anything for league-wide parity statistically.

In the soft-cap NBA the situation is very different. As you recall, the cap was implemented in full force prior to the 1984-85 season. It's been 18 full seasons since then, so we've broken the NBA history into 9-year segments:

In only one 9-year segment prior to the implementation of the cap (the 9 years featuring the Celtic dynasty of the 50's and 60's) did fewer teams win the NBA championship than have in the two segments of the cap era. More different teams have made the finals, it is true, but since the cap came into being only one NBA champion has failed to defend their crown at least once. Even more interesting is that the cap was implemented to protect smaller market teams and increase their competitiveness. The 4 largest markets in the United States are New York, Los Angeles, Chicago and Houston. In the 18 years since the cap was brought into effect, the Chicago Bulls, LA Lakers and Houston Rockets have won 14 championships (Chicago 6, LA 6 and Houston 2). Teams from New York, LA and Houston have also lost in the finals 6 times (the Knicks and Lakers twice, the Nets once and the Rockets once) for a total of 20 appearances. In the 18 years prior to the cap, the 4 largest markets only won 5 titles in 13 appearances and the only repeat champions were the smaller market Boston Celtics of 67/68 and 68/69.

As we saw with the NFL, the championships are not a perfect measurement of overall balance, so let's look at the standard deviations for the NBA. It's been 18 years since the inception of the NBA cap, so we're using 18 year segments, not 8.

The NFL had a difference between the two of roughly .02% in favor of the cap. In the NBA the difference is just over 2.0%, and shows that the pre-cap era actually had more competitive balance. Where the NFL shows no change in competitive balance league wide and a measurable improvement in top-end competitiveness, the NBA shows a significant decrease in competitiveness both at the top and across the board. Based on these numbers the soft cap is a resounding failure at promoting parity.

We've looked at the hard-cap league (doesn't promote league-wide parity) and a soft-cap league (apparently not effective at improving competitive balance and may actually restrain it). Now let's look at the non-capped leagues to see how they measure up. First up is baseball, also broken down into 9 year segments to make comparisons easier (don't forget that the 1904 and 1994 World Series were cancelled).

Based on this data, baseball is and always has been more competitive than basketball (with or without the cap) if we look at the number of teams making the Finals and winning the World Series. It is also almost as competitive the post cap NFL at times. Despite all the uproar about the Yankees and "buying the World Series", teams from the big 4 markets have won only 6 Championships in 8 appearances in the last 18 years, fewer than basketball although still twice as many as in the NFL.

Looking at league-wide numbers also shows interesting numbers. Here are the average of the standard deviations since the 1995 season (the inception of the most recent salary cap: the NFL's)

Despite the popular outcry against the unfairness of baseball, it is significantly more competitive league wide than basketball or football (or hockey, as we'll see in a minute). MLB also has good variety in the number of different finalists in the World Series. Despite the lack of a cap and only limited revenue sharing, Major League Baseball delivers excellent competitive balance. The sheer number of games (162) and length of the season (7 months), combined with the very small number of teams to make the playoffs (8 total, compared with 16 in hockey and basketball, and 12 in the NFL) magnify the statistically small differences between good and bad teams in baseball. The fact remains that baseball is, based on the numbers, more competitive than any of the other major sports leagues.

Finally we come to the NHL. The NHL took over the Stanley Cup prior to the 1926-27 season. Since then, here are the numbers:

Despite the complaints about free agency, small market teams, rising payrolls and such the league has been more competitive over the last 20 years than it has at any time in its history, based on this criterion. Looking at the standard deviations supports this claim.

The National Hockey League has produced a wide variety of Stanley Cup finalists as well as a very even league, overall, since the strike-shortened 1994/1995 season.
Here's a comparison of the average of annual standard deviations of team records since the inception of the NFL salary cap:

Despite claims of restrictions on team spending, the two leagues with serious caps and revenue sharing show the least competitive balance league wide. The NFL number is skewed high due to the relatively small number of games they play (16 per season), while MLB is skewed low due to their very long season (162 games), but the NHL and NBA both play 82 games and the difference between them is marked. The two leagues without caps and significant revenue sharing seem to produce more balanced competition than the leagues with cost-of-labor restrictions over the last 8 years. In addition, we saw that the restrictions did little if anything to improve competition in the NFL and may have actually hurt the NBA's pursuit of parity.

Player Salaries

"Players don't deserve all the money they're getting, but the owners don't deserve it even more."
-Author and former Major League pitcher Jim Bouton

"And then we come into a year like this where we get five-year contracts - (at) $9 million (per year). There are ways to run a business and that's why there's going to be a lockout. Because we're not running the business well."
-Toronto Head Coach/GM Pat Quinn, July 2002 (who has a 6 year deal with center Mats Sundin for $8.75 mil per)

Proponents of restricting the price of labor in professional sports argue that salary caps, revenue sharing and other such methods will reduce player salaries. Capping the amount that teams may spend on payroll supposedly limits the ability of the owners to pay huge salaries to individual players and prevents salaries from spiraling out of control.

Taking a list of every player who started the season with a major league team, here are the numbers on player salaries (2001-02 for the NFL and NBA, 2002 for MLB and 2002-03 for the NHL). Figure 14 lists the average salary for each of the major leagues, the median salary (the one in the middle), and the standard deviation on salaries (how much variation is there).

Here is a listing of the top five individual salaries in each sport:

The NBA is far and away the best league for player salaries. The NFL has the lowest average rate of pay (and doesn't have guaranteed contracts either). It is interesting to note that the two leagues without salary caps are the two middle ones in terms of average salary, not the highest. Despite complaints about skyrocketing salaries in the NHL, the top moneymaker in hockey is Jaromir Jagr of the Washington Capitals. His salary of $11,483,333 per year would put him sixth in the NFL, sixteenth in MLB and eighteenth in the NBA. Here's the breakdown by league of how many players (number and as a percentage of the whole) make more than certain amounts:

Despite (or more accurately because of) being the smallest league in terms of players, the NBA has the same number of individuals making $10 million or more as the rest of the pro leagues combined. The leagues with larger team rosters have more players making under $1 million/year. Having a salary cap doesn't seem to affect how much money top end players make. Owners in all four leagues offer the top talent the most money to play for their team. To cut payroll they generally go after the mid-range players, not the stars. Even in the NFL with the hard cap most stars have their contracts re-negotiated to stall their impact on the cap, cuts tend to be less important, middle-pay range type players. The NHL's Washington Capitals are a good example of this. After signing Jaromir Jagr and Robert Lang to long term, big money deals over the last two years, they reduced payroll by trading or releasing the likes of Andrei Nikolishin, Sylvain Cote, Chris Simon and Joe Sacco. Serviceable players, yes, but older players all of whom were making more money than the Caps were happy with. They have been replaced mostly by minor-leaguers from within the Washington system and being paid close to the league minimum. The young players are competent but probably not as effective as the vets would have been, so far at least.

Just like in un-capped sports, salary caps don't strongly influence the escalation of player salaries. Players perceived as stars are being paid tremendous sums in all four sports (hockey least of all) while the salaries and job status of the low end and mid range players are adjusted to compensate.

The Proverbial Family of Four

"The average cost to attend a game: $303. And what do you get for these bucks? Four tickets, two small draft beers, four small soft drinks, four regular-sized hot dogs, parking, two game programs and two of the least-expensive adult-sized caps. Kind of takes away your appetite, doesn't it? That is not my idea of what constitutes an affordable afternoon. It's hard to enjoy yourself when you keep thinking about how many lunches you will have to skip just to get the budget back on track."
-Paul Attner, The Sporting News, "Attending an NFL game will smash the credit line", 11 September, 2001

Ticket prices have increased in all the major leagues, there is no question about that. The Team Marketing Report's "Fan Cost Index" is the popular way to judge how pricey sporting events are. The survey tracks the cost of attendance for a family of four. The FCI includes: two adult average price tickets; two child average price tickets; four small soft drinks; two small beers; four hot dogs; two programs; parking; and two adult-size caps. The FCIs for the 2002 or 2002/03 seasons are:

While the numbers used to come up with the FCI are certainly legitimate and the Index does give a good look at the increasing costs at the arena, it has flaws. First it assumes that the family will sit in "average price" tickets. Many teams offer discounted family packs that include most of the above items. The NHL's Capitals, for example, have the "WB50 Family Pack" that consists of 4 tickets, 4 hamburgers, 4 bags of chips, 4 medium sodas, and 4 Caps hats. It starts at $78 for the whole package but runs as high as $190 for premium side seats. The Fan Cost Index is similar to the Consumer Price Index: it measures a basket of goods, only a fraction of which are applicable to all consumers. As such it gives a good overall picture, but not a precise view of the actual state of affairs for individuals.

The argument that it costs much more today to attend a game then it did 5, 10 or 20 years ago is also open to debate. Allen Sanderson is a senior lecturer in economics at the University of Chicago and senior study director at the National Opinion Research Center (NORC). When not answering the sports-and-economics questions of reporters from the Chicago Tribune to the New York Times, he teaches a two-quarter introductory economics sequence in the College, and works on NORC survey research projects related to labor markets and higher education. Here is his response to the FCI and complaints about the cost of attending baseball games from his paper "Bottom-Line Drive" (published in 1994):

"It now costs over $100 for a family of four to attend a game. Hasn't baseball priced itself beyond the reach of the average fan?
Whether something is thought to be dear or cheap depends on its price increases over time, its price relative to goods or services we could substitute for it, and our ability to pay for it. On any of these criteria, tickets to baseball games are cheap and getting cheaper.
We have to remember that prices in the U.S. have been increasing, on average, for 40 years. Since 1950, the cost of living, as measured by the Consumer Price Index, has increased about fivefold. In inflation-adjusted dollars, the price of an average seat to a Major League game has actually decreased over this period; tickets were lower in real terms in 1993 than in 1950. Relative to other forms of entertainment -- professional basketball or football, amusement parks, concerts and plays, and even restaurant meals -- baseball tickets are less expensive and have risen more slowly in price. At the same time, we are more than twice as wealthy now as we were 40 years ago. Converting salaries into baseball-ticket equivalents, the average 1950 family income could have purchased about 1,200 tickets; today the corresponding figure would be 3,000.
The assertion that it costs a family of four $100 to attend a game is worse than comparing apples with oranges. That supposed "hit" includes four tickets, four hot dogs and soft drinks, a couple of beers, ice cream, parking, two souvenir caps, and a game program. With the exception of alcohol, fans are permitted to bring their own food into most ballparks (try that at a movie theater!). If you do eat at the game, you are simply substituting a meal there for one at home or at a fast-food restaurant, not buying a second dinner. And souvenir caps and pennants are not something we purchase at each and every game. In fact, no one is forcing fans to buy anything beyond the actual ticket."

Most sporting event attendees aren't families of four, either. Businesses buy tickets, groups buy tickets for outings, the most popular teams sell most of their seats to long-time season ticket holders. Another factor to consider is the concept of "good seats". The new arenas that have sprung up over the last 10 years or so generally don't have a bad seat in the house. Despite somewhat cramped quarters, they rarely have obstructed sightlines and even the top row of the upper deck in most NHL arenas is close enough to see the action. Older arenas have a higher number of less desirable seats, but the concept of the "bad seat" is not as prevalent as it was years ago. The average cost of a ticket tends to be weighted towards the upper end of the scale regardless due to the cost of club seats, premium rink-side tickets and luxury boxes. If I purchase a $150 ticket at center ice and four $25 dollar seats in the first row of the upper deck then the average ticket price for the five seats I bought is $50 thanks to the one pricey seat.

Ticket prices have increased, there is no doubt, and they continue to increase. Relative to incomes, relative to purchasing power and relative to other entertainment options, however, they have actually gotten cheaper. To quote Sanderson again, "..paraphrasing Twain, it's not so much what you don't know that gets you into trouble, it's what you think you know that turns out to be wrong".

Did I Do That? The Law of Unintended Consequences

"It is not from the benevolence of the butcher, or the baker, that we expect our dinner, but from regard to their own self interest."
-Adam Smith, "The Wealth of Nations"

The law of unintended consequences, often cited but rarely defined, is that actions of people, , organizations, and governments always have effects that are unanticipated or "unintended." Economists and other social scientists have heeded its power for centuries; for just as long, politicians and popular opinion have largely ignored it.
Any major change to the structure of a league will result in major consequences, intended and unintended. Take player movement as an example. Baseball and hockey feature trades by the dozen. The count down to the trade deadline in those two leagues is tremendously exciting as teams do their best to patch holes for the playoffs or to get rid of dead weight in exchange for players who can help next year. In football and basketball, though, the trade deadline is hardly noticed.

The NFL features virtually no trades in a given year. Teams generally swap draft picks when deals are made, rarely is a player actually sent from one team to another. In addition, the very early trade deadline means that few teams know what kind of players they actually need for the current season, or even if they are on track to make the playoffs at all. Given that most teams max out their cap space early, trading for an impact player is next to impossible if space under the cap must be cleared.

The NBA typically sees a large number of trades, but few meaningful ones. The majority take place during the off-season, especially on draft day, and are moves for draft picks. Because of the way the NBA cap is structured, trades must be balanced in terms of salary, a difficult medium to reach. The cap is set up to encourage players to remain with one team, making it even harder to deal established vets.