This is almost certainly going to be a disjointed and somewhat rambling response to zknower's excellent post Deep Pockets Required for Success in MLB. This is an indulgence on my part; I have no fundamental right to respond to his research, and it does not follow that because one person makes a good argument, someone else gets to make a counterargument. That said, his article brought up many reactions in me that I felt compelled to put into words, so if you feel like reading a lot of them, please follow me below the fold.
Of course, this isn't how it works. For a wide variety of reasons, market forces, and market failures, baseball players are not paid primarily on their expected or actual production. This is, as I mentioned before, an injustice. It isn't an injustice that I expect many people to feel hugely sympathetic over; these are all highly paid professional baseball players after all. Pujols and Lincecum got paid eight figure salaries last year, even if neither were among the top 25 in salary, and both have the promise of eclipsing the names currently ahead of them on this list in the near term. But it is important to recognize that the platonic ideal isn't a collectivized baseball, where all players are paid equally, regardless of performance. The best players deserve to be paid the most money.
While the best players aren't always or even usually the ones who are paid the most money, good players are paid more, generally, than mediocre and bad players. Just like there is an obvious correlation between how much a team spends on payroll and their success, there is a correlation between how good a player is and how much they are paid.
There are a lot of interesting things about which teams in Major League Baseball are big spenders, and which ones aren't. Obviously, the number one team in player payroll is the New York Yankees, the flagship franchise of the most populous city in America. The Yankees make the most money because of this in a variety of ways: there are more people demanding Yankees' tickets, so they can charge much more than other teams; the commercials during games are much more lucrative because they reach so many more eyeballs; bigger companies pay much bigger prices for luxury suites. New York City is so far and away the largest city in the United States that its population of 8,391,881 is more than twice that of the next largest city, Los Angeles, with a population of 3,813,868. But the team with the next highest payroll over the period measured in zknower's article, the Boston Red Sox, don't hail from Los Angeles. Boston isn't third in population either; in fact, Boston is the 20th most heavily populated city in America, with 645,169 residents in 2009. Boston has a smaller population than seven cities who don't even have a major league baseball team (San Antonio, San Jose, Jacksonville, Indianapolis, Austin, Charlotte, and Memphis). There are more than ten New Yorkers for every Bostonian.
The city with the next highest population after Boston, 21st in the country, is Baltimore, who happens to have the third worst winning percentage in baseball over the period measured.
Of course, that isn't the only interesting thing that you can find by looking at city population and the dataset that zknower presented. For example, after the Yankees and Red Sox, the three teams with the best winning percentages over the period measured are, in order, Atlanta, St. Louis and Oakland. Oakland won with one of the lowest payrolls in baseball, while both Atlanta and St. Louis were among the top ten in Major League Baseball in payroll over that period. Yet Oakland has a significantly higher population (409,189) than St. Louis (356,587) and Atlanta is only the 33rd largest city by population in the US. In fact, St. Louis, one of baseball's bigger spenders, has over 100,000 fewer residents than Kansas City, the team with the lowest average payroll over the measured period.
I could go on and on here, with the findings being of diminishing interest as I do so. But what the data shows is that outside of the very largest markets of New York and Los Angeles, and the very smallest (Pittsburgh, the smallest city with a major league team, is second to last in payroll) market size is not a determining factor in team payroll. But if payroll isn't determined by market size, what is it determined by?
There are a lot of reasons why certain teams might spend more money than other teams on player payroll, many of which are difficult to quantify. The Twins, for example, under the ownership of the late Carl Pohlad, were among the most tight-fisted in baseball, despite years of success on the field and a respectable population shared between Minneapolis and St. Paul. Since Pohlad's death, the Twins have sent their payroll rocketing over the $100 million mark. So ownership's willingness to spend can be one factor. Another factor is relative affluence; Boston may be barely bigger than Baltimore, but Boston is a much wealthier city. Bostonians can afford to pay more for tickets to games than can folks from Baltimore, can spend more on jerseys and other merchandise, and can be sold more expensive products during games by advertisers. Fan enthusiasm is another major factor; the Cubs had a worse record over the measured period than the White Sox, while sharing the third largest market size in the US. But the Cubs spent considerably more on payroll than the White Sox, who spent merely the league average on player salaries. This is certainly in some part because despite their century of mostly terrible performance, the Cubs are very popular, while the White Sox aren't. Ballparks are also a factor; regardless of their effect on attendance, newer, generally taxpayer-subsidized stadiums correlate with higher payroll expenditures (witness the rise in payroll of the Twins), while teams with antiquated and fan-unfriendly parks like the Rays, Marlins and A's continually lag behind in payroll.
The question of the income of residents of a metropolitan area is one that deserves more explanation. While Boston is not a large city by population, the Boston metropolitan area is the fifth wealthiest by median household income in the United States, according to the Census Bureau, with a median household income of over $52,000 a year. Atlanta is only a little bit behind, with a median household income of $51,948. But while this is a compelling explanation, it does not fully satisfy. St. Louis is not only smaller by population than Kansas City, but is also poorer; St. Louis has a median household income of more than $2,000 less than Kansas City. Meanwhile, the generally tight-fisted Twins reside in the metropolitan area with the fourth highest per capita median income in the United States, with the average resident of the Twin Cities earning more than the median of Boston, New York, and Chicago.
Perhaps the most compelling explanation to me, however, is the converse of zknower's argument: teams are generally able and willing to spend more money when they are winning over an extended period of time.
The Braves and Cardinals are the two teams with the smallest city populations and the highest payrolls over the measured period. They represent metropolitan areas on opposite ends of the median income spectrum as well, with Atlanta being rich and St. Louis being poor. They outspend even teams with winning records from much bigger cities, such as the Diamondbacks and Astros. They outspend teams from wealthier areas like the Giants and Rockies. The Braves and Cardinals also have the third and fourth best winning percentages over the measured period in all of baseball. Over that same period, both teams have fielded a number of players with strong Hall of Fame cases: Chipper Jones, Andruw Jones, Albert Pujols, Mark McGwire, Larry Walker, Scott Rolen, Greg Maddux, Tom Glavine and John Smoltz among them. Both teams have also featured Hall of Fame managers in Tony La Russa and Bobby Cox. Over the measured period, the Braves won their division in over half the seasons and the Cardinals won one World Series and lost another.
This prolonged success not only increased the revenue of the Braves and Cardinals above what their market size would suggest, but also gave these teams incentive to spend more. Year in and year out, the marginal value of a win for these two franchises is very high. Both teams also have high downside risk in not continuing to field a contender; both clubs have significant percentages of their revenues coming from playoff appearances and ticket and ad sales late in the regular season. Matt Holliday's contract is much more affordable when subsidized by playoff shares. The financial rewards of winning for franchises are large, although we can observe that they are not large enough to supplant entirely other factors limiting team spending, such as in Tampa or Oakland.
Baseball is a complicated economic system, where a multitude of factors influence what teams do and how they spend their money. Pat answers cannot satisfy us. And only when we recognize how elusive actual truths are will we properly value the few truths we can legitimately claim to know.