I recently came across the basketball-analysis site hoopism.com  Even though I love the site, the article that prompted me to blather was something I think is if at least not just wrong, completely missing the point. He argues that the best players will migrate to the big cities for the endorsement income. Below is my response.
"Boston, L.A., New York, Miami will compete each year for the title while the Clippers and Golden State incubate talent.."
This sentence refutes itself. To clarify, insert "LA" in front of Clippers. Also, it seem the Nets play in a larger market than the Lakers, yet the Lakers have been slightly more successful for, oh, all time. And the Clippers are a.. Okay, just.. The Clippers. Competent management would seem to be a nontrivial factor.
The natural evolution of the situation as described would be for teams to continue gravitating to larger markets, e.g. Kings –> LA market, Fort Wayne Zollner Pistons –> Detroit. (For the Sacramento Lachrymentalists crying about the gosh-darn unfairness of it all, I’ll append the Rochester Royals –> Cincinnati, Cincinnati –> Kansas City, etc…) But it is this same LA destination market that is the home of the Incubators. (Michael Olowokandi wasn’t so bad — he was just being incubated.) So obviously something more is at work here. Other owners can see that the Clippers are proof that a team doesn’t have to be good to make money. The equilibrium should occur when the major metro areas have multiple teams and less dense areas share teams (The Rocky Mountain Sun-Nuggets, e.g.). True parity will come when every basketball player represents approximately the same number of citizens. If this seems unfair because of historical biases, one need only ask oneself if we were to start from scratch would it be fair to bless the average citizen of Memphis with SEVEN TIMES as many NBA franchises as the average citizen of Chicago?
There is another factor which subverts the standard large-market migration endorsement income theory. While Amare Stoudemire might sell more shoes from NYC, his primary source of income is now taxed not just by the Feds, but (for at least his home games) by the state of NY and NYC as well. If the marginal tax liability for the entire Knicks roster doesn’t exceed the marginal endorsement post-tax income, it's certainly on the same order. If there were a true salary cap in place, the large-market tax consequences might actually reverse the great migration. Perhaps a future NBA finals will see the Cheyenne Sharks rumbling with the Juneau Jets.
The dominant factor driving the large/small market dichotomy is the soft salary cap. Having a luxury tax in place of a true salary cap really will amplify the advantages of the large market teams. In fact, you couldn’t devise a better system for promoting the large markets. I don't think that this is even a happy accident. The league tilts the playing field while simultaneously chirping about egalite. At the level of the team, the marginal payroll from exceeding the fake salary cap will surely wipe out the ancillary income that is claimed to be driving the great migration. A large market owner with the latitude to spread around $20M would seem to be a lot more appealing to a star (or perhaps even more so for the ACGreen-type vital role-player) than spending his "nights and weekends" doing ads for Whataburger. There is a natural experiment that completely blows the large-market ancillary income myth out of the water. It's name is Lamar Odom. The reason Lamar Odom is a Laker is not because he can now get more endorsements playing in Los Angeles than back when he used to play in, um, Los Angeles. It’s because the soft salary clap allows the Lakers to pay an all-star to come off the bench.
 This is the coolest data representation thing I've seen in some time:
 My favorite is Rasheed Wallace