I tend to think of parity in terms of the NFL. The definition I remember was that any team could beat any other team on a given Sunday. A more mathematical way of stating it would be all the teams have the same intrinsic winning percentage.

(Now, they don't have to be .500 clubs. The NL teams had a losing record to the AL in interleague play this year, indicating they all may be slightly less than .500 teams. But as long as they have about the same winning percentage, it works out the same. So for the purposes of this article, we'll assume parity exists when all teams have an intrinsic winning percentage of .500.)

What is an intrinsic winning percentage? It's the performance we'd expect from a team over a very large number of games. That is, if teams could play enough games to smooth out luck, the intrinsic winning percentage is the result we expect to see.

Now sometimes what looks like parity isn't. Here's a trivial example. Suppose all the teams in the NL played four games, and each team was 2-2. While that certainly looks like parity, it likely isn't. The binomial distribution tells us that after four games in a league in parity, we're most likely to get one team 4-0, four teams 3-1, six teams 2-2, four teams 1-3 and one team 0-4. Sixteen teams at 2-2 is a low probability event in a league in parity (.0004). So we can test for parity by looking at the distribution of wins in the league. So do we have a parity situation in the NL right now?

The average team in the NL has played about 132 games. If each team had the same intrinsic winning percentage, we'd expect there to be 13 teams with between 59 and 73 wins. In fact, there are 12 such teams in the NL. That's pretty close.

Another way of testing for parity is through simulations. There is such a simulator at Baseball Musings. Keep clicking enter, and you'll see very close wild card races, similar to what's going on now.

So we have two separate tests, each indicating that the National League is close to parity. Now, the Mets are really good and the Pirates and Cubs really bad, but otherwise the model holds. It's Pete Rozelle's dream.

Now, is this a good thing? On the pro side of the argument, lots of teams are in the race and that should keep more fans interested. People like to watch winners. If there's a high probability your team is going to lose on a regular basis, there's less of a chance of you going to the park (see Kansas City and Tampa Bay).

The downside is the quality of play, since there are no great teams. There's a certain artistry in watching a top flight club. I remember in the early 70's watching the Yankees play the Oakland A's. Oakland was a great team, winning three World Series in a row. The Yankees were getting better, but not the team they would become at the end of the decade. The A's came to town and dispatched New York easily. On defense, on offense, on the mound, the Athletics showed they were better. Teams could watch them play and learn how to go about playing baseball. They hit mistakes, they fielded cleanly, they made good pitches. They were winners and they knew it. Outside of the Mets this season, I don't think there's another National League team like that.

There's another kind of parity, one that we saw from the beginning of free agency to the end of the CBS TV contract and the strike. For lack of a better term, I'll call it revenue parity. The money from National TV was enough to even out the disparity in local revenue at that time. It was parity of opportunity vs. parity of outcome, if you will. All the teams had enough money to build a winner, and from 1978 to 1992 a different team won the World Series every year. The teams weren't evenly match, but the resources were.

This is the parity I prefer, where great teams are created, teams that others can strive to beat. Lousy teams come into being as well, teams from whose mistakes others can learn. It's tough to appreciate greatness without the corresponding failure. With revenue sharing and new National TV contracts, I hope we're getting there again.

David Pinto is the author of Baseball Musings. David worked for STATS, Inc. for eleven years, ten as the lead researcher for Baseball Tonight on ESPN. He's also hosted Baseball Tonight online at ESPN.com and is a former employee of Baseball Info Solutions.