This week's Miscellaneous Musings column got me to thinking about position scarcity. In a lot of ways, I analyze these things similarly to the way Chris does, but I have a few twists on the notion of position scarcity that I think are worth sharing.

The first thing I do every year, in every league I'm in, is figure out exactly how many players at each position will be on the fantasy league rosters in any league I'm in.

In a standard ten team NL-only roto league, for example, you'll see some real discrepancies in the way various positions are distributed. If you start three middle infielders (a second baseman, a shortstop, and a middle infielder), that means your league will start thirty middle infielders. The real national league teams will start a total of thirty-two, so there will be two major league starters who don't even warrant a starting fantasy job. This is why I got Dan Uggla off the waiver wire a few weeks into last season.

Outfielders, on the other hand, exceed the limit in my league. We each start five, so that means a total of fifty players start in our fantasy league. In the real National League, there are only forty-eight starting outfielders.

In most of the leagues I play, starting pitching is the position with the least scarcity from a pure numbers standpoint. One of my leagues, a ten team NL-only head-to-head points league, allows fantasy players to use up to seven starting pitchers. Even if every team took this option, that only adds up to seventy starting pitchers. In the era of five-man rotations, eighty pitchers are currently in starting rotations at any one time. In most of my leagues, inning limits hold you to about five starting pitchers, in which case nearly forty percent of the starting pitchers in the league won't be starting on fantasy teams. This can be true even in very deep leagues.

However, the vast dropoff in talent in starting pitching, along with the total uselessness of most fifth starters, makes analyzing the scarcity more difficult. What I do for each position (not just starting pitchers), is try to get a handle on what kind of performance can be expected out of the worst starter at each position. In other words, if I spend my money on everything else and get stuck with the last middle infielder, how will his performance compare to the rest of the middle infielders. This is more easily done in points leagues than roto leagues, obviously.

The other thing I do is determine both the average and the mean performance at a given position. What you often see in doing this analysis is that there isn't a whole lot of difference between the fifteenth middle infielder and the thirtieth. If that's the case, you know you want to use your money or draft picks elsewhere once the top middle infielders are gone.

Once I've figured out where the soft spots in the league may be structurally, I begin to try to rank the players more formally. At some positions, you are able to find very sizeable gaps between players, and you may find that it's not just the difference between the fifteenth and thirtieth ranked middle infielders isn't very large -- you may find that the cutoff for real differentiation is even higher up the rankings than that. In other words, if your eighth ranked middle infielder is still available, along with the twenty-first ranked outfielder, there's more to the analysis than just looking at who ranks more highly at the position. You need to look at how both players compare to what you may be able to get further down the road.

Next week, with the start of Spring Training, I'll start looking at specific players and positions for 2007.