Strength of Schedule

Projecting pitchers is the goal for these MLP's, at least.  But, as anyone who has seen pitcher projections in the past would attest, projecting pitchers is maddeningly difficult.  There are two primary factors which contribute highly to making them imprecise.  First and foremost is injuries.  Unlike with hitters, almost any injury to any part of a pitcher's body can dramatically reduct his ability to get maximal velocity while controlling his stuff.  Those are not considered at all in MLP's, and if anyone plans to use them for reference, manual adjustments need to be made.  The other factor is more complicated to explain, and can be loosely described as the “pyramid” nature of minor league talent, with there being many more players qualified to play at each lower level than the level above it.

As with a “pyramid marketing scheme”, sports are cut-throat, with players doing whatever they can to get to the top level of the “pyramid”, and staying there.  Also, like a scheme, there is some fluidity to the people involved, with players on their way up, and players on their way out at any level at any given time.  With better data, pitchers would be further analyzed to see how they did against the various “levels” of hitters which they faced in a minor-league season.  This would separate out some of the pitchers who are able to get out “minor-league hitters” almost routinely, but don't have the pure “stuff” to get MLB-quality hitters out regularly.  Conversely, pitchers who are “their own worst enemy” (usually measured in a rudimentary manner by looking at walk totals) are “penalized” less for their wildness than this wildness would actually impact them in games.  This is based on the fact that one area where pitchers do show measurable growth over the years is in their control. 

All caveats aside, here's a list of the top pitchers with 10+ starts in the minors in 2007.   Minor-league relief pitching numbers have much less predictive value than SP numbers, so relief pitchers aren't included. 

Using aging curves, modified by a variety of factors, Major League Projections (MLPs) attempt to project a given set of data compiled in a given minor-league season out to the “prime years” of a player's major-league career.  To me, the easiest way of thinking about these target values is an average of the “middle three” seasons from ages 26-30, meaning to discard the high and low outliers from this grouping.  Various factors are taken into account, such as park factors, BABIP, and other stats which can influence how much growing a player has left before age starts him into a decline phase.  Obviously, MLPs are geared toward the mythical “neutral park”, since a lot of things can change between the time a guy plays in A ball and when he's in his prime in the majors.

And others – as with MLB players as well – have wild fluctuations, though with the adjustments for BABIP, some of the fluky batting-average variations are smoothed out.  Obviously, this system also assumes that players will follow a typical growth pattern, based on their previous stats, and not pull a Hanley Ramirez explosion nor a Sean Burroughs collapse. 

Note: SB and CS numbers are NOT projected, rather they are simply the minor-league totals.