Thursday, February 24, 2011

Taking a Look at Prospect Rankings

*Edit 3/1: Keith Law brought to my attention that Jenrry Mejia was excluded from his list because he is no longer a rookie and thus not eligible. This has been changed. Also, the Baseball Prospectus rankings have been included.

I have been trying to obtain information on MLB prospects since the beginning of this year (thanks, dynasty league fantasy baseball). Sure, I had heard of Bryce Harper and Stephen Strasburg, but it was literally a surprise to me when some new prospect would come along and win a starting job. People like Carlos Santana. Buster Posey. Even Jason Heyward. Pathetic, right? So now I know a tiny bit more. I know that Devin Mesoraco used to be highly thought of, had a few pretty poor years in the minors, was nearly written off, supposedly re-tooled his swing, had a good year last year and is now back on prospect lists. Not too in-depth, but it's a start.

Anyways, that's not the point. I, like many others, like lists. I also like making spreadsheets with all of the lists put together and comparing them. So that's what I did with prospect top 100 lists. I looked at AOL Fanhouse (AOL), Baseball America (BA), Baseball Prospectus (BP), Keith Law (Insider) (KL)MLBProspect Junkies (PJ) and Project Prospect (PP). I will also be looking at Fangraphs once they are available. Of course, the MLB rankings only go to 50, so they do not contain as much information. Here is the spreadsheet.

I took all of these rankings and averaged them to get an overall rank for each prospect. The averaged list starts out with Mike Trout with an average rank of 2, contains 156 players and ends with Jacob Petricka, James Jones, Joe Benson and Tyrell Jenkins. Each of these four players were only listed as the number 100 prospect and only on one list. Good start. Now comes the fun part. What can I do with all of this information? Well, the first thing is look at how each list likes or dislikes certain players. The average for a player may be 63 (Jaff Decker), but he may be ranked at 26 on one list (PP), 97th on another (BP) and not even show up on others (AOL, BA, MLB). As it turns out, Jaff has the greatest standard deviation among players ranked at least twice (isn't cherry picking fun?).

For each ranking, I looked at the players it had ranked higher or lower than at least 20 spots from the average of players that were included on at least 3 lists. Why 20? Because. I also listed the players who were surprisingly included or excluded. A surprising inclusion is defined as a player who was only on one or two lists and was ranked higher than 80. A surprising exclusion is defined as a player who was on at least 4 prospect lists, but did not make that specific list. Another note: by "likes" or "doesn't like" I simply mean ranks higher or lower than average. It is definitely not to be taken as "Keith Law doesn't like Chris Sale as a person." The wording is simply odd since "ranked highly" could mean "ranked 100" or "ranked 1." Just to be clear. First, a summary of each system:


So here we go.