I posted my views on the desirability of drafting a college pitcher vs. drafting a high school pitcher about a week ago, but I was traveling quite a bit and didn’t have to time to follow/rebut the conversation.  I wanted to have another go at it, so I figured I needed to do some additional research in order to make it worth anyone’s time to read it.

First, I want to address a couple of points that were raised in response to the original post.  I thought I actually addressed one of them, but I guess not since it didn’t seem to get through.  Maybe I am the only one who thinks it makes any sense.  That point would be that only measuring performance for the years a player is under team control is an incomplete analysis.  Obviously, the first sixish years are ones every team can count on and should be looked at closely, but my point is that the only chance the Cardinals have to obtain a top of the rotation starter is to develop and retain one.  If you are one of the small market teams that is going to lose every single player the minute they are eligible for free agency then it makes sense to look at those years only. 

I hope the Cardinals are a team that can make a run at retaining their best players.  If someone knows for a fact that they aren’t please tell me now so I can stop yammering and quit being a Cardinal fan.  After all, the Cardinals have signed Adam Wainwright through his first two free agent years, at least if you count club options.  I think the ability to have a Wainwright-type pitcher signed for what should be significantly below market for two additional years shows that the Cardinals are a team that needs to look beyond the team control years in making value assessments.  In fact, it is probably their only hope to be successful in the long run.    

I think it is particularly important to emphasize that strategy as the Cardinals have utterly failed to sign a top-tier free agent starting pitcher that was available on the open market.  I can’t remember a top starting pitcher that has ever signed with the Cardinals unless they were already on the team.  Chris Carpenter was a dumpster dive, Adam Wainwright was obtained via trade, Daryl Kile was also a trade, Matt Morris was a draft pick.  However, each of these pitchers has signed contracts to remain with the Cardinals and bypassed potential free agency.

Secondly, someone raised the issue that looking at the statistical leaders in an attempt to find out where the elite pitchers come from is somehow a small sample size issue.  I may be obtuse, but I don’t see how looking at a three-year running total of all innings pitched by all major league pitchers is a small sample problem just because you are only interested in who the best ones are.  Especially if you have already analyzed fifteen years of data to attempt to identify where top of the rotation starters come from and have data that supports the same conclusion.  Anyway, I digress. 

I have reviewed fangraphs WAR data for all MLB pitchers since 2002 and isolated a running group of pitchers that were the top 35 in WAR for each three year interval – e.g. 2002-2004, 2003-2005, 2004-2006, etc through the 2008 season.  It made sense to me that sustained success would win out over any single outstanding season.  The results clearly indicate to me that high school pitchers have significantly more upside than college pitchers and should be considered at least as desirable when making a draft decision. 

For each of the running three year intervals there were about twice as many HS pitchers in the top 35 as college pitchers in spite of the fact that there are on average about 50% more college pitchers are drafted and signed in each draft, especially in the first five rounds.  Of the top 35 starters in these five intervals two-thirds of the pitchers were consistently players drafted and signed out of high school while only one third were drafted out of college.  The percentage of WAR achieved by HS players ranged from 62.1% in 2002-2004 to 70.4% in 2004-2006.  For the most recent period, 2006-2008, HS pitchers accounted for 238.6 WAR while the college guys only earned 125.8.

When the comparison is limited just to pitchers drafted in the first round the results are very interesting.  In the first interval examined (2002-2004) there were more than twice as many 1st round college pitchers in the top 35 as there were HS pitchers with the college guys earning 89.9 WAR to only 40.3 for the HS kids.  However, the trend has been in favor of the HS guys as both the number of 1st round pitchers and their respective WAR are have increased substantially.  By 2006-2008, the roles were reversed with 60% of the first round picks represented being from HS and the WAR advantage being 82.1 to 44.6.  For the entire period from 2002 to 2008 there were 22 1st round picks from college and 25 from HS. 

Maybe all I succeed in doing here is showing my lack of statistical sophistication and wasting everyone’s time, but I think these data show that any club that bypasses HS pitchers in the draft in favor of college pitchers is missing out on a superior talent pool.  This may not prove that HS pitchers are better, but I don’t see how anyone can view this information and still believe it is a better idea to draft college pitchers.   Unless, of course, you are Pittsburgh, Kansas City, or some other hopeless small market team who can’t/won’t sign your best players.  By the way, Chris Carpenter, Adam Wainwright, and Kyle Lohse were signed out of high school.