Valuing the draft (Part 2)


In my last article, I took a look at the draft pick value a type-A free agent gives. In this article, I’m going to follow a similar framework and take a look at the college versus high school split in the draft. I did not look at any junior college players due to the lack of a good sample size. In Rany Jazayerli’s previous draft study, he found that from 1992 to 1999, college players hold a slight edge over high school players. However, the advantage saw a big drop from 1984 to 1991.

To test the college vs. high school split, I broke drafted players into three tiers: first-round picks (1-30), supplemental/second-round picks (31-70), and third-round picks (71-100). I then separated players based on school level and position players and pitchers. From there, I found the average production of each group in a player’s cost-controlled years (his first six major league seasons).


First-Round Picks:
College Hitters: .76 WAB/year
High School Hitters: .75 WAB/year
College Pitchers: .49 WAB/year
High School Pitchers: .35 WAB/year

For first-round picks, hitters have a significant edge over pitchers. There doesn’t seem to be much of a difference between college and high school hitters. However, high school hitters do take longer to develop than college hitters. When we take a look at drafted pitchers, we see that college pitchers drafted seem to hold a decent edge over high school pitchers. So for first-round draft picks, college players have a slight edge with almost of that edge generated from pitchers.

Supplemental/Second-Round Picks:
College Hitters: .20 WAB/year
High School Hitters: .14 WAB/year
College Pitchers: .18 WAB/year
High School Pitchers: .16 WAB/year

For the supplement/second round, college and high school players are pretty bunched up. College hitters hold a small edge over high school hitters while college pitchers also hold a small edge over high school pitchers. The edge hitters had over pitchers evaporated in this round. Overall, college players were able to maintain an edge over high school players.

Third-Round Picks:
College Hitters- .04 WAB/year
High School Hitters- .10 WAB/year
College Pitchers- .11 WAB/year
High School Pitchers- .08 WAB/year

In the third round, college hitters take a hit, producing just slightly above replacement. High school hitters now have a slight edge over college hitters while college pitchers have an even smaller edge over high school pitchers. High school players have a slight overall advantage over college players in this round, though that advantage would probably disappear when you consider development time.

There are a few interesting things to point out from this data. In my article in 2009 THT Annual, I wrote that a pretty good trading strategy would be to go for elite hitting prospects and stock up on lower tier pitching prospects. The draft data show a similar result. Hitters have a large advantage over pitchers in the first round. That large edge disappears in the second and third rounds.

When you consider that pitchers have a higher $/WAB than hitters, you can see that a good drafting strategy would be to go for hitters in the first round and then bulk up on pitchers. Note that this is identical to the strategy I proposed in my Annual article. So overall, I feel pretty confident in saying that a general talent acquisition strategy would be to go for hitters as the top player in a trade or in a draft and then bulk up on pitchers. It’s also interesting to note that Paul DePodesta made a similar assertion in Moneyball.

Something else that is interesting to note is that college hitters lose much of their advantage in the second and third rounds. Contrast that with Rany Jazayerli’s draft rule #7:

“College hitters enjoy a sizable advantage over every other class of draft pick, in both eras, and in every round. “

My research contradicts this statement. In fact, my research shows that college hitters are the worst draft pick in the third round. I can’t really explain why our results differed so much. My 1990-1997 data set should have almost the exact same players as his 1992-1999 data set. It could have something to do with using WSAB instead of WARP. It could also have something to do with Jazayerli’s projections for players from more recent years. I chose not to use 1998 and 1999 in my sample because I did not have enough players from those years who had finished their first six major league years.


I think the most important information one can get from this article is that the “hitters first, pitcher next” player acquisition strategy holds in the draft. This information is particularly useful for teams looking to rebuild as it goes against the conventional wisdom of drafting and acquiring young pitchers early. Also, note that my research shows that college hitters do not maintain an advantage over other players in each round, contrary to past research. I’ll take a look at another aspect of the draft in another article, but I’m not sure what yet exactly. I’m open to suggestions.

Print Friendly
 Share on Facebook0Tweet about this on Twitter0Share on Google+0Share on Reddit0Email this to someone
« Previous: Lost in Transactions 1/26-2/2/09: Young pitchers sign long-term deals
Next: Draft strategy: Injury risks can lead to big rewards (Part 1) »

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>