The recent February ennui thread featured a long discussion about the benefits and risks of various drafting strategies, with a few tangents into using minor-league performance to project major-league success. This wasn't the first thread we've had on this subject, and they all tend to get rather repetitive:
Person A: X is true.
Person B: There is no evidence for X. Y is true.
Person C: Y is obviously not true.
[I've partaken in these discussions many times myself, so cut me some slack on the parody.]
Here are some paraphrased assertions from the February ennui thread; we've all heard them stated and denied many times over:
- College players have less upside than high school players.
- College players are less risky than high school players.
- College players have a higher expected value than high school players.
- There exist identifiable groups of minor league players who can be expected to overperform or underperform statistical projections when advanced to a higher level.
- College players make it to the majors more quickly than high schoolers.
- Focusing on one specific group of players improves the quality of a team's drafts.
- Focusing on one specific group of players decreases the quality of a team's drafts.
- Tools matter.
- Tools don't matter.
Part of the problem with this sort of discussion is that, as far as I know, there is no organization that attempts to keep track of all studies on subjects such as these. I'm a medical student by day. If I want to find out whether a patient presenting to an emergency department with chest pain is more likely to be having a heart attack if he has a documented history of coronary artery disease, I can connect to Medline and search the entire literature; abstracts are freely available for almost all recently published articles. (In fact, a history of coronary artery disease has not been shown to significantly increase the probability that a patient is having a heart attack.) In baseball analysis, however, there's no well-organized, exhuastive repository of information to which one can turn for answers.
Ideally, I'd like to see someone create such a well-organized, exhuastive repository. However, I don't see that happening any time soon, so let's aim for a poorly organized and incomplete one. If you know of a good study on some aspect of drafting or prospecting, post a citation in this thread, together with a brief synopsis if you please. The study can be freely available on the Internet (e.g. at Primer), available on the Internet for a fee (e.g. at BA or BP), or published in a book (e.g. Bill James' Baseball Abstracts.)
Do high K/BB and K/IP ratios predict success for minor league pitchers? Do Baseball America's top 100 prospects outperform their PECOTA projections in aggregate? Is the speed of a pitcher's fastball a predictor of future success independent of his stats? Is his height a predictor of future success? Do high-school pitchers drafted in the first round have less success than their college peers? I really don't know the answers to any of these questions. I've heard it said that some of these questions have been conclusively answered, but I've never seen the studies with my own eyes, or if I have, I've forgotten where to find them. But perhaps you haven't.
One last request: let's try to restrict this thread to citations of studies and constructive criticisms of the studies cited. I love arguing about Brian Grant and Jamie Vermilyea as much as the next guy, but there's already an active thread for that discussion.
https://www.battersbox.ca/article.php?story=20040215060157999