The MLB amateur draft is one of the biggest crapshoots in professional sports, with only a small percentage of draftees ever seeing major league time, and even fewer having any type of sustained success. The book (and now film) Moneyball would have you believe that the Oakland Athletics figured out how to game the draft: by focusing on high-performing college players with good plate discipline like Jeremy Brown at the expense of toolsy high schoolers like BJ Upton. In retrospect, the famous Moneyball draft didn’t go as well as planned, and the more successful draft classes in recent years often included these two characteristics: possessing many early draft picks and paying large overslot bonuses to keep talented high schoolers from going to college. Wealthy teams such as the Yankees and Red Sox in particular have been able to take advantage of the “suggestive” nature of draft slotting to bring talented players into the system despite having lower 1st-round picks.
With most teams these days willing to exceed slot recommendations and increase their draft budgets, spending money in the draft is no longer as significant of an advantage for the Yankees and other big-market teams. So what is the next frontier in improving the value of your draft choices? Rany Jazayerli of Baseball Prospectus has an intriguing hypothesis, based in part on the incredible success of some recent high school draftees who were among the younger players in their draft class:
…given two players taken at the same slot in the draft, the younger player returned greater value. In other words, even accounting for the fact that teams took age into consideration—presumably, a player who is particularly young for his draft class might get picked earlier—I wondered if those players were still undervalued.
Jazayerli focused only on high school hitters given the unpredictability associated with pitchers, and the theory that age had less of an impact as players entered college. He compared the expected value of the 5 oldest and 5 youngest players in the top 100 picks each draft class from 1965-1996 (based on their draft slot) with the actual value generated by these players in their major league careers. His finding was pretty stunning:
A team that drafted one of the five youngest high school hitters selected among the top 100 picks could expect MORE THAN TWICE AS MUCH VALUE from him as a team that selected one of the five oldest high school hitters. And that’s not a small sample size fluke; that’s a result derived from 32 years of the draft, looking at 160 players from both camps.
This is a pretty incredible finding. One might expect younger players to have more room for improvement because there theoretically would be more time for physical maturation, and younger players would have slightly less experience. Given the additional projection present in younger players, we would expect teams to draft a younger player higher than a similarly talented player who is a year older, and indeed they often do. However, the results of the study suggest that teams are not accounting for age adequately in their evaluation of players.
This finding is very intriguing, but there are some important limitations. It only looks at high school hitters, which limits the application of these findings to only this cohort. Additionally, by looking at only the 5 oldest and 5 youngest players drafted in the top 100, the results may be a little overstated due to a focus on outliers. It is unclear how much of an age difference would be important in evaluating players, especially considering that kids develop differently. Finally, the study only looks at draftees up to 1996, so the findings wouldn’t account for teams changing their draft strategies in the last 15 years or players drafted in recent years (though Trout and Heyward, along with Bryce Harper, do provide compelling anecdotal evidence).
So what does this finding mean for the Yankees? Assuming the Yankees underestimate the importance of age as much as other teams, it would make sense to value the talents of younger draftees even more highly than they currently do, and discount the value of older draftees. These results make me curious about some of the Yankees’ younger draftees, to see whether they were possibly undervalued in the draft process, either by the Yankees or the traditional prospect-ranking authorities (BA, Keith Law, etc.).
The two that immediately jump to mind were both from the 2010 draft: 1st-rounder Cito Culver (birthdate: August 26, 1992 and 2nd-rounder Angelo Gumbs (birthdate: October 13, 1992). The Culver selection was universally criticized as a reach, and lot of people were confused by the Gumbs pick as well. Both were drafted at 17, and made their professional debuts in rookie ball at that age. They then played for Staten Island this season against many older players at the tender age of 18. It is possible that their young age led them to be underrated in the draft process, allowing the Yankees to get great value by getting them when they did (and explaining why they were not on the BA radar for where they were selected).
Compare this to fellow 2010 draftee Mason Williams, whose August 1991 birthdate made him about a year older than Culver, and 14 months older than Gumbs. Williams definitely out-performed Culver and Gumbs this year, but considering the 1-year age gap, the disparity was maybe not as wide as we are led to believe. What kind of performance would Culver and Gumbs put up in Staten Island at Williams’ age? We likely will never find out, because they will presumably move up to Charleston next year along with Williams.
Jazayerli will have a second article on draftee age tomorrow, and I will be interested to see if he ends up with similar findings.
If you enjoyed this article, please consider sharing it!
I have always been a fan of drafting the more toosly high school player over a polished college player for a team like the Yankees. We can absorb misses in the draft more so than another team, and I would rather have the talented high school pick growing up in a farm system facing other high talent high schoolers and college athletes. Again I havent done the research to see which type of player works out more often but in my opinion its a low risk high reward risk a team like the Yankees should take.
Most polished college players with high ceilings aren’t available when the Yankees typically pick in the 1st round, so I agree that often the high schooler may be a better choice.
My first impression is that this is a self-selecting group, in that scouts are out looking for skills, and the most talented players show those skills at a very early age. Therefore, the earlier you’re drafted, the more talented you probably were to begin with. Whether its a high school coach making a phone call to a scout he knows about a kid who’s a man among boys on his team, or a scout finding a kid on a sandlot, this information will get filtered up well before the draft occurs, and always has.
I’ve heard stories of DiMaggio, Mays and Mantle playing above their level in high school, being on the radar of scouts by the time they were teenagers. Ruth was in a reform school/orphanage when he was discovered. This just looks like more of the same to me. I don’t think age in and of itself is the predictor here (and therefore something to be exploited) but rather the innate talent level revealing itself at an early age.
Agreed. However, the study seems to indicate that even with this knowledge, teams and scouts are still underrating the younger players.
I think it has to do with being scared of the risk. I think it’s probably much easier to convince an owner/GM to draft a player as a scout if you have more information to go off of, and it’s certainly easier to sell a college kid who could be in the majors in 2-3 years over a high school kid who could take 3-5.
It’s worth the risk, but sometimes jobs are on the line now, and you have to sell something to someone to keep it.
True, I would imagine that scouts and GM’s are fairly risk-averse, which could prevent them from taking big risks with an early pick. However, if you miss out on a future superstar, you look pretty bad too.
I certainly agree. I would think though that many are looking to look good now and hope it turns out OK, rather than be fired now and have a steal turn up afterwards.
It’s not the best way for the fans, but I would imagine it’s the most used method.