Making Reasonable Comparisons

The Rookies of the Year will be announced about the time this post goes up, or shortly thereafter. I do not know who will win, but there are a number of good candidates. Among those candidates should be a pair of 21 year-old first basemen: Kansas City’s Eric Hosmer and Atlanta’s Freddie Freeman. Controversial defensive metrics aside, their 2011 hitting performances (114 wRC+ for Hosmer, 118 wRC+ for Freeman) have rightly earned them recognition. Apart the passing glory of the Rookie of the Year award, what is most exciting for fans of both players is how such performances at such a young age bode for their respective futures in the game. While those numbers are unspectacular for first basemen, to have done so at an age when most of their peers were in college or the minors is most promising. This leads to columns like this, which compares Hosmer’s 2011 performance with that of other 21-year olds who hit as well or better. It is an impressive list, of course, and the general point is sound. But by only comparing Hosmer with players who hit as well or better at 21 it also skews the perception of Hosmer’s season. It makes Hosmer’s 2011 seem closer to Frank Robinson‘s age 21 performance (139 wRC+ in 1957) while leaving out that it is much closer to Jim Fregosi’s 1963 (111 wRC+). That should not be taken to demean Hosmer or Freeman. However, looking at more reasonable comparisons can leave optimistic impressions without being unrealistic.

[Before I go any further, without taking anything away from Hosmer and Freeman, it is worth noting that Mike Stanton, who is the same age as they are, put up a 118 wRC+ in 396 2010 major-league plate appearances, while Hosmer and Freeman were both in the minors. This season, he put up a 138 wRC+. If you want to [loosely] compare someone to Frank Robinson and players of similar career stature, Stanton would be a better choice.]

It is probably obvious that although wRC+ is a great “all-in-one” offensive stat, when trying to make comparisons aimed at getting a sense of possible futures of young players, one should be more specific in terms of not just specific component stats (walk rate, strikeout rate, home run rate, etc.) but also position, handedness, body type, and other more specific traits. However, it would be unfair to criticize the performance-at-this-age comparisons on that basis — they are not trying to be that fine-tuned, but are simply making a general point.

As I said above, simply looking for comparisons to Hosmer and Freeman from players who his as well or better in their age-21 seasons is distorting for obvious reasons. Doing a search of age 21 offensive seasons since 1955 with 500 or more plate appearances, the best three I find according to wRC+ are Cesar Cedeno (168 in 1972), Albert Pujols (158 in 2001), and Ken Griffey, Jr. 149 in 1991). These are the sorts of names that get listed, which implies to some that Freeman and Hosmer are in line for that sort of career, which simply is pushing things too far. This is not to say one has to be that good to have a great career (check out Mike Schmidt’s rookie season, for example), simply that such comparisons are obviously stretched and set up unrealistic expectations.

One can still find optimistic comparisons by looking at players who had offensive performances that were closer to the level of Hosmer and Freeman (and, yes, there is selection bias at work in our favor, too). So I did the same basic query (age 21 seasons with 500 or more plate appearances since 1955) but set the minimum wRC+ at just 105. The query turned up 48 players who met the criteria. Who were the three worst?

Bob Bailey was number 46 in the rankings with a 109 wRC+ in 1964, his second full season in the league. Bailey played third and first for a number of teams, and while he was never a superstar, he had a long and productive career, including five seasons of three WAR or more, three seasons of four WAR or more, and a great 5.9 WAR season in 1973 with Montreal.

While Gregg Jefferies (number 47) ended up spending time at first, third, and left field, in 1989 he was a 21 year old second baseman with the Mets. He was not much with the glove (which is probably why he moved around so much), but he hit well, for a 106 wRC+. While his career was less notable than Bailey’s, he did have some good years in his twenties, and was an absolute monster in 1993, and made the All-Star team both in 1993 and 1994.

The “worst” player on our list only hit for a 105 wRC+ in his age-21 season. What ever happened to Roberto Clemente?

Clemente aside, readers are likely nonplussed by the career of Bailey and Jefferies. I am not asserting that they are good comparisons for Eric Hosmer or Freddie Freeman. They were not Hall of Famers or even superstars of their time in any lasting way. However, they were not bad players by a long shot — they both were good hitters for a decent numbers of years, and each had one or more very good seasons. Hosmer and Freeman each hit better than Bailey, Clemente, or Jefferies did at 21. That does not mean that Hosmer and Freeman will definitely be better than even Jefferies. However, it strikes me as much reasonable to set expectations for good young hitters such as Hosmer and Freeman”better than Jeffries” than to put them on a list with A-Rod and Orlando Cepeda.