June 1, 2007
The Once and Future Superstar Shortstops
Last weekend in Miami, the Mets swept the Marlins by scoring 20 runs to the Fish's eight in three games. The series wasn't really a divisional showdown-the Marlins were 6 ½ games behind the Mets when they opened play, and despite last year's storyline and the weight of expectations, they're no better than the fourth-best team in the NL East. However, it did feature a head-to-head battle of a different sort: Jose Reyes vs. Hanley Ramirez. The two shortstops, who as recently as 14 months ago were the subjects of skepticism about their abilities, have become the two best players at their position in the National League, each not only providing valuable performance, but improving many aspects of their games to become well-rounded stars who now look like they have the potential to be MVPs.
Of course, that's not good enough for us statheads. We want to know the ultimate answer: which one is better, or more accurately, which one will be better to have over their careers?
Let's start by pointing out one potential problem with what follows: the mechanisms we use to determine a player's baseball age are generally good ones, picking a cutoff of July 1 or August 1 and using the age on that date as the player's season age. However, for players whose birthdate falls close to the cutoff, that can be misleading.
When you look at the DT pages for Reyes and Ramirez, you'll find that Reyes is listed as a 24-year-old this year, while Ramirez is 23. Their birthdays are just six months apart, however: Reyes was born June 11, 1983, and Ramirez on December 23 that same year. (Pause for the writer to weep for his spent youth.) The actual difference between them is just six months, but because of the mechanisms we use, they look like they're a year apart on the pages. This makes a considerable difference when it comes to projecting players at these ages. The availability of an extra year of development, of growth, can have an exponential effect on a player's projected peak and projected career totals.
It's for this reason that I'm throwing out my original idea of comparing the two players year-by-year by their age. Doing so would unfairly penalize Reyes, who isn't really a year behind Ramirez when it comes to his baseball age and potential development. How can you compare the two? Let's go back to the waning days of the 20th century first. Both players were signed not long after their 16th birthdays, Reyes by the Mets in the summer of 1999, Ramirez by the Red Sox in the summer of 2000.
Reyes debuted in the States in 2000 at 17, drawing 20 walks in 159 plate appearances in the New York-Penn League. He would advance through the Mets' system at about a level a year, mostly getting noticed for his raw speed, but showing signs of power (42 extra-base hits at 18 in the Sally League, 57 a year later in the Florida State and Eastern league) and occasional plate discipline (30 walks in a half-season in the FSL). He was always among the youngest players in his league, and reached the majors on the day before his 20th birthday, instantly becoming the youngest position player in the NL in 2003. Reyes was the Mets' regular shortstop from his debut until he injured his ankle at the end of August. He hit .307 in 274 at-bats, adding 21 extra-base hits and 13/3 SB/CS. His walk rate and strikeout-to-walk ratio were poor (13 walks, 36 strikeouts), and caused concern among analysts. In retrospect, I don't think I gave Reyes enough credit for what he was-a productive regular shortstop at the age of 20-and focused too much on what he wasn't.
Reyes' career detoured in 2004. The Mets signed Kazuo Matsui from Japan, and moved their shortstop prospect to second base that spring. The move did not take; Reyes was not a good second baseman, while Matsui was a complete disaster, raising questions as to why the Mets had ever shifted their potential superstar. Reyes missed two months with a hamstring injury, and fell apart at the plate upon his return, with 27 strikeouts and four walks in 188 at-bats. He played through a painful leg injury for much of that time before finally going on the DL in August. Reyes' raw speed and power potential were still evident, but on the whole, 2004 was a lost year.
As Reyes was climbing through the Mets' system, Ramirez was making his way through that of the Red Sox. He was never quite as young for his leagues as Reyes was, but in 2002 and 2003 he was showing the same blend of speed and preternatural power as the Met was. What was different was that Ramirez picked up some labels along the way, clashing with his coaches and managers, and gaining a reputation as a difficult player. Cause and effect are always unclear in cases like this, but what we do know is that Ramirez's performance stagnated a bit in 2004 and 2005. He hit for average, but his power was down and his speed wasn't translating as well on the basepaths. He lost ground among prospect mavens, not least here at BP, where he bounced around from 33rd on our list in 2003 to unranked in 2004, back to #24 in 2005, and rating only an honorable mention in 2006.
Traded from the Red Sox to the Marlins in November of 2005, Ramirez joined an organization with no established shortstop and no reason not to let the talented prospect play. Once again, I got caught up in what he wasn't:
[T]he centerpiece of the return package, Hanley Ramirez, is an overrated prospect, a tools guy who has yet to convince me that he's going to become a baseball player. Ramirez' speed hasn't translated well on the field so far--he's 112/45 on the bases in five seasons, he doesn't get a ton of speed hits or triples, and he didn't adapt well when the Sox tried him in center field. His bat stagnated at Double-A this year, and while you can point to his age (just 21) as a factor, .271/.335/.385 with 26/13 on the bases doesn't leave much to be excited about.
I'm not as upset about this evaluation as I am about what I said about Reyes, who was a regular in the majors, and not a bad one, at 20. Ramirez had more negative markers, fewer positive ones, and through his age-21 season had yet to prove he could handle Double-A. Ramirez, of course, would go on to win the NL Rookie of the Year award last season, hitting .292/.353/.480 with 51 steals and good defense at shortstop. He seemed to change his approach at the plate, trading some contact for power, and it worked.
Reyes also improved considerably in 2006 over 2005. After drawing 27 walks and posting a .300 OBP while playing in all but one game in '05, there were questions as to whether Reyes would ever be more than a 21st-century Shawon Dunston. Dunston was a productive player for a decade, but his inability to take pitches limited his value as a hitter. Reyes had power (48 XBH) and speed (60 steals at an 80% success rate), but the lack of OBP, especially for a nominal leadoff guy, was crippling. He would have to improve in that area or his upside would be severely constrained.
Reyes improved across the board last year, but most impressively in getting on base. He nearly doubled his unintentional walk rate while holding his strikeout rate steady, hitting for average and showing improved power. The 54-point jump in his OBP was as big a reason as any for the Mets' division title last season.
So if we allow that Reyes was well ahead of Ramirez through 2005-he'd played better and advanced further at younger ages, and had established himself as a regular in the major leagues-where were they after 2006?
AVG OBP SLG SB CS BB SO EqA WARP FRAR Reyes .300 .354 .487 64 17 53 81 .288 5.6 5 Ramirez .292 .353 .480 51 15 56 128 .288 6.9 17
The two players were essentially the same at the plate, with Ramirez's apparent edge in the field the difference in his favor. There's enough wiggle room in one-year measurements of defense to question whether that FRAR number should be enough to make his case. If you look at other methods, you find that the difference between them isn't as large; David Pinto's Probablistic Model of Range has them both right around average, with a slight edge to Ramirez. Baseball Info Solutions gives a small edge to Reyes, again with both around average. (Thanks to Marc Normandin for that information.)
Both players have improved in 2007. In fact, they've been two of the top ten players in baseball:
AVG OBP SLG SB CS BB SO EqA WARP FRAR Reyes .305 .388 .465 28 5 29 29 .310 3.8 13 Ramirez .314 .388 .502 15 4 21 39 .316 2.7 2
Again, the two players are comparable hitters, with the edge in WARP going to the player who's played better in the field according to Clay's system. Reyes' improvement in his plate discipline indicators is truly amazing. Four seasons ago, he walked once every 46 times he stepped to the plate; he now walks about once every ten times (intentionals aside). He's bumped his OBP 100 points in that timeframe, and he hasn't done much damage to his contact rate in the process. (You would expect a player walking more to strike out a bit more as well, as he puts himself in more two-strike situations.)
If you run these guys through PECOTA, you'll find that they're projected almost evenly for the 2007-2011 period, with Ramirez holding a slight edge, but well within the margin of error for this kind of exercise. Ramirez has shown more power than Reyes has, and has had somewhat better walk rates on the whole. On the other hand, PECOTA is seeing a one-year age difference where in fact we only have a six-month one, so it may be overstating Ramirez's potential for growth relative to Reyes. (This shows up a bit more in PECOTA's Upside projections.)
The facts are that Ramirez is six months younger than Reyes and he has been a slightly better player over the past 210 games. While he has less speed than Reyes-everyone does-his overall physical tools are superior. Whereas Reyes is going to hit for some power, Ramirez is going to hit for a lot of power, having 30-homer seasons and slugging .530 or better throughout his peak. Neither player has a significant edge in plate discipline right now, and Ramirez's power is likely to gain him some walks along the way.
When I was noodling about doing this article, I thought it would be clear that Reyes was the better player due to his improved walk rate-which has enormous value given his ability to steal bases and run the bases-and developing power. When you compare the two, however, you find that while Reyes was ahead of Ramirez for most of their careers, the latter player not only closed the gap in 2006, he started 2007 in a way that shows last year wasn't a fluke. There's no degradation here, no slip in underlying indicators that would herald a decline. The biggest difference between the two, power and projected power, is an edge in Ramirez's favor.
This is a very tough call, and one that I'm tempted to pass on making. I think Reyes is the safer pick, the player who will have less variance from year to year and less chance of collapsing. However, Ramirez's power potential, coupled with the rest of his game, gives him enough of an edge that, were I to choose, I would choose him from this point forward.