« Happy New Year | Main | Weekend Photo »

January 05, 2011

What becomes of Third Basemen at Michael Young's Age?

As of today, Adrian Beltre has replaced Michael Young as Texas's third baseman. Young was a shortstop in March 2007 when he signed a five-year extension covering the 2009-2013 seasons. The winter before his extension kicked in, Texas shunted him to third in favor of 20-year-old, Double-A glove wizard Elvis Andrus. Now, two years, later, Young is a player without portfolio. Assuming he stays in Texas, he'll mostly DH and also substitute at multiple positions. While I doubt many people (including the front office) expected him to serve all five years of his extension at shortstop, it's remarkable that he'll serve none of them there, and only two of the five at a specific defensive position. Regardless of the wisdom of management kicking Young to the easy side of the defensive spectrum, it's a huge disappointment relative to contractual expectations.

To answer the title question, I first create a group of Young's peers with the following criteria:

  • Hitters with at least 750 total plate appearances during their Age 32-33 seasons (Young's most recent, and only seasons at 3B), and,
  • Played at least 50% of those games at third base.

88 players fit the bill, including Young himself and six other active players: Brandon Inge, Pedro Feliz, Chipper Jones, Melvin Mora, Alex Rodriguez, and Scott Rolen.

Rehearsals For Retirement

Of the 81 retired 3Bs, seven never took the field after their Age 33 season, including Travis Fryman and David Bell, the two worst 3Bs of the group by WAR (the Baseball Reference version, not Fangraphs). All of those seven were among the bottom half by WAR, and none was close to Young's combined 6.0 WAR for 2009-2010. Another ten departed after Age 34, most prominently Ron Santo and Joe Sewell.

The average exited MLB at Age 36, which in Young's case would be the same year his current contract expires. Put another, more depressing way: among this group of not-terribly-old third basemen whom I deliberately selected for stability in the field and at the plate, 59% didn't last to Age 37. Only nine (11% of the group) held a bat at Age 40.

Note that I've made no adjustments for playing time in the above discussion. Players with minimal service time during their final season (e.g., Pie Traynor's five games at Age 38) receive as much credit as those who die with their boots on (e.g., George Brett, 145 games at Age 40).

Offensive Stamina

Henceforth, I'm focusing on what Young's peers accomplished during Ages 34-36, coinciding with the three remaining years of Young's contract.

Of my original 88 third basemen, 83 completed their Age 32-33 seasons long enough ago to play through Age 36 (or retire). How many of them achieved 1,506 plate appearances, which equals the minimum need to qualify for the batting title (502) multiplied by three seasons? Only 19. Another 24 reached a total of 1,000 PAs. The other 40 fell short of 1,000. The majority of players fell to part-time or lesser status in the three years after turning 33.

Incidentally, Young is 1,152 hits shy of 3,000. If he wants to reach the summit before turning 40, he'll need 192 in each of his next six seasons. The only player in the group to equal or surpass 192 hits per season at Ages 34-36 was... Pete Rose. Even the younger version of Young stopped surpassing 192 hits after 2007. He'll need to play into his 40s to reach 3,000.

Defensive Stamina

Chucking the seven early retirees and the too-recent examples of Age 32-33 third baseman (Inge, A-Rod, Young himself) leaves 78 players. How many of them stayed put in their Age 34-36 seasons? More specifically, how many averaged at least 50% of their team's games during those three seasons (in total, not in each season)? 35 of 79, or about 44%.

Another 32 spent at least a plurality of their personal playing time at third, but fewer than 50% of his team's total games. So, the players in my data set were far more likely to maintain their current defensive position than their batting acumen or health.

The other 11 players (14%) would spend a plurality or majority of their Age 34-36 games at a different position. Young may end up in this least common group.

Baseball is a young man's game:


Young's combined 2009-2010 WAR of 6.0 ranks comfortably among the top half, if not the top third, of the 88 3B regulars during their Age 32-33 seasons. The group had an average two-year value of 4.5 WAR and a median of 4.2.

During the following three years, at Ages 34-36, this group averaged a total of 3.3 WAR and a median of just 2.4. Again, those are three-year totals. On a per-season basis, the average 3B fell from 2.3 to 1.1, and the median fell from 1.4 to 0.8. These figures do include the aforementioned seven who never played a day past Age 33. Obviously, Young's not finished, so excluding the youngest retirees presents a more accurate picture. Actually, it makes little difference, as the average three-year WAR jumps only by 0.3 to 3.6.

The line is the result of a linear regression, which confirms about what you'd expect from eyeballing the data points (a clear correlation between past and future performance [r^2 = .41, t-value = 7.2] but a high standard error [3.3 WAR] ). The regression also confirms that the 3Bs' per-season value drops by about one-half from Ages 32-33 to Ages 34-36. The regression guesses Young will be worth a total of 4.5 WAR during 2011-2013. That's over $10 million per WAR, more than double the going rate for free agents according to Fangraphs. (Baseball Reference and Fangraphs don't calculate WAR identically, so I'm loathe to make more than a general assertion in this regard. Nevertheless, it's painfully clear that Young will almost certainly be grossly overpaid during the next three seasons.)

Unlike my Cliff Lee research, I found no indication that the elite 3Bs aged more gracefully than average players. Everyone suffers. In any case, Young isn't in the elite group.

Defense and Value, Together

As you'd guess, the players who started at least 50% of their team's games during their Age 34-36 seasons were the most valuable. They provided a total WAR of 5.8 (average) or 5.1 (median) during the three years. On the whole, the other players were far less valuable. Those still playing mostly at 3B but not as frequently provided an average three-year WAR of 1.6. Those playing mostly at other positions had an average three-year WAR of 1.4 and a median of zero. The most notable outliers among ex-3Bs were George Brett (9.2 WAR), who shifted to 1B at Age 34, and Darrell Evans (9.3 WAR), who spent a very narrow majority of his time at first. The other nine ex-3Bs provided a combined three-year WAR of 6.4.

Is (Beltre + Young) > (Young + Guerrero) ?

Though Young is a poor defensive 3B, the switch to primary DH will still negatively impact his value. For example, pretending that Young spent all of 2010at DH decreases his Fangraphs WAR from 2.4 to 1.3. The difference between the defensive contribution of an average 3B and a DH is about two wins, and Young's defense isn't so terrible as to make the switch to DH a wash. (Not yet, anyway.) Regardless of the team-wide impact, Beltre's acquisition harms Young personally unless you believe Young's defensive ability is on the cusp of total collapse. What little chance he has of providing value commensurate to his contract has vanished. He doesn't have a $16 million bat.

Does Beltre make the team better? I think so, at least in the short run. The defensive difference between Young and Beltre is about two wins per season, and Beltre is a solid hitter overall, albeit prone to wild variance from year to year. Some back-of-the-envelope calculations:

Beltre: 4-5 WAR
Young: 1-2 WAR (spending 2/3 of time at DH and 1/3 at various positions)


Young: 2-3 WAR (as a 3B)
Guerrero: 1-2 WAR (as a DH)

The Beltre configuration adds about two wins in 2011 at a cost of, say, $8-10 million, assuming Guerrero would have signed for $6-8. That's close to the market average of dollars per win. As for 2012 and beyond, well...


Adrian Beltre is entering his Year 32 season. In two years, and with three years plus a vesting option remaining on his contract, we can apply this analysis to him.

My gut feel is that Beltre should give Texas three good years. The back end of the deal will be ugly. Not as ugly as Young's is shaping up, but ugly nonetheless.

Fairport Convention, "Who Knows Where The Time Goes?" from Unhalfbricking, 1969

Posted by Lucas at January 5, 2011 01:04 PM